Provide and easier way to generate magics and pre-post hooks

This introduce a code base way of modifying the ast.
This allow to use Template strings with the two special variable names
names `__code__` and `__ret__` allowing to quickly write a magic, or
hooks that modify the ast.

This also introduce a `%code_wrap` cell magic to on the fly wrap code.
It is this easy to for example modify IPython to say time each block of
code, or retry them, or wrap them in try/except and analyse the error
message, profile...

Note that this is not new, but simply convenience function and
utilities, especially around hygiene.

IPython/core/magics/ast_mod.py

@@ -0,0 +1,320 b''
+"""
+This module contains utility function and classes to inject simple ast
+transformations based on code strings into IPython. While it is already possible
+with ast-transformers it is not easy to directly manipulate ast.
+
+
+IPython has pre-code and post-code hooks, but are ran from within the IPython
+machinery so may be inappropriate, for example for performance mesurement.
+
+This module give you tools to simplify this, and expose 2 classes:
+
+- `ReplaceCodeTransformer` which is a simple ast transformer based on code
+  template,
+
+and for advance case:
+
+- `Mangler` which is a simple ast transformer that mangle names in the ast.
+
+
+Example, let's try to make a simple version of the ``timeit`` magic, that run a
+code snippet 10 times and print the average time taken.
+
+Basically we want to run :
+
+.. code-block:: python
+
+    from time import perf_counter
+    now = perf_counter()
+    for i in range(10):
+        __code__ # our code
+    print(f"Time taken: {(perf_counter() - now)/10}")
+    __ret__ # the result of the last statement
+
+Where ``__code__`` is the code snippet we want to run, and ``__ret__`` is the
+result, so that if we for example run `dataframe.head()` IPython still display
+the head of dataframe instead of nothing.
+
+Here is a complete example of a file `timit2.py` that define such a magic:
+
+.. code-block:: python
+
+    from IPython.core.magic import (
+        Magics,
+        magics_class,
+        line_cell_magic,
+    )
+    from IPython.core.magics.ast_mod import ReplaceCodeTransformer
+    from textwrap import dedent
+    import ast
+
+    template = template = dedent('''
+        from time import perf_counter
+        now = perf_counter()
+        for i in range(10):
+            __code__
+        print(f"Time taken: {(perf_counter() - now)/10}")
+        __ret__
+    '''
+    )
+
+
+    @magics_class
+    class AstM(Magics):
+        @line_cell_magic
+        def t2(self, line, cell):
+            transformer = ReplaceCodeTransformer.from_string(template)
+            transformer.debug = True
+            transformer.mangler.debug = True
+            new_code = transformer.visit(ast.parse(cell))
+            return exec(compile(new_code, "<ast>", "exec"))
+
+
+    def load_ipython_extension(ip):
+        ip.register_magics(AstM)
+
+
+
+.. code-block:: python
+
+    In [1]: %load_ext timit2
+
+    In [2]: %%t2
+       ...: import time
+       ...: time.sleep(0.05)
+       ...:
+       ...:
+    Time taken: 0.05435649999999441
+
+
+If you wish to ran all the code enter in IPython in an ast transformer, you can
+do so as well:
+
+.. code-block:: python
+
+    In [1]: from IPython.core.magics.ast_mod import ReplaceCodeTransformer
+       ...:
+       ...: template = '''
+       ...: from time import perf_counter
+       ...: now = perf_counter()
+       ...: __code__
+       ...: print(f"Code ran in {perf_counter()-now}")
+       ...: __ret__'''
+       ...:
+       ...: get_ipython().ast_transformers.append(ReplaceCodeTransformer.from_string(template))
+
+    In [2]: 1+1
+    Code ran in 3.40410006174352e-05
+    Out[2]: 2
+
+
+
+Hygiene and Mangling
+--------------------
+
+The ast transformer above is not hygienic, it may not work if the user code use
+the same variable names as the ones used in the template. For example.
+
+To help with this by default the `ReplaceCodeTransformer` will mangle all names
+staring with 3 underscores. This is a simple heuristic that should work in most
+case, but can be cumbersome in some case. We provide a `Mangler` class that can
+be overridden to change the mangling heuristic, or simply use the `mangle_all`
+utility function. It will _try_ to mangle all names (except `__ret__` and
+`__code__`), but this include builtins (``print``, ``range``, ``type``) and
+replace those by invalid identifiers py prepending ``mangle-``:
+``mangle-print``, ``mangle-range``, ``mangle-type`` etc. This is not a problem
+as currently Python AST support invalid identifiers, but it may not be the case
+in the future.
+
+You can set `ReplaceCodeTransformer.debug=True` and
+`ReplaceCodeTransformer.mangler.debug=True` to see the code after mangling and
+transforming:
+
+.. code-block:: python
+
+
+    In [1]: from IPython.core.magics.ast_mod import ReplaceCodeTransformer, mangle_all
+       ...:
+       ...: template = '''
+       ...: from builtins import type, print
+       ...: from time import perf_counter
+       ...: now = perf_counter()
+       ...: __code__
+       ...: print(f"Code ran in {perf_counter()-now}")
+       ...: __ret__'''
+       ...:
+       ...: transformer = ReplaceCodeTransformer.from_string(template, mangling_predicate=mangle_all)
+
+
+    In [2]: transformer.debug = True
+       ...: transformer.mangler.debug = True
+       ...: get_ipython().ast_transformers.append(transformer)
+
+    In [3]: 1+1
+    Mangling Alias mangle-type
+    Mangling Alias mangle-print
+    Mangling Alias mangle-perf_counter
+    Mangling now
+    Mangling perf_counter
+    Not mangling __code__
+    Mangling print
+    Mangling perf_counter
+    Mangling now
+    Not mangling __ret__
+    ---- Transformed code ----
+    from builtins import type as mangle-type, print as mangle-print
+    from time import perf_counter as mangle-perf_counter
+    mangle-now = mangle-perf_counter()
+    ret-tmp = 1 + 1
+    mangle-print(f'Code ran in {mangle-perf_counter() - mangle-now}')
+    ret-tmp
+    ---- ---------------- ----
+    Code ran in 0.00013654199938173406
+    Out[3]: 2
+
+
+"""
+
+__skip_doctest__ = True
+
+
+from ast import NodeTransformer, Store, Load, Name, Expr, Assign, Module
+import ast
+import copy
+
+from typing import Dict, Optional
+
+
+mangle_all = lambda name: False if name in ("__ret__", "__code__") else True
+
+
+class Mangler(NodeTransformer):
+    """
+    Mangle given names in and ast tree to make sure they do not conflict with
+    user code.
+    """
+
+    enabled: bool = True
+    debug: bool = False
+
+    def log(self, *args, **kwargs):
+        if self.debug:
+            print(*args, **kwargs)
+
+    def __init__(self, predicate=None):
+        if predicate is None:
+            predicate = lambda name: name.startswith("___")
+        self.predicate = predicate
+
+    def visit_Name(self, node):
+        if self.predicate(node.id):
+            self.log("Mangling", node.id)
+            # Once in the ast we do not need
+            # names to be valid identifiers.
+            node.id = "mangle-" + node.id
+        else:
+            self.log("Not mangling", node.id)
+        return node
+
+    def visit_FunctionDef(self, node):
+        if self.predicate(node.name):
+            self.log("Mangling", node.name)
+            node.name = "mangle-" + node.name
+        else:
+            self.log("Not mangling", node.name)
+
+        for arg in node.args.args:
+            if self.predicate(arg.arg):
+                self.log("Mangling function arg", arg.arg)
+                arg.arg = "mangle-" + arg.arg
+            else:
+                self.log("Not mangling function arg", arg.arg)
+        return self.generic_visit(node)
+
+    def visit_ImportFrom(self, node):
+        return self._visit_Import_and_ImportFrom(node)
+
+    def visit_Import(self, node):
+        return self._visit_Import_and_ImportFrom(node)
+
+    def _visit_Import_and_ImportFrom(self, node):
+        for alias in node.names:
+            asname = alias.name if alias.asname is None else alias.asname
+            if self.predicate(asname):
+                new_name: str = "mangle-" + asname
+                self.log("Mangling Alias", new_name)
+                alias.asname = new_name
+            else:
+                self.log("Not mangling Alias", alias.asname)
+        return node
+
+
+class ReplaceCodeTransformer(NodeTransformer):
+    enabled: bool = True
+    debug: bool = False
+    mangler: Mangler
+
+    def __init__(
+        self, template: Module, mapping: Optional[Dict] = None, mangling_predicate=None
+    ):
+        assert isinstance(mapping, (dict, type(None)))
+        assert isinstance(mangling_predicate, (type(None), type(lambda: None)))
+        assert isinstance(template, ast.Module)
+        self.template = template
+        self.mangler = Mangler(predicate=mangling_predicate)
+        if mapping is None:
+            mapping = {}
+        self.mapping = mapping
+
+    @classmethod
+    def from_string(
+        cls, template: str, mapping: Optional[Dict] = None, mangling_predicate=None
+    ):
+        return cls(
+            ast.parse(template), mapping=mapping, mangling_predicate=mangling_predicate
+        )
+
+    def visit_Module(self, code):
+        if not self.enabled:
+            return code
+        # if not isinstance(code, ast.Module):
+        # recursively called...
+        #    return generic_visit(self, code)
+        last = code.body[-1]
+        if isinstance(last, Expr):
+            code.body.pop()
+            code.body.append(Assign([Name("ret-tmp", ctx=Store())], value=last.value))
+            ast.fix_missing_locations(code)
+            ret = Expr(value=Name("ret-tmp", ctx=Load()))
+            ret = ast.fix_missing_locations(ret)
+            self.mapping["__ret__"] = ret
+        else:
+            self.mapping["__ret__"] = ast.parse("None").body[0]
+        self.mapping["__code__"] = code.body
+        tpl = ast.fix_missing_locations(self.template)
+
+        tx = copy.deepcopy(tpl)
+        tx = self.mangler.visit(tx)
+        node = self.generic_visit(tx)
+        node_2 = ast.fix_missing_locations(node)
+        if self.debug:
+            print("---- Transformed code ----")
+            print(ast.unparse(node_2))
+            print("---- ---------------- ----")
+        return node_2
+
+    # this does not work as the name might be in a list and one might want to extend the list.
+    # def visit_Name(self, name):
+    #     if name.id in self.mapping and name.id == "__ret__":
+    #         print(name, "in mapping")
+    #         if isinstance(name.ctx, ast.Store):
+    #             return Name("tmp", ctx=Store())
+    #         else:
+    #             return copy.deepcopy(self.mapping[name.id])
+    #     return name
+
+    def visit_Expr(self, expr):
+        if isinstance(expr.value, Name) and expr.value.id in self.mapping:
+            if self.mapping[expr.value.id] is not None:
+                return copy.deepcopy(self.mapping[expr.value.id])
+        return self.generic_visit(expr)

IPython/core/interactiveshell.py

@@ -1,3910 +1,3913 b''
 # -*- coding: utf-8 -*-
 """Main IPython class."""
 
 #-----------------------------------------------------------------------------
 #  Copyright (C) 2001 Janko Hauser <[e-mail]>
 #  Copyright (C) 2001-2007 Fernando Perez. <[e-mail]>
 #  Copyright (C) 2008-2011  The IPython Development Team
 #
 #  Distributed under the terms of the BSD License.  The full license is in
 #  the file COPYING, distributed as part of this software.
 #-----------------------------------------------------------------------------
 
 
 import abc
 import ast
 import atexit
 import bdb
 import builtins as builtin_mod
 import functools
 import inspect
 import os
 import re
 import runpy
 import subprocess
 import sys
 import tempfile
 import traceback
 import types
 import warnings
 from ast import stmt
 from io import open as io_open
 from logging import error
 from pathlib import Path
 from typing import Callable
 from typing import List as ListType, Dict as DictType, Any as AnyType
 from typing import Optional, Sequence, Tuple
 from warnings import warn
 
 from pickleshare import PickleShareDB
 from tempfile import TemporaryDirectory
 from traitlets import (
     Any,
     Bool,
     CaselessStrEnum,
     Dict,
     Enum,
     Instance,
     Integer,
     List,
     Type,
     Unicode,
     default,
     observe,
     validate,
 )
 from traitlets.config.configurable import SingletonConfigurable
 from traitlets.utils.importstring import import_item
 
 import IPython.core.hooks
 from IPython.core import magic, oinspect, page, prefilter, ultratb
 from IPython.core.alias import Alias, AliasManager
 from IPython.core.autocall import ExitAutocall
 from IPython.core.builtin_trap import BuiltinTrap
 from IPython.core.compilerop import CachingCompiler
 from IPython.core.debugger import InterruptiblePdb
 from IPython.core.display_trap import DisplayTrap
 from IPython.core.displayhook import DisplayHook
 from IPython.core.displaypub import DisplayPublisher
 from IPython.core.error import InputRejected, UsageError
 from IPython.core.events import EventManager, available_events
 from IPython.core.extensions import ExtensionManager
 from IPython.core.formatters import DisplayFormatter
 from IPython.core.history import HistoryManager
 from IPython.core.inputtransformer2 import ESC_MAGIC, ESC_MAGIC2
 from IPython.core.logger import Logger
 from IPython.core.macro import Macro
 from IPython.core.payload import PayloadManager
 from IPython.core.prefilter import PrefilterManager
 from IPython.core.profiledir import ProfileDir
 from IPython.core.usage import default_banner
 from IPython.display import display
 from IPython.paths import get_ipython_dir
 from IPython.testing.skipdoctest import skip_doctest
 from IPython.utils import PyColorize, io, openpy, py3compat
 from IPython.utils.decorators import undoc
 from IPython.utils.io import ask_yes_no
 from IPython.utils.ipstruct import Struct
 from IPython.utils.path import ensure_dir_exists, get_home_dir, get_py_filename
 from IPython.utils.process import getoutput, system
 from IPython.utils.strdispatch import StrDispatch
 from IPython.utils.syspathcontext import prepended_to_syspath
 from IPython.utils.text import DollarFormatter, LSString, SList, format_screen
 from IPython.core.oinspect import OInfo
 
 
 sphinxify: Optional[Callable]
 
 try:
     import docrepr.sphinxify as sphx
 
     def sphinxify(oinfo):
         wrapped_docstring = sphx.wrap_main_docstring(oinfo)
 
         def sphinxify_docstring(docstring):
             with TemporaryDirectory() as dirname:
                 return {
                     "text/html": sphx.sphinxify(wrapped_docstring, dirname),
                     "text/plain": docstring,
                 }
 
         return sphinxify_docstring
 except ImportError:
     sphinxify = None
 
 
 class ProvisionalWarning(DeprecationWarning):
     """
     Warning class for unstable features
     """
     pass
 
 from ast import Module
 
 _assign_nodes = (ast.AugAssign, ast.AnnAssign, ast.Assign)
 _single_targets_nodes = (ast.AugAssign, ast.AnnAssign)
 
 #-----------------------------------------------------------------------------
 # Await Helpers
 #-----------------------------------------------------------------------------
 
 # we still need to run things using the asyncio eventloop, but there is no
 # async integration
 from .async_helpers import (
     _asyncio_runner,
     _curio_runner,
     _pseudo_sync_runner,
     _should_be_async,
     _trio_runner,
 )
 
 #-----------------------------------------------------------------------------
 # Globals
 #-----------------------------------------------------------------------------
 
 # compiled regexps for autoindent management
 dedent_re = re.compile(r'^\s+raise|^\s+return|^\s+pass')
 
 #-----------------------------------------------------------------------------
 # Utilities
 #-----------------------------------------------------------------------------
 
 
 def is_integer_string(s: str):
     """
     Variant of "str.isnumeric()" that allow negative values and other ints.
     """
     try:
         int(s)
         return True
     except ValueError:
         return False
     raise ValueError("Unexpected error")
 
 
 @undoc
 def softspace(file, newvalue):
     """Copied from code.py, to remove the dependency"""
 
     oldvalue = 0
     try:
         oldvalue = file.softspace
     except AttributeError:
         pass
     try:
         file.softspace = newvalue
     except (AttributeError, TypeError):
         # "attribute-less object" or "read-only attributes"
         pass
     return oldvalue
 
 @undoc
 def no_op(*a, **kw):
     pass
 
 
 class SpaceInInput(Exception): pass
 
 
 class SeparateUnicode(Unicode):
     r"""A Unicode subclass to validate separate_in, separate_out, etc.
 
     This is a Unicode based trait that converts '0'->'' and ``'\\n'->'\n'``.
     """
 
     def validate(self, obj, value):
         if value == '0': value = ''
         value = value.replace('\\n','\n')
         return super(SeparateUnicode, self).validate(obj, value)
 
 
 @undoc
 class DummyMod(object):
     """A dummy module used for IPython's interactive module when
     a namespace must be assigned to the module's __dict__."""
     __spec__ = None
 
 
 class ExecutionInfo(object):
     """The arguments used for a call to :meth:`InteractiveShell.run_cell`
 
     Stores information about what is going to happen.
     """
     raw_cell = None
     store_history = False
     silent = False
     shell_futures = True
     cell_id = None
 
     def __init__(self, raw_cell, store_history, silent, shell_futures, cell_id):
         self.raw_cell = raw_cell
         self.store_history = store_history
         self.silent = silent
         self.shell_futures = shell_futures
         self.cell_id = cell_id
 
     def __repr__(self):
         name = self.__class__.__qualname__
         raw_cell = (
             (self.raw_cell[:50] + "..") if len(self.raw_cell) > 50 else self.raw_cell
         )
         return (
             '<%s object at %x, raw_cell="%s" store_history=%s silent=%s shell_futures=%s cell_id=%s>'
             % (
                 name,
                 id(self),
                 raw_cell,
                 self.store_history,
                 self.silent,
                 self.shell_futures,
                 self.cell_id,
             )
         )
 
 
 class ExecutionResult(object):
     """The result of a call to :meth:`InteractiveShell.run_cell`
 
     Stores information about what took place.
     """
     execution_count = None
     error_before_exec = None
     error_in_exec: Optional[BaseException] = None
     info = None
     result = None
 
     def __init__(self, info):
         self.info = info
 
     @property
     def success(self):
         return (self.error_before_exec is None) and (self.error_in_exec is None)
 
     def raise_error(self):
         """Reraises error if `success` is `False`, otherwise does nothing"""
         if self.error_before_exec is not None:
             raise self.error_before_exec
         if self.error_in_exec is not None:
             raise self.error_in_exec
 
     def __repr__(self):
         name = self.__class__.__qualname__
         return '<%s object at %x, execution_count=%s error_before_exec=%s error_in_exec=%s info=%s result=%s>' %\
                 (name, id(self), self.execution_count, self.error_before_exec, self.error_in_exec, repr(self.info), repr(self.result))
 
 @functools.wraps(io_open)
 def _modified_open(file, *args, **kwargs):
     if file in {0, 1, 2}:
         raise ValueError(
             f"IPython won't let you open fd={file} by default "
             "as it is likely to crash IPython. If you know what you are doing, "
             "you can use builtins' open."
         )
 
     return io_open(file, *args, **kwargs)
 
 class InteractiveShell(SingletonConfigurable):
     """An enhanced, interactive shell for Python."""
 
     _instance = None
 
     ast_transformers = List([], help=
         """
         A list of ast.NodeTransformer subclass instances, which will be applied
         to user input before code is run.
         """
     ).tag(config=True)
 
     autocall = Enum((0,1,2), default_value=0, help=
         """
         Make IPython automatically call any callable object even if you didn't
         type explicit parentheses. For example, 'str 43' becomes 'str(43)'
         automatically. The value can be '0' to disable the feature, '1' for
         'smart' autocall, where it is not applied if there are no more
         arguments on the line, and '2' for 'full' autocall, where all callable
         objects are automatically called (even if no arguments are present).
         """
     ).tag(config=True)
 
     autoindent = Bool(True, help=
         """
         Autoindent IPython code entered interactively.
         """
     ).tag(config=True)
 
     autoawait = Bool(True, help=
         """
         Automatically run await statement in the top level repl.
         """
     ).tag(config=True)
 
     loop_runner_map ={
         'asyncio':(_asyncio_runner, True),
         'curio':(_curio_runner, True),
         'trio':(_trio_runner, True),
         'sync': (_pseudo_sync_runner, False)
     }
 
     loop_runner = Any(default_value="IPython.core.interactiveshell._asyncio_runner",
         allow_none=True,
         help="""Select the loop runner that will be used to execute top-level asynchronous code"""
     ).tag(config=True)
 
     @default('loop_runner')
     def _default_loop_runner(self):
         return import_item("IPython.core.interactiveshell._asyncio_runner")
 
     @validate('loop_runner')
     def _import_runner(self, proposal):
         if isinstance(proposal.value, str):
             if proposal.value in self.loop_runner_map:
                 runner, autoawait = self.loop_runner_map[proposal.value]
                 self.autoawait = autoawait
                 return runner
             runner = import_item(proposal.value)
             if not callable(runner):
                 raise ValueError('loop_runner must be callable')
             return runner
         if not callable(proposal.value):
             raise ValueError('loop_runner must be callable')
         return proposal.value
 
     automagic = Bool(True, help=
         """
         Enable magic commands to be called without the leading %.
         """
     ).tag(config=True)
 
     banner1 = Unicode(default_banner,
         help="""The part of the banner to be printed before the profile"""
     ).tag(config=True)
     banner2 = Unicode('',
         help="""The part of the banner to be printed after the profile"""
     ).tag(config=True)
 
     cache_size = Integer(1000, help=
         """
         Set the size of the output cache.  The default is 1000, you can
         change it permanently in your config file.  Setting it to 0 completely
         disables the caching system, and the minimum value accepted is 3 (if
         you provide a value less than 3, it is reset to 0 and a warning is
         issued).  This limit is defined because otherwise you'll spend more
         time re-flushing a too small cache than working
         """
     ).tag(config=True)
     color_info = Bool(True, help=
         """
         Use colors for displaying information about objects. Because this
         information is passed through a pager (like 'less'), and some pagers
         get confused with color codes, this capability can be turned off.
         """
     ).tag(config=True)
     colors = CaselessStrEnum(('Neutral', 'NoColor','LightBG','Linux'),
                              default_value='Neutral',
         help="Set the color scheme (NoColor, Neutral, Linux, or LightBG)."
     ).tag(config=True)
     debug = Bool(False).tag(config=True)
     disable_failing_post_execute = Bool(False,
         help="Don't call post-execute functions that have failed in the past."
     ).tag(config=True)
     display_formatter = Instance(DisplayFormatter, allow_none=True)
     displayhook_class = Type(DisplayHook)
     display_pub_class = Type(DisplayPublisher)
     compiler_class = Type(CachingCompiler)
     inspector_class = Type(
         oinspect.Inspector, help="Class to use to instantiate the shell inspector"
     ).tag(config=True)
 
     sphinxify_docstring = Bool(False, help=
         """
         Enables rich html representation of docstrings. (This requires the
         docrepr module).
         """).tag(config=True)
 
     @observe("sphinxify_docstring")
     def _sphinxify_docstring_changed(self, change):
         if change['new']:
             warn("`sphinxify_docstring` is provisional since IPython 5.0 and might change in future versions." , ProvisionalWarning)
 
     enable_html_pager = Bool(False, help=
         """
         (Provisional API) enables html representation in mime bundles sent
         to pagers.
         """).tag(config=True)
 
     @observe("enable_html_pager")
     def _enable_html_pager_changed(self, change):
         if change['new']:
             warn("`enable_html_pager` is provisional since IPython 5.0 and might change in future versions.", ProvisionalWarning)
 
     data_pub_class = None
 
     exit_now = Bool(False)
     exiter = Instance(ExitAutocall)
     @default('exiter')
     def _exiter_default(self):
         return ExitAutocall(self)
     # Monotonically increasing execution counter
     execution_count = Integer(1)
     filename = Unicode("<ipython console>")
     ipython_dir= Unicode('').tag(config=True) # Set to get_ipython_dir() in __init__
 
     # Used to transform cells before running them, and check whether code is complete
     input_transformer_manager = Instance('IPython.core.inputtransformer2.TransformerManager',
                                          ())
 
     @property
     def input_transformers_cleanup(self):
         return self.input_transformer_manager.cleanup_transforms
 
     input_transformers_post = List([],
         help="A list of string input transformers, to be applied after IPython's "
              "own input transformations."
     )
 
     @property
     def input_splitter(self):
         """Make this available for backward compatibility (pre-7.0 release) with existing code.
 
         For example, ipykernel ipykernel currently uses
         `shell.input_splitter.check_complete`
         """
         from warnings import warn
         warn("`input_splitter` is deprecated since IPython 7.0, prefer `input_transformer_manager`.",
              DeprecationWarning, stacklevel=2
         )
         return self.input_transformer_manager
 
     logstart = Bool(False, help=
         """
         Start logging to the default log file in overwrite mode.
         Use `logappend` to specify a log file to **append** logs to.
         """
     ).tag(config=True)
     logfile = Unicode('', help=
         """
         The name of the logfile to use.
         """
     ).tag(config=True)
     logappend = Unicode('', help=
         """
         Start logging to the given file in append mode.
         Use `logfile` to specify a log file to **overwrite** logs to.
         """
     ).tag(config=True)
     object_info_string_level = Enum((0,1,2), default_value=0,
     ).tag(config=True)
     pdb = Bool(False, help=
         """
         Automatically call the pdb debugger after every exception.
         """
     ).tag(config=True)
     display_page = Bool(False,
         help="""If True, anything that would be passed to the pager
         will be displayed as regular output instead."""
     ).tag(config=True)
 
 
     show_rewritten_input = Bool(True,
         help="Show rewritten input, e.g. for autocall."
     ).tag(config=True)
 
     quiet = Bool(False).tag(config=True)
 
     history_length = Integer(10000,
         help='Total length of command history'
     ).tag(config=True)
 
     history_load_length = Integer(1000, help=
         """
         The number of saved history entries to be loaded
         into the history buffer at startup.
         """
     ).tag(config=True)
 
     ast_node_interactivity = Enum(['all', 'last', 'last_expr', 'none', 'last_expr_or_assign'],
                                   default_value='last_expr',
                                   help="""
         'all', 'last', 'last_expr' or 'none', 'last_expr_or_assign' specifying
         which nodes should be run interactively (displaying output from expressions).
         """
     ).tag(config=True)
 
     warn_venv = Bool(
         True,
         help="Warn if running in a virtual environment with no IPython installed (so IPython from the global environment is used).",
     ).tag(config=True)
 
     # TODO: this part of prompt management should be moved to the frontends.
     # Use custom TraitTypes that convert '0'->'' and '\\n'->'\n'
     separate_in = SeparateUnicode('\n').tag(config=True)
     separate_out = SeparateUnicode('').tag(config=True)
     separate_out2 = SeparateUnicode('').tag(config=True)
     wildcards_case_sensitive = Bool(True).tag(config=True)
     xmode = CaselessStrEnum(('Context', 'Plain', 'Verbose', 'Minimal'),
                             default_value='Context',
                             help="Switch modes for the IPython exception handlers."
                             ).tag(config=True)
 
     # Subcomponents of InteractiveShell
     alias_manager = Instance('IPython.core.alias.AliasManager', allow_none=True)
     prefilter_manager = Instance('IPython.core.prefilter.PrefilterManager', allow_none=True)
     builtin_trap = Instance('IPython.core.builtin_trap.BuiltinTrap', allow_none=True)
     display_trap = Instance('IPython.core.display_trap.DisplayTrap', allow_none=True)
     extension_manager = Instance('IPython.core.extensions.ExtensionManager', allow_none=True)
     payload_manager = Instance('IPython.core.payload.PayloadManager', allow_none=True)
     history_manager = Instance('IPython.core.history.HistoryAccessorBase', allow_none=True)
     magics_manager = Instance('IPython.core.magic.MagicsManager', allow_none=True)
 
     profile_dir = Instance('IPython.core.application.ProfileDir', allow_none=True)
     @property
     def profile(self):
         if self.profile_dir is not None:
             name = os.path.basename(self.profile_dir.location)
             return name.replace('profile_','')
 
 
     # Private interface
     _post_execute = Dict()
 
     # Tracks any GUI loop loaded for pylab
     pylab_gui_select = None
 
     last_execution_succeeded = Bool(True, help='Did last executed command succeeded')
 
     last_execution_result = Instance('IPython.core.interactiveshell.ExecutionResult', help='Result of executing the last command', allow_none=True)
 
     def __init__(self, ipython_dir=None, profile_dir=None,
                  user_module=None, user_ns=None,
                  custom_exceptions=((), None), **kwargs):
         # This is where traits with a config_key argument are updated
         # from the values on config.
         super(InteractiveShell, self).__init__(**kwargs)
         if 'PromptManager' in self.config:
             warn('As of IPython 5.0 `PromptManager` config will have no effect'
                  ' and has been replaced by TerminalInteractiveShell.prompts_class')
         self.configurables = [self]
 
         # These are relatively independent and stateless
         self.init_ipython_dir(ipython_dir)
         self.init_profile_dir(profile_dir)
         self.init_instance_attrs()
         self.init_environment()
 
         # Check if we're in a virtualenv, and set up sys.path.
         self.init_virtualenv()
 
         # Create namespaces (user_ns, user_global_ns, etc.)
         self.init_create_namespaces(user_module, user_ns)
         # This has to be done after init_create_namespaces because it uses
         # something in self.user_ns, but before init_sys_modules, which
         # is the first thing to modify sys.
         # TODO: When we override sys.stdout and sys.stderr before this class
         # is created, we are saving the overridden ones here. Not sure if this
         # is what we want to do.
         self.save_sys_module_state()
         self.init_sys_modules()
 
         # While we're trying to have each part of the code directly access what
         # it needs without keeping redundant references to objects, we have too
         # much legacy code that expects ip.db to exist.
         self.db = PickleShareDB(os.path.join(self.profile_dir.location, 'db'))
 
         self.init_history()
         self.init_encoding()
         self.init_prefilter()
 
         self.init_syntax_highlighting()
         self.init_hooks()
         self.init_events()
         self.init_pushd_popd_magic()
         self.init_user_ns()
         self.init_logger()
         self.init_builtins()
 
         # The following was in post_config_initialization
         self.init_inspector()
         self.raw_input_original = input
         self.init_completer()
         # TODO: init_io() needs to happen before init_traceback handlers
         # because the traceback handlers hardcode the stdout/stderr streams.
         # This logic in in debugger.Pdb and should eventually be changed.
         self.init_io()
         self.init_traceback_handlers(custom_exceptions)
         self.init_prompts()
         self.init_display_formatter()
         self.init_display_pub()
         self.init_data_pub()
         self.init_displayhook()
         self.init_magics()
         self.init_alias()
         self.init_logstart()
         self.init_pdb()
         self.init_extension_manager()
         self.init_payload()
         self.events.trigger('shell_initialized', self)
         atexit.register(self.atexit_operations)
 
         # The trio runner is used for running Trio in the foreground thread. It
         # is different from `_trio_runner(async_fn)` in `async_helpers.py`
         # which calls `trio.run()` for every cell. This runner runs all cells
         # inside a single Trio event loop. If used, it is set from
         # `ipykernel.kernelapp`.
         self.trio_runner = None
 
     def get_ipython(self):
         """Return the currently running IPython instance."""
         return self
 
     #-------------------------------------------------------------------------
     # Trait changed handlers
     #-------------------------------------------------------------------------
     @observe('ipython_dir')
     def _ipython_dir_changed(self, change):
         ensure_dir_exists(change['new'])
 
     def set_autoindent(self,value=None):
         """Set the autoindent flag.
 
         If called with no arguments, it acts as a toggle."""
         if value is None:
             self.autoindent = not self.autoindent
         else:
             self.autoindent = value
 
     def set_trio_runner(self, tr):
         self.trio_runner = tr
 
     #-------------------------------------------------------------------------
     # init_* methods called by __init__
     #-------------------------------------------------------------------------
 
     def init_ipython_dir(self, ipython_dir):
         if ipython_dir is not None:
             self.ipython_dir = ipython_dir
             return
 
         self.ipython_dir = get_ipython_dir()
 
     def init_profile_dir(self, profile_dir):
         if profile_dir is not None:
             self.profile_dir = profile_dir
             return
         self.profile_dir = ProfileDir.create_profile_dir_by_name(
             self.ipython_dir, "default"
         )
 
     def init_instance_attrs(self):
         self.more = False
 
         # command compiler
         self.compile = self.compiler_class()
 
         # Make an empty namespace, which extension writers can rely on both
         # existing and NEVER being used by ipython itself.  This gives them a
         # convenient location for storing additional information and state
         # their extensions may require, without fear of collisions with other
         # ipython names that may develop later.
         self.meta = Struct()
 
         # Temporary files used for various purposes.  Deleted at exit.
         # The files here are stored with Path from Pathlib
         self.tempfiles = []
         self.tempdirs = []
 
         # keep track of where we started running (mainly for crash post-mortem)
         # This is not being used anywhere currently.
         self.starting_dir = os.getcwd()
 
         # Indentation management
         self.indent_current_nsp = 0
 
         # Dict to track post-execution functions that have been registered
         self._post_execute = {}
 
     def init_environment(self):
         """Any changes we need to make to the user's environment."""
         pass
 
     def init_encoding(self):
         # Get system encoding at startup time.  Certain terminals (like Emacs
         # under Win32 have it set to None, and we need to have a known valid
         # encoding to use in the raw_input() method
         try:
             self.stdin_encoding = sys.stdin.encoding or 'ascii'
         except AttributeError:
             self.stdin_encoding = 'ascii'
 
 
     @observe('colors')
     def init_syntax_highlighting(self, changes=None):
         # Python source parser/formatter for syntax highlighting
         pyformat = PyColorize.Parser(style=self.colors, parent=self).format
         self.pycolorize = lambda src: pyformat(src,'str')
 
     def refresh_style(self):
         # No-op here, used in subclass
         pass
 
     def init_pushd_popd_magic(self):
         # for pushd/popd management
         self.home_dir = get_home_dir()
 
         self.dir_stack = []
 
     def init_logger(self):
         self.logger = Logger(self.home_dir, logfname='ipython_log.py',
                              logmode='rotate')
 
     def init_logstart(self):
         """Initialize logging in case it was requested at the command line.
         """
         if self.logappend:
             self.magic('logstart %s append' % self.logappend)
         elif self.logfile:
             self.magic('logstart %s' % self.logfile)
         elif self.logstart:
             self.magic('logstart')
 
 
     def init_builtins(self):
         # A single, static flag that we set to True.  Its presence indicates
         # that an IPython shell has been created, and we make no attempts at
         # removing on exit or representing the existence of more than one
         # IPython at a time.
         builtin_mod.__dict__['__IPYTHON__'] = True
         builtin_mod.__dict__['display'] = display
 
         self.builtin_trap = BuiltinTrap(shell=self)
 
     @observe('colors')
     def init_inspector(self, changes=None):
         # Object inspector
         self.inspector = self.inspector_class(
             oinspect.InspectColors,
             PyColorize.ANSICodeColors,
             self.colors,
             self.object_info_string_level,
         )
 
     def init_io(self):
         # implemented in subclasses, TerminalInteractiveShell does call
         # colorama.init().
         pass
 
     def init_prompts(self):
         # Set system prompts, so that scripts can decide if they are running
         # interactively.
         sys.ps1 = 'In : '
         sys.ps2 = '...: '
         sys.ps3 = 'Out: '
 
     def init_display_formatter(self):
         self.display_formatter = DisplayFormatter(parent=self)
         self.configurables.append(self.display_formatter)
 
     def init_display_pub(self):
         self.display_pub = self.display_pub_class(parent=self, shell=self)
         self.configurables.append(self.display_pub)
 
     def init_data_pub(self):
         if not self.data_pub_class:
             self.data_pub = None
             return
         self.data_pub = self.data_pub_class(parent=self)
         self.configurables.append(self.data_pub)
 
     def init_displayhook(self):
         # Initialize displayhook, set in/out prompts and printing system
         self.displayhook = self.displayhook_class(
             parent=self,
             shell=self,
             cache_size=self.cache_size,
         )
         self.configurables.append(self.displayhook)
         # This is a context manager that installs/revmoes the displayhook at
         # the appropriate time.
         self.display_trap = DisplayTrap(hook=self.displayhook)
 
     @staticmethod
     def get_path_links(p: Path):
         """Gets path links including all symlinks
 
         Examples
         --------
         In [1]: from IPython.core.interactiveshell import InteractiveShell
 
         In [2]: import sys, pathlib
 
         In [3]: paths = InteractiveShell.get_path_links(pathlib.Path(sys.executable))
 
         In [4]: len(paths) == len(set(paths))
         Out[4]: True
 
         In [5]: bool(paths)
         Out[5]: True
         """
         paths = [p]
         while p.is_symlink():
             new_path = Path(os.readlink(p))
             if not new_path.is_absolute():
                 new_path = p.parent / new_path
             p = new_path
             paths.append(p)
         return paths
 
     def init_virtualenv(self):
         """Add the current virtualenv to sys.path so the user can import modules from it.
         This isn't perfect: it doesn't use the Python interpreter with which the
         virtualenv was built, and it ignores the --no-site-packages option. A
         warning will appear suggesting the user installs IPython in the
         virtualenv, but for many cases, it probably works well enough.
 
         Adapted from code snippets online.
 
         http://blog.ufsoft.org/2009/1/29/ipython-and-virtualenv
         """
         if 'VIRTUAL_ENV' not in os.environ:
             # Not in a virtualenv
             return
         elif os.environ["VIRTUAL_ENV"] == "":
             warn("Virtual env path set to '', please check if this is intended.")
             return
 
         p = Path(sys.executable)
         p_venv = Path(os.environ["VIRTUAL_ENV"])
 
         # fallback venv detection:
         # stdlib venv may symlink sys.executable, so we can't use realpath.
         # but others can symlink *to* the venv Python, so we can't just use sys.executable.
         # So we just check every item in the symlink tree (generally <= 3)
         paths = self.get_path_links(p)
 
         # In Cygwin paths like "c:\..." and '\cygdrive\c\...' are possible
         if p_venv.parts[1] == "cygdrive":
             drive_name = p_venv.parts[2]
             p_venv = (drive_name + ":/") / Path(*p_venv.parts[3:])
 
         if any(p_venv == p.parents[1] for p in paths):
             # Our exe is inside or has access to the virtualenv, don't need to do anything.
             return
 
         if sys.platform == "win32":
             virtual_env = str(Path(os.environ["VIRTUAL_ENV"], "Lib", "site-packages"))
         else:
             virtual_env_path = Path(
                 os.environ["VIRTUAL_ENV"], "lib", "python{}.{}", "site-packages"
             )
             p_ver = sys.version_info[:2]
 
             # Predict version from py[thon]-x.x in the $VIRTUAL_ENV
             re_m = re.search(r"\bpy(?:thon)?([23])\.(\d+)\b", os.environ["VIRTUAL_ENV"])
             if re_m:
                 predicted_path = Path(str(virtual_env_path).format(*re_m.groups()))
                 if predicted_path.exists():
                     p_ver = re_m.groups()
 
             virtual_env = str(virtual_env_path).format(*p_ver)
         if self.warn_venv:
             warn(
                 "Attempting to work in a virtualenv. If you encounter problems, "
                 "please install IPython inside the virtualenv."
             )
         import site
         sys.path.insert(0, virtual_env)
         site.addsitedir(virtual_env)
 
     #-------------------------------------------------------------------------
     # Things related to injections into the sys module
     #-------------------------------------------------------------------------
 
     def save_sys_module_state(self):
         """Save the state of hooks in the sys module.
 
         This has to be called after self.user_module is created.
         """
         self._orig_sys_module_state = {'stdin': sys.stdin,
                                        'stdout': sys.stdout,
                                        'stderr': sys.stderr,
                                        'excepthook': sys.excepthook}
         self._orig_sys_modules_main_name = self.user_module.__name__
         self._orig_sys_modules_main_mod = sys.modules.get(self.user_module.__name__)
 
     def restore_sys_module_state(self):
         """Restore the state of the sys module."""
         try:
             for k, v in self._orig_sys_module_state.items():
                 setattr(sys, k, v)
         except AttributeError:
             pass
         # Reset what what done in self.init_sys_modules
         if self._orig_sys_modules_main_mod is not None:
             sys.modules[self._orig_sys_modules_main_name] = self._orig_sys_modules_main_mod
 
     #-------------------------------------------------------------------------
     # Things related to the banner
     #-------------------------------------------------------------------------
 
     @property
     def banner(self):
         banner = self.banner1
         if self.profile and self.profile != 'default':
             banner += '\nIPython profile: %s\n' % self.profile
         if self.banner2:
             banner += '\n' + self.banner2
         return banner
 
     def show_banner(self, banner=None):
         if banner is None:
             banner = self.banner
         sys.stdout.write(banner)
 
     #-------------------------------------------------------------------------
     # Things related to hooks
     #-------------------------------------------------------------------------
 
     def init_hooks(self):
         # hooks holds pointers used for user-side customizations
         self.hooks = Struct()
 
         self.strdispatchers = {}
 
         # Set all default hooks, defined in the IPython.hooks module.
         hooks = IPython.core.hooks
         for hook_name in hooks.__all__:
             # default hooks have priority 100, i.e. low; user hooks should have
             # 0-100 priority
             self.set_hook(hook_name, getattr(hooks, hook_name), 100)
 
         if self.display_page:
             self.set_hook('show_in_pager', page.as_hook(page.display_page), 90)
 
     def set_hook(self, name, hook, priority=50, str_key=None, re_key=None):
         """set_hook(name,hook) -> sets an internal IPython hook.
 
         IPython exposes some of its internal API as user-modifiable hooks.  By
         adding your function to one of these hooks, you can modify IPython's
         behavior to call at runtime your own routines."""
 
         # At some point in the future, this should validate the hook before it
         # accepts it.  Probably at least check that the hook takes the number
         # of args it's supposed to.
 
         f = types.MethodType(hook,self)
 
         # check if the hook is for strdispatcher first
         if str_key is not None:
             sdp = self.strdispatchers.get(name, StrDispatch())
             sdp.add_s(str_key, f, priority )
             self.strdispatchers[name] = sdp
             return
         if re_key is not None:
             sdp = self.strdispatchers.get(name, StrDispatch())
             sdp.add_re(re.compile(re_key), f, priority )
             self.strdispatchers[name] = sdp
             return
 
         dp = getattr(self.hooks, name, None)
         if name not in IPython.core.hooks.__all__:
             print("Warning! Hook '%s' is not one of %s" % \
                   (name, IPython.core.hooks.__all__ ))
 
         if name in IPython.core.hooks.deprecated:
             alternative = IPython.core.hooks.deprecated[name]
             raise ValueError(
                 "Hook {} has been deprecated since IPython 5.0. Use {} instead.".format(
                     name, alternative
                 )
             )
 
         if not dp:
             dp = IPython.core.hooks.CommandChainDispatcher()
 
         try:
             dp.add(f,priority)
         except AttributeError:
             # it was not commandchain, plain old func - replace
             dp = f
 
         setattr(self.hooks,name, dp)
 
     #-------------------------------------------------------------------------
     # Things related to events
     #-------------------------------------------------------------------------
 
     def init_events(self):
         self.events = EventManager(self, available_events)
 
         self.events.register("pre_execute", self._clear_warning_registry)
 
     def register_post_execute(self, func):
         """DEPRECATED: Use ip.events.register('post_run_cell', func)
 
         Register a function for calling after code execution.
         """
         raise ValueError(
             "ip.register_post_execute is deprecated since IPython 1.0, use "
             "ip.events.register('post_run_cell', func) instead."
         )
 
     def _clear_warning_registry(self):
         # clear the warning registry, so that different code blocks with
         # overlapping line number ranges don't cause spurious suppression of
         # warnings (see gh-6611 for details)
         if "__warningregistry__" in self.user_global_ns:
             del self.user_global_ns["__warningregistry__"]
 
     #-------------------------------------------------------------------------
     # Things related to the "main" module
     #-------------------------------------------------------------------------
 
     def new_main_mod(self, filename, modname):
         """Return a new 'main' module object for user code execution.
 
         ``filename`` should be the path of the script which will be run in the
         module. Requests with the same filename will get the same module, with
         its namespace cleared.
 
         ``modname`` should be the module name - normally either '__main__' or
         the basename of the file without the extension.
 
         When scripts are executed via %run, we must keep a reference to their
         __main__ module around so that Python doesn't
         clear it, rendering references to module globals useless.
 
         This method keeps said reference in a private dict, keyed by the
         absolute path of the script. This way, for multiple executions of the
         same script we only keep one copy of the namespace (the last one),
         thus preventing memory leaks from old references while allowing the
         objects from the last execution to be accessible.
         """
         filename = os.path.abspath(filename)
         try:
             main_mod = self._main_mod_cache[filename]
         except KeyError:
             main_mod = self._main_mod_cache[filename] = types.ModuleType(
                         modname,
                         doc="Module created for script run in IPython")
         else:
             main_mod.__dict__.clear()
             main_mod.__name__ = modname
 
         main_mod.__file__ = filename
         # It seems pydoc (and perhaps others) needs any module instance to
         # implement a __nonzero__ method
         main_mod.__nonzero__ = lambda : True
 
         return main_mod
 
     def clear_main_mod_cache(self):
         """Clear the cache of main modules.
 
         Mainly for use by utilities like %reset.
 
         Examples
         --------
         In [15]: import IPython
 
         In [16]: m = _ip.new_main_mod(IPython.__file__, 'IPython')
 
         In [17]: len(_ip._main_mod_cache) > 0
         Out[17]: True
 
         In [18]: _ip.clear_main_mod_cache()
 
         In [19]: len(_ip._main_mod_cache) == 0
         Out[19]: True
         """
         self._main_mod_cache.clear()
 
     #-------------------------------------------------------------------------
     # Things related to debugging
     #-------------------------------------------------------------------------
 
     def init_pdb(self):
         # Set calling of pdb on exceptions
         # self.call_pdb is a property
         self.call_pdb = self.pdb
 
     def _get_call_pdb(self):
         return self._call_pdb
 
     def _set_call_pdb(self,val):
 
         if val not in (0,1,False,True):
             raise ValueError('new call_pdb value must be boolean')
 
         # store value in instance
         self._call_pdb = val
 
         # notify the actual exception handlers
         self.InteractiveTB.call_pdb = val
 
     call_pdb = property(_get_call_pdb,_set_call_pdb,None,
                         'Control auto-activation of pdb at exceptions')
 
     def debugger(self,force=False):
         """Call the pdb debugger.
 
         Keywords:
 
           - force(False): by default, this routine checks the instance call_pdb
             flag and does not actually invoke the debugger if the flag is false.
             The 'force' option forces the debugger to activate even if the flag
             is false.
         """
 
         if not (force or self.call_pdb):
             return
 
         if not hasattr(sys,'last_traceback'):
             error('No traceback has been produced, nothing to debug.')
             return
 
         self.InteractiveTB.debugger(force=True)
 
     #-------------------------------------------------------------------------
     # Things related to IPython's various namespaces
     #-------------------------------------------------------------------------
     default_user_namespaces = True
 
     def init_create_namespaces(self, user_module=None, user_ns=None):
         # Create the namespace where the user will operate.  user_ns is
         # normally the only one used, and it is passed to the exec calls as
         # the locals argument.  But we do carry a user_global_ns namespace
         # given as the exec 'globals' argument,  This is useful in embedding
         # situations where the ipython shell opens in a context where the
         # distinction between locals and globals is meaningful.  For
         # non-embedded contexts, it is just the same object as the user_ns dict.
 
         # FIXME. For some strange reason, __builtins__ is showing up at user
         # level as a dict instead of a module. This is a manual fix, but I
         # should really track down where the problem is coming from. Alex
         # Schmolck reported this problem first.
 
         # A useful post by Alex Martelli on this topic:
         # Re: inconsistent value from __builtins__
         # Von: Alex Martelli <[e-mail]>
         # Datum: Freitag 01 Oktober 2004 04:45:34 nachmittags/abends
         # Gruppen: comp.lang.python
 
         # Michael Hohn <[e-mail]> wrote:
         # > >>> print type(builtin_check.get_global_binding('__builtins__'))
         # > <type 'dict'>
         # > >>> print type(__builtins__)
         # > <type 'module'>
         # > Is this difference in return value intentional?
 
         # Well, it's documented that '__builtins__' can be either a dictionary
         # or a module, and it's been that way for a long time. Whether it's
         # intentional (or sensible), I don't know. In any case, the idea is
         # that if you need to access the built-in namespace directly, you
         # should start with "import __builtin__" (note, no 's') which will
         # definitely give you a module. Yeah, it's somewhat confusing:-(.
 
         # These routines return a properly built module and dict as needed by
         # the rest of the code, and can also be used by extension writers to
         # generate properly initialized namespaces.
         if (user_ns is not None) or (user_module is not None):
             self.default_user_namespaces = False
         self.user_module, self.user_ns = self.prepare_user_module(user_module, user_ns)
 
         # A record of hidden variables we have added to the user namespace, so
         # we can list later only variables defined in actual interactive use.
         self.user_ns_hidden = {}
 
         # Now that FakeModule produces a real module, we've run into a nasty
         # problem: after script execution (via %run), the module where the user
         # code ran is deleted.  Now that this object is a true module (needed
         # so doctest and other tools work correctly), the Python module
         # teardown mechanism runs over it, and sets to None every variable
         # present in that module.  Top-level references to objects from the
         # script survive, because the user_ns is updated with them.  However,
         # calling functions defined in the script that use other things from
         # the script will fail, because the function's closure had references
         # to the original objects, which are now all None.  So we must protect
         # these modules from deletion by keeping a cache.
         #
         # To avoid keeping stale modules around (we only need the one from the
         # last run), we use a dict keyed with the full path to the script, so
         # only the last version of the module is held in the cache.  Note,
         # however, that we must cache the module *namespace contents* (their
         # __dict__).  Because if we try to cache the actual modules, old ones
         # (uncached) could be destroyed while still holding references (such as
         # those held by GUI objects that tend to be long-lived)>
         #
         # The %reset command will flush this cache.  See the cache_main_mod()
         # and clear_main_mod_cache() methods for details on use.
 
         # This is the cache used for 'main' namespaces
         self._main_mod_cache = {}
 
         # A table holding all the namespaces IPython deals with, so that
         # introspection facilities can search easily.
         self.ns_table = {'user_global':self.user_module.__dict__,
                          'user_local':self.user_ns,
                          'builtin':builtin_mod.__dict__
                          }
 
     @property
     def user_global_ns(self):
         return self.user_module.__dict__
 
     def prepare_user_module(self, user_module=None, user_ns=None):
         """Prepare the module and namespace in which user code will be run.
 
         When IPython is started normally, both parameters are None: a new module
         is created automatically, and its __dict__ used as the namespace.
 
         If only user_module is provided, its __dict__ is used as the namespace.
         If only user_ns is provided, a dummy module is created, and user_ns
         becomes the global namespace. If both are provided (as they may be
         when embedding), user_ns is the local namespace, and user_module
         provides the global namespace.
 
         Parameters
         ----------
         user_module : module, optional
             The current user module in which IPython is being run. If None,
             a clean module will be created.
         user_ns : dict, optional
             A namespace in which to run interactive commands.
 
         Returns
         -------
         A tuple of user_module and user_ns, each properly initialised.
         """
         if user_module is None and user_ns is not None:
             user_ns.setdefault("__name__", "__main__")
             user_module = DummyMod()
             user_module.__dict__ = user_ns
 
         if user_module is None:
             user_module = types.ModuleType("__main__",
                 doc="Automatically created module for IPython interactive environment")
 
         # We must ensure that __builtin__ (without the final 's') is always
         # available and pointing to the __builtin__ *module*.  For more details:
         # http://mail.python.org/pipermail/python-dev/2001-April/014068.html
         user_module.__dict__.setdefault('__builtin__', builtin_mod)
         user_module.__dict__.setdefault('__builtins__', builtin_mod)
 
         if user_ns is None:
             user_ns = user_module.__dict__
 
         return user_module, user_ns
 
     def init_sys_modules(self):
         # We need to insert into sys.modules something that looks like a
         # module but which accesses the IPython namespace, for shelve and
         # pickle to work interactively. Normally they rely on getting
         # everything out of __main__, but for embedding purposes each IPython
         # instance has its own private namespace, so we can't go shoving
         # everything into __main__.
 
         # note, however, that we should only do this for non-embedded
         # ipythons, which really mimic the __main__.__dict__ with their own
         # namespace.  Embedded instances, on the other hand, should not do
         # this because they need to manage the user local/global namespaces
         # only, but they live within a 'normal' __main__ (meaning, they
         # shouldn't overtake the execution environment of the script they're
         # embedded in).
 
         # This is overridden in the InteractiveShellEmbed subclass to a no-op.
         main_name = self.user_module.__name__
         sys.modules[main_name] = self.user_module
 
     def init_user_ns(self):
         """Initialize all user-visible namespaces to their minimum defaults.
 
         Certain history lists are also initialized here, as they effectively
         act as user namespaces.
 
         Notes
         -----
         All data structures here are only filled in, they are NOT reset by this
         method.  If they were not empty before, data will simply be added to
         them.
         """
         # This function works in two parts: first we put a few things in
         # user_ns, and we sync that contents into user_ns_hidden so that these
         # initial variables aren't shown by %who.  After the sync, we add the
         # rest of what we *do* want the user to see with %who even on a new
         # session (probably nothing, so they really only see their own stuff)
 
         # The user dict must *always* have a __builtin__ reference to the
         # Python standard __builtin__ namespace,  which must be imported.
         # This is so that certain operations in prompt evaluation can be
         # reliably executed with builtins.  Note that we can NOT use
         # __builtins__ (note the 's'),  because that can either be a dict or a
         # module, and can even mutate at runtime, depending on the context
         # (Python makes no guarantees on it).  In contrast, __builtin__ is
         # always a module object, though it must be explicitly imported.
 
         # For more details:
         # http://mail.python.org/pipermail/python-dev/2001-April/014068.html
         ns = {}
 
         # make global variables for user access to the histories
         ns['_ih'] = self.history_manager.input_hist_parsed
         ns['_oh'] = self.history_manager.output_hist
         ns['_dh'] = self.history_manager.dir_hist
 
         # user aliases to input and output histories.  These shouldn't show up
         # in %who, as they can have very large reprs.
         ns['In']  = self.history_manager.input_hist_parsed
         ns['Out'] = self.history_manager.output_hist
 
         # Store myself as the public api!!!
         ns['get_ipython'] = self.get_ipython
 
         ns['exit'] = self.exiter
         ns['quit'] = self.exiter
         ns["open"] = _modified_open
 
         # Sync what we've added so far to user_ns_hidden so these aren't seen
         # by %who
         self.user_ns_hidden.update(ns)
 
         # Anything put into ns now would show up in %who.  Think twice before
         # putting anything here, as we really want %who to show the user their
         # stuff, not our variables.
 
         # Finally, update the real user's namespace
         self.user_ns.update(ns)
 
     @property
     def all_ns_refs(self):
         """Get a list of references to all the namespace dictionaries in which
         IPython might store a user-created object.
 
         Note that this does not include the displayhook, which also caches
         objects from the output."""
         return [self.user_ns, self.user_global_ns, self.user_ns_hidden] + \
                [m.__dict__ for m in self._main_mod_cache.values()]
 
     def reset(self, new_session=True, aggressive=False):
         """Clear all internal namespaces, and attempt to release references to
         user objects.
 
         If new_session is True, a new history session will be opened.
         """
         # Clear histories
         self.history_manager.reset(new_session)
         # Reset counter used to index all histories
         if new_session:
             self.execution_count = 1
 
         # Reset last execution result
         self.last_execution_succeeded = True
         self.last_execution_result = None
 
         # Flush cached output items
         if self.displayhook.do_full_cache:
             self.displayhook.flush()
 
         # The main execution namespaces must be cleared very carefully,
         # skipping the deletion of the builtin-related keys, because doing so
         # would cause errors in many object's __del__ methods.
         if self.user_ns is not self.user_global_ns:
             self.user_ns.clear()
         ns = self.user_global_ns
         drop_keys = set(ns.keys())
         drop_keys.discard('__builtin__')
         drop_keys.discard('__builtins__')
         drop_keys.discard('__name__')
         for k in drop_keys:
             del ns[k]
 
         self.user_ns_hidden.clear()
 
         # Restore the user namespaces to minimal usability
         self.init_user_ns()
         if aggressive and not hasattr(self, "_sys_modules_keys"):
             print("Cannot restore sys.module, no snapshot")
         elif aggressive:
             print("culling sys module...")
             current_keys = set(sys.modules.keys())
             for k in current_keys - self._sys_modules_keys:
                 if k.startswith("multiprocessing"):
                     continue
                 del sys.modules[k]
 
         # Restore the default and user aliases
         self.alias_manager.clear_aliases()
         self.alias_manager.init_aliases()
 
         # Now define aliases that only make sense on the terminal, because they
         # need direct access to the console in a way that we can't emulate in
         # GUI or web frontend
         if os.name == 'posix':
             for cmd in ('clear', 'more', 'less', 'man'):
                 if cmd not in self.magics_manager.magics['line']:
                     self.alias_manager.soft_define_alias(cmd, cmd)
 
         # Flush the private list of module references kept for script
         # execution protection
         self.clear_main_mod_cache()
 
     def del_var(self, varname, by_name=False):
         """Delete a variable from the various namespaces, so that, as
         far as possible, we're not keeping any hidden references to it.
 
         Parameters
         ----------
         varname : str
             The name of the variable to delete.
         by_name : bool
             If True, delete variables with the given name in each
             namespace. If False (default), find the variable in the user
             namespace, and delete references to it.
         """
         if varname in ('__builtin__', '__builtins__'):
             raise ValueError("Refusing to delete %s" % varname)
 
         ns_refs = self.all_ns_refs
 
         if by_name:                    # Delete by name
             for ns in ns_refs:
                 try:
                     del ns[varname]
                 except KeyError:
                     pass
         else:                         # Delete by object
             try:
                 obj = self.user_ns[varname]
             except KeyError as e:
                 raise NameError("name '%s' is not defined" % varname) from e
             # Also check in output history
             ns_refs.append(self.history_manager.output_hist)
             for ns in ns_refs:
                 to_delete = [n for n, o in ns.items() if o is obj]
                 for name in to_delete:
                     del ns[name]
 
             # Ensure it is removed from the last execution result
             if self.last_execution_result.result is obj:
                 self.last_execution_result = None
 
             # displayhook keeps extra references, but not in a dictionary
             for name in ('_', '__', '___'):
                 if getattr(self.displayhook, name) is obj:
                     setattr(self.displayhook, name, None)
 
     def reset_selective(self, regex=None):
         """Clear selective variables from internal namespaces based on a
         specified regular expression.
 
         Parameters
         ----------
         regex : string or compiled pattern, optional
             A regular expression pattern that will be used in searching
             variable names in the users namespaces.
         """
         if regex is not None:
             try:
                 m = re.compile(regex)
             except TypeError as e:
                 raise TypeError('regex must be a string or compiled pattern') from e
             # Search for keys in each namespace that match the given regex
             # If a match is found, delete the key/value pair.
             for ns in self.all_ns_refs:
                 for var in ns:
                     if m.search(var):
                         del ns[var]
 
     def push(self, variables, interactive=True):
         """Inject a group of variables into the IPython user namespace.
 
         Parameters
         ----------
         variables : dict, str or list/tuple of str
             The variables to inject into the user's namespace.  If a dict, a
             simple update is done.  If a str, the string is assumed to have
             variable names separated by spaces.  A list/tuple of str can also
             be used to give the variable names.  If just the variable names are
             give (list/tuple/str) then the variable values looked up in the
             callers frame.
         interactive : bool
             If True (default), the variables will be listed with the ``who``
             magic.
         """
         vdict = None
 
         # We need a dict of name/value pairs to do namespace updates.
         if isinstance(variables, dict):
             vdict = variables
         elif isinstance(variables, (str, list, tuple)):
             if isinstance(variables, str):
                 vlist = variables.split()
             else:
                 vlist = variables
             vdict = {}
             cf = sys._getframe(1)
             for name in vlist:
                 try:
                     vdict[name] = eval(name, cf.f_globals, cf.f_locals)
                 except:
                     print('Could not get variable %s from %s' %
                            (name,cf.f_code.co_name))
         else:
             raise ValueError('variables must be a dict/str/list/tuple')
 
         # Propagate variables to user namespace
         self.user_ns.update(vdict)
 
         # And configure interactive visibility
         user_ns_hidden = self.user_ns_hidden
         if interactive:
             for name in vdict:
                 user_ns_hidden.pop(name, None)
         else:
             user_ns_hidden.update(vdict)
 
     def drop_by_id(self, variables):
         """Remove a dict of variables from the user namespace, if they are the
         same as the values in the dictionary.
 
         This is intended for use by extensions: variables that they've added can
         be taken back out if they are unloaded, without removing any that the
         user has overwritten.
 
         Parameters
         ----------
         variables : dict
             A dictionary mapping object names (as strings) to the objects.
         """
         for name, obj in variables.items():
             if name in self.user_ns and self.user_ns[name] is obj:
                 del self.user_ns[name]
                 self.user_ns_hidden.pop(name, None)
 
     #-------------------------------------------------------------------------
     # Things related to object introspection
     #-------------------------------------------------------------------------
     @staticmethod
     def _find_parts(oname: str) -> Tuple[bool, ListType[str]]:
         """
         Given an object name, return a list of parts of this object name.
 
         Basically split on docs when using attribute access,
         and extract the value when using square bracket.
 
 
         For example foo.bar[3].baz[x] -> foo, bar, 3, baz, x
 
 
         Returns
         -------
         parts_ok: bool
             wether we were properly able to parse parts.
         parts: list of str
             extracted parts
 
 
 
         """
         raw_parts = oname.split(".")
         parts = []
         parts_ok = True
         for p in raw_parts:
             if p.endswith("]"):
                 var, *indices = p.split("[")
                 if not var.isidentifier():
                     parts_ok = False
                     break
                 parts.append(var)
                 for ind in indices:
                     if ind[-1] != "]" and not is_integer_string(ind[:-1]):
                         parts_ok = False
                         break
                     parts.append(ind[:-1])
                 continue
 
             if not p.isidentifier():
                 parts_ok = False
             parts.append(p)
 
         return parts_ok, parts
 
     def _ofind(
         self, oname: str, namespaces: Optional[Sequence[Tuple[str, AnyType]]] = None
     ) -> OInfo:
         """Find an object in the available namespaces.
 
 
         Returns
         -------
         OInfo with fields:
           - ismagic
           - isalias
           - found
           - obj
           - namespac
           - parent
 
         Has special code to detect magic functions.
         """
         oname = oname.strip()
         parts_ok, parts = self._find_parts(oname)
 
         if (
             not oname.startswith(ESC_MAGIC)
             and not oname.startswith(ESC_MAGIC2)
             and not parts_ok
         ):
             return OInfo(
                 ismagic=False,
                 isalias=False,
                 found=False,
                 obj=None,
                 namespace=None,
                 parent=None,
             )
 
         if namespaces is None:
             # Namespaces to search in:
             # Put them in a list. The order is important so that we
             # find things in the same order that Python finds them.
             namespaces = [ ('Interactive', self.user_ns),
                            ('Interactive (global)', self.user_global_ns),
                            ('Python builtin', builtin_mod.__dict__),
                            ]
 
         ismagic = False
         isalias = False
         found = False
         ospace = None
         parent = None
         obj = None
 
 
         # Look for the given name by splitting it in parts.  If the head is
         # found, then we look for all the remaining parts as members, and only
         # declare success if we can find them all.
         oname_parts = parts
         oname_head, oname_rest = oname_parts[0],oname_parts[1:]
         for nsname,ns in namespaces:
             try:
                 obj = ns[oname_head]
             except KeyError:
                 continue
             else:
                 for idx, part in enumerate(oname_rest):
                     try:
                         parent = obj
                         # The last part is looked up in a special way to avoid
                         # descriptor invocation as it may raise or have side
                         # effects.
                         if idx == len(oname_rest) - 1:
                             obj = self._getattr_property(obj, part)
                         else:
                             if is_integer_string(part):
                                 obj = obj[int(part)]
                             else:
                                 obj = getattr(obj, part)
                     except:
                         # Blanket except b/c some badly implemented objects
                         # allow __getattr__ to raise exceptions other than
                         # AttributeError, which then crashes IPython.
                         break
                 else:
                     # If we finish the for loop (no break), we got all members
                     found = True
                     ospace = nsname
                     break  # namespace loop
 
         # Try to see if it's magic
         if not found:
             obj = None
             if oname.startswith(ESC_MAGIC2):
                 oname = oname.lstrip(ESC_MAGIC2)
                 obj = self.find_cell_magic(oname)
             elif oname.startswith(ESC_MAGIC):
                 oname = oname.lstrip(ESC_MAGIC)
                 obj = self.find_line_magic(oname)
             else:
                 # search without prefix, so run? will find %run?
                 obj = self.find_line_magic(oname)
                 if obj is None:
                     obj = self.find_cell_magic(oname)
             if obj is not None:
                 found = True
                 ospace = 'IPython internal'
                 ismagic = True
                 isalias = isinstance(obj, Alias)
 
         # Last try: special-case some literals like '', [], {}, etc:
         if not found and oname_head in ["''",'""','[]','{}','()']:
             obj = eval(oname_head)
             found = True
             ospace = 'Interactive'
 
         return OInfo(
             obj=obj,
             found=found,
             parent=parent,
             ismagic=ismagic,
             isalias=isalias,
             namespace=ospace,
         )
 
     @staticmethod
     def _getattr_property(obj, attrname):
         """Property-aware getattr to use in object finding.
 
         If attrname represents a property, return it unevaluated (in case it has
         side effects or raises an error.
 
         """
         if not isinstance(obj, type):
             try:
                 # `getattr(type(obj), attrname)` is not guaranteed to return
                 # `obj`, but does so for property:
                 #
                 # property.__get__(self, None, cls) -> self
                 #
                 # The universal alternative is to traverse the mro manually
                 # searching for attrname in class dicts.
                 if is_integer_string(attrname):
                     return obj[int(attrname)]
                 else:
                     attr = getattr(type(obj), attrname)
             except AttributeError:
                 pass
             else:
                 # This relies on the fact that data descriptors (with both
                 # __get__ & __set__ magic methods) take precedence over
                 # instance-level attributes:
                 #
                 #    class A(object):
                 #        @property
                 #        def foobar(self): return 123
                 #    a = A()
                 #    a.__dict__['foobar'] = 345
                 #    a.foobar  # == 123
                 #
                 # So, a property may be returned right away.
                 if isinstance(attr, property):
                     return attr
 
         # Nothing helped, fall back.
         return getattr(obj, attrname)
 
     def _object_find(self, oname, namespaces=None) -> OInfo:
         """Find an object and return a struct with info about it."""
         return self._ofind(oname, namespaces)
 
     def _inspect(self, meth, oname, namespaces=None, **kw):
         """Generic interface to the inspector system.
 
         This function is meant to be called by pdef, pdoc & friends.
         """
         info: OInfo = self._object_find(oname, namespaces)
         if self.sphinxify_docstring:
             if sphinxify is None:
                 raise ImportError("Module ``docrepr`` required but missing")
             docformat = sphinxify(self.object_inspect(oname))
         else:
             docformat = None
         if info.found or hasattr(info.parent, oinspect.HOOK_NAME):
             pmethod = getattr(self.inspector, meth)
             # TODO: only apply format_screen to the plain/text repr of the mime
             # bundle.
             formatter = format_screen if info.ismagic else docformat
             if meth == 'pdoc':
                 pmethod(info.obj, oname, formatter)
             elif meth == 'pinfo':
                 pmethod(
                     info.obj,
                     oname,
                     formatter,
                     info,
                     enable_html_pager=self.enable_html_pager,
                     **kw,
                 )
             else:
                 pmethod(info.obj, oname)
         else:
             print('Object `%s` not found.' % oname)
             return 'not found'  # so callers can take other action
 
     def object_inspect(self, oname, detail_level=0):
         """Get object info about oname"""
         with self.builtin_trap:
             info = self._object_find(oname)
             if info.found:
                 return self.inspector.info(info.obj, oname, info=info,
                             detail_level=detail_level
                 )
             else:
                 return oinspect.object_info(name=oname, found=False)
 
     def object_inspect_text(self, oname, detail_level=0):
         """Get object info as formatted text"""
         return self.object_inspect_mime(oname, detail_level)['text/plain']
 
     def object_inspect_mime(self, oname, detail_level=0, omit_sections=()):
         """Get object info as a mimebundle of formatted representations.
 
         A mimebundle is a dictionary, keyed by mime-type.
         It must always have the key `'text/plain'`.
         """
         with self.builtin_trap:
             info = self._object_find(oname)
             if info.found:
                 docformat = (
                     sphinxify(self.object_inspect(oname))
                     if self.sphinxify_docstring
                     else None
                 )
                 return self.inspector._get_info(
                     info.obj,
                     oname,
                     info=info,
                     detail_level=detail_level,
                     formatter=docformat,
                     omit_sections=omit_sections,
                 )
             else:
                 raise KeyError(oname)
 
     #-------------------------------------------------------------------------
     # Things related to history management
     #-------------------------------------------------------------------------
 
     def init_history(self):
         """Sets up the command history, and starts regular autosaves."""
         self.history_manager = HistoryManager(shell=self, parent=self)
         self.configurables.append(self.history_manager)
 
     #-------------------------------------------------------------------------
     # Things related to exception handling and tracebacks (not debugging)
     #-------------------------------------------------------------------------
 
     debugger_cls = InterruptiblePdb
 
     def init_traceback_handlers(self, custom_exceptions):
         # Syntax error handler.
         self.SyntaxTB = ultratb.SyntaxTB(color_scheme='NoColor', parent=self)
 
         # The interactive one is initialized with an offset, meaning we always
         # want to remove the topmost item in the traceback, which is our own
         # internal code. Valid modes: ['Plain','Context','Verbose','Minimal']
         self.InteractiveTB = ultratb.AutoFormattedTB(mode = 'Plain',
                                                      color_scheme='NoColor',
                                                      tb_offset = 1,
                                    debugger_cls=self.debugger_cls, parent=self)
 
         # The instance will store a pointer to the system-wide exception hook,
         # so that runtime code (such as magics) can access it.  This is because
         # during the read-eval loop, it may get temporarily overwritten.
         self.sys_excepthook = sys.excepthook
 
         # and add any custom exception handlers the user may have specified
         self.set_custom_exc(*custom_exceptions)
 
         # Set the exception mode
         self.InteractiveTB.set_mode(mode=self.xmode)
 
     def set_custom_exc(self, exc_tuple, handler):
         """set_custom_exc(exc_tuple, handler)
 
         Set a custom exception handler, which will be called if any of the
         exceptions in exc_tuple occur in the mainloop (specifically, in the
         run_code() method).
 
         Parameters
         ----------
         exc_tuple : tuple of exception classes
             A *tuple* of exception classes, for which to call the defined
             handler.  It is very important that you use a tuple, and NOT A
             LIST here, because of the way Python's except statement works.  If
             you only want to trap a single exception, use a singleton tuple::
 
                 exc_tuple == (MyCustomException,)
 
         handler : callable
             handler must have the following signature::
 
                 def my_handler(self, etype, value, tb, tb_offset=None):
                     ...
                     return structured_traceback
 
             Your handler must return a structured traceback (a list of strings),
             or None.
 
             This will be made into an instance method (via types.MethodType)
             of IPython itself, and it will be called if any of the exceptions
             listed in the exc_tuple are caught. If the handler is None, an
             internal basic one is used, which just prints basic info.
 
             To protect IPython from crashes, if your handler ever raises an
             exception or returns an invalid result, it will be immediately
             disabled.
 
         Notes
         -----
         WARNING: by putting in your own exception handler into IPython's main
         execution loop, you run a very good chance of nasty crashes.  This
         facility should only be used if you really know what you are doing.
         """
 
         if not isinstance(exc_tuple, tuple):
             raise TypeError("The custom exceptions must be given as a tuple.")
 
         def dummy_handler(self, etype, value, tb, tb_offset=None):
             print('*** Simple custom exception handler ***')
             print('Exception type :', etype)
             print('Exception value:', value)
             print('Traceback      :', tb)
 
         def validate_stb(stb):
             """validate structured traceback return type
 
             return type of CustomTB *should* be a list of strings, but allow
             single strings or None, which are harmless.
 
             This function will *always* return a list of strings,
             and will raise a TypeError if stb is inappropriate.
             """
             msg = "CustomTB must return list of strings, not %r" % stb
             if stb is None:
                 return []
             elif isinstance(stb, str):
                 return [stb]
             elif not isinstance(stb, list):
                 raise TypeError(msg)
             # it's a list
             for line in stb:
                 # check every element
                 if not isinstance(line, str):
                     raise TypeError(msg)
             return stb
 
         if handler is None:
             wrapped = dummy_handler
         else:
             def wrapped(self,etype,value,tb,tb_offset=None):
                 """wrap CustomTB handler, to protect IPython from user code
 
                 This makes it harder (but not impossible) for custom exception
                 handlers to crash IPython.
                 """
                 try:
                     stb = handler(self,etype,value,tb,tb_offset=tb_offset)
                     return validate_stb(stb)
                 except:
                     # clear custom handler immediately
                     self.set_custom_exc((), None)
                     print("Custom TB Handler failed, unregistering", file=sys.stderr)
                     # show the exception in handler first
                     stb = self.InteractiveTB.structured_traceback(*sys.exc_info())
                     print(self.InteractiveTB.stb2text(stb))
                     print("The original exception:")
                     stb = self.InteractiveTB.structured_traceback(
                                             (etype,value,tb), tb_offset=tb_offset
                     )
                 return stb
 
         self.CustomTB = types.MethodType(wrapped,self)
         self.custom_exceptions = exc_tuple
 
     def excepthook(self, etype, value, tb):
         """One more defense for GUI apps that call sys.excepthook.
 
         GUI frameworks like wxPython trap exceptions and call
         sys.excepthook themselves.  I guess this is a feature that
         enables them to keep running after exceptions that would
         otherwise kill their mainloop. This is a bother for IPython
         which expects to catch all of the program exceptions with a try:
         except: statement.
 
         Normally, IPython sets sys.excepthook to a CrashHandler instance, so if
         any app directly invokes sys.excepthook, it will look to the user like
         IPython crashed.  In order to work around this, we can disable the
         CrashHandler and replace it with this excepthook instead, which prints a
         regular traceback using our InteractiveTB.  In this fashion, apps which
         call sys.excepthook will generate a regular-looking exception from
         IPython, and the CrashHandler will only be triggered by real IPython
         crashes.
 
         This hook should be used sparingly, only in places which are not likely
         to be true IPython errors.
         """
         self.showtraceback((etype, value, tb), tb_offset=0)
 
     def _get_exc_info(self, exc_tuple=None):
         """get exc_info from a given tuple, sys.exc_info() or sys.last_type etc.
 
         Ensures sys.last_type,value,traceback hold the exc_info we found,
         from whichever source.
 
         raises ValueError if none of these contain any information
         """
         if exc_tuple is None:
             etype, value, tb = sys.exc_info()
         else:
             etype, value, tb = exc_tuple
 
         if etype is None:
             if hasattr(sys, 'last_type'):
                 etype, value, tb = sys.last_type, sys.last_value, \
                                    sys.last_traceback
 
         if etype is None:
             raise ValueError("No exception to find")
 
         # Now store the exception info in sys.last_type etc.
         # WARNING: these variables are somewhat deprecated and not
         # necessarily safe to use in a threaded environment, but tools
         # like pdb depend on their existence, so let's set them.  If we
         # find problems in the field, we'll need to revisit their use.
         sys.last_type = etype
         sys.last_value = value
         sys.last_traceback = tb
 
         return etype, value, tb
 
     def show_usage_error(self, exc):
         """Show a short message for UsageErrors
 
         These are special exceptions that shouldn't show a traceback.
         """
         print("UsageError: %s" % exc, file=sys.stderr)
 
     def get_exception_only(self, exc_tuple=None):
         """
         Return as a string (ending with a newline) the exception that
         just occurred, without any traceback.
         """
         etype, value, tb = self._get_exc_info(exc_tuple)
         msg = traceback.format_exception_only(etype, value)
         return ''.join(msg)
 
     def showtraceback(self, exc_tuple=None, filename=None, tb_offset=None,
                       exception_only=False, running_compiled_code=False):
         """Display the exception that just occurred.
 
         If nothing is known about the exception, this is the method which
         should be used throughout the code for presenting user tracebacks,
         rather than directly invoking the InteractiveTB object.
 
         A specific showsyntaxerror() also exists, but this method can take
         care of calling it if needed, so unless you are explicitly catching a
         SyntaxError exception, don't try to analyze the stack manually and
         simply call this method."""
 
         try:
             try:
                 etype, value, tb = self._get_exc_info(exc_tuple)
             except ValueError:
                 print('No traceback available to show.', file=sys.stderr)
                 return
 
             if issubclass(etype, SyntaxError):
                 # Though this won't be called by syntax errors in the input
                 # line, there may be SyntaxError cases with imported code.
                 self.showsyntaxerror(filename, running_compiled_code)
             elif etype is UsageError:
                 self.show_usage_error(value)
             else:
                 if exception_only:
                     stb = ['An exception has occurred, use %tb to see '
                            'the full traceback.\n']
                     stb.extend(self.InteractiveTB.get_exception_only(etype,
                                                                      value))
                 else:
                     try:
                         # Exception classes can customise their traceback - we
                         # use this in IPython.parallel for exceptions occurring
                         # in the engines. This should return a list of strings.
                         if hasattr(value, "_render_traceback_"):
                             stb = value._render_traceback_()
                         else:
                             stb = self.InteractiveTB.structured_traceback(
                                 etype, value, tb, tb_offset=tb_offset
                             )
 
                     except Exception:
                         print(
                             "Unexpected exception formatting exception. Falling back to standard exception"
                         )
                         traceback.print_exc()
                         return None
 
                     self._showtraceback(etype, value, stb)
                     if self.call_pdb:
                         # drop into debugger
                         self.debugger(force=True)
                     return
 
                 # Actually show the traceback
                 self._showtraceback(etype, value, stb)
 
         except KeyboardInterrupt:
             print('\n' + self.get_exception_only(), file=sys.stderr)
 
     def _showtraceback(self, etype, evalue, stb: str):
         """Actually show a traceback.
 
         Subclasses may override this method to put the traceback on a different
         place, like a side channel.
         """
         val = self.InteractiveTB.stb2text(stb)
         try:
             print(val)
         except UnicodeEncodeError:
             print(val.encode("utf-8", "backslashreplace").decode())
 
     def showsyntaxerror(self, filename=None, running_compiled_code=False):
         """Display the syntax error that just occurred.
 
         This doesn't display a stack trace because there isn't one.
 
         If a filename is given, it is stuffed in the exception instead
         of what was there before (because Python's parser always uses
         "<string>" when reading from a string).
 
         If the syntax error occurred when running a compiled code (i.e. running_compile_code=True),
         longer stack trace will be displayed.
         """
         etype, value, last_traceback = self._get_exc_info()
 
         if filename and issubclass(etype, SyntaxError):
             try:
                 value.filename = filename
             except:
                 # Not the format we expect; leave it alone
                 pass
 
         # If the error occurred when executing compiled code, we should provide full stacktrace.
         elist = traceback.extract_tb(last_traceback) if running_compiled_code else []
         stb = self.SyntaxTB.structured_traceback(etype, value, elist)
         self._showtraceback(etype, value, stb)
 
     # This is overridden in TerminalInteractiveShell to show a message about
     # the %paste magic.
     def showindentationerror(self):
         """Called by _run_cell when there's an IndentationError in code entered
         at the prompt.
 
         This is overridden in TerminalInteractiveShell to show a message about
         the %paste magic."""
         self.showsyntaxerror()
 
     @skip_doctest
     def set_next_input(self, s, replace=False):
         """ Sets the 'default' input string for the next command line.
 
         Example::
 
             In [1]: _ip.set_next_input("Hello Word")
             In [2]: Hello Word_  # cursor is here
         """
         self.rl_next_input = s
 
     def _indent_current_str(self):
         """return the current level of indentation as a string"""
         return self.input_splitter.get_indent_spaces() * ' '
 
     #-------------------------------------------------------------------------
     # Things related to text completion
     #-------------------------------------------------------------------------
 
     def init_completer(self):
         """Initialize the completion machinery.
 
         This creates completion machinery that can be used by client code,
         either interactively in-process (typically triggered by the readline
         library), programmatically (such as in test suites) or out-of-process
         (typically over the network by remote frontends).
         """
         from IPython.core.completer import IPCompleter
         from IPython.core.completerlib import (
             cd_completer,
             magic_run_completer,
             module_completer,
             reset_completer,
         )
 
         self.Completer = IPCompleter(shell=self,
                                      namespace=self.user_ns,
                                      global_namespace=self.user_global_ns,
                                      parent=self,
                                      )
         self.configurables.append(self.Completer)
 
         # Add custom completers to the basic ones built into IPCompleter
         sdisp = self.strdispatchers.get('complete_command', StrDispatch())
         self.strdispatchers['complete_command'] = sdisp
         self.Completer.custom_completers = sdisp
 
         self.set_hook('complete_command', module_completer, str_key = 'import')
         self.set_hook('complete_command', module_completer, str_key = 'from')
         self.set_hook('complete_command', module_completer, str_key = '%aimport')
         self.set_hook('complete_command', magic_run_completer, str_key = '%run')
         self.set_hook('complete_command', cd_completer, str_key = '%cd')
         self.set_hook('complete_command', reset_completer, str_key = '%reset')
 
     @skip_doctest
     def complete(self, text, line=None, cursor_pos=None):
         """Return the completed text and a list of completions.
 
         Parameters
         ----------
         text : string
             A string of text to be completed on.  It can be given as empty and
             instead a line/position pair are given.  In this case, the
             completer itself will split the line like readline does.
         line : string, optional
             The complete line that text is part of.
         cursor_pos : int, optional
             The position of the cursor on the input line.
 
         Returns
         -------
         text : string
             The actual text that was completed.
         matches : list
             A sorted list with all possible completions.
 
         Notes
         -----
         The optional arguments allow the completion to take more context into
         account, and are part of the low-level completion API.
 
         This is a wrapper around the completion mechanism, similar to what
         readline does at the command line when the TAB key is hit.  By
         exposing it as a method, it can be used by other non-readline
         environments (such as GUIs) for text completion.
 
         Examples
         --------
         In [1]: x = 'hello'
 
         In [2]: _ip.complete('x.l')
         Out[2]: ('x.l', ['x.ljust', 'x.lower', 'x.lstrip'])
         """
 
         # Inject names into __builtin__ so we can complete on the added names.
         with self.builtin_trap:
             return self.Completer.complete(text, line, cursor_pos)
 
     def set_custom_completer(self, completer, pos=0) -> None:
         """Adds a new custom completer function.
 
         The position argument (defaults to 0) is the index in the completers
         list where you want the completer to be inserted.
 
         `completer` should have the following signature::
 
             def completion(self: Completer, text: string) -> List[str]:
                 raise NotImplementedError
 
         It will be bound to the current Completer instance and pass some text
         and return a list with current completions to suggest to the user.
         """
 
         newcomp = types.MethodType(completer, self.Completer)
         self.Completer.custom_matchers.insert(pos,newcomp)
 
     def set_completer_frame(self, frame=None):
         """Set the frame of the completer."""
         if frame:
             self.Completer.namespace = frame.f_locals
             self.Completer.global_namespace = frame.f_globals
         else:
             self.Completer.namespace = self.user_ns
             self.Completer.global_namespace = self.user_global_ns
 
     #-------------------------------------------------------------------------
     # Things related to magics
     #-------------------------------------------------------------------------
 
     def init_magics(self):
         from IPython.core import magics as m
         self.magics_manager = magic.MagicsManager(shell=self,
                                    parent=self,
                                    user_magics=m.UserMagics(self))
         self.configurables.append(self.magics_manager)
 
         # Expose as public API from the magics manager
         self.register_magics = self.magics_manager.register
 
         self.register_magics(m.AutoMagics, m.BasicMagics, m.CodeMagics,
             m.ConfigMagics, m.DisplayMagics, m.ExecutionMagics,
             m.ExtensionMagics, m.HistoryMagics, m.LoggingMagics,
             m.NamespaceMagics, m.OSMagics, m.PackagingMagics,
             m.PylabMagics, m.ScriptMagics,
         )
         self.register_magics(m.AsyncMagics)
 
         # Register Magic Aliases
         mman = self.magics_manager
         # FIXME: magic aliases should be defined by the Magics classes
         # or in MagicsManager, not here
         mman.register_alias('ed', 'edit')
         mman.register_alias('hist', 'history')
         mman.register_alias('rep', 'recall')
         mman.register_alias('SVG', 'svg', 'cell')
         mman.register_alias('HTML', 'html', 'cell')
         mman.register_alias('file', 'writefile', 'cell')
 
         # FIXME: Move the color initialization to the DisplayHook, which
         # should be split into a prompt manager and displayhook. We probably
         # even need a centralize colors management object.
         self.run_line_magic('colors', self.colors)
 
     # Defined here so that it's included in the documentation
     @functools.wraps(magic.MagicsManager.register_function)
     def register_magic_function(self, func, magic_kind='line', magic_name=None):
         self.magics_manager.register_function(
             func, magic_kind=magic_kind, magic_name=magic_name
         )
 
     def _find_with_lazy_load(self, /, type_, magic_name: str):
         """
         Try to find a magic potentially lazy-loading it.
 
         Parameters
         ----------
 
         type_: "line"|"cell"
             the type of magics we are trying to find/lazy load.
         magic_name: str
             The name of the magic we are trying to find/lazy load
 
 
         Note that this may have any side effects
         """
         finder = {"line": self.find_line_magic, "cell": self.find_cell_magic}[type_]
         fn = finder(magic_name)
         if fn is not None:
             return fn
         lazy = self.magics_manager.lazy_magics.get(magic_name)
         if lazy is None:
             return None
 
         self.run_line_magic("load_ext", lazy)
         res = finder(magic_name)
         return res
 
     def run_line_magic(self, magic_name: str, line, _stack_depth=1):
         """Execute the given line magic.
 
         Parameters
         ----------
         magic_name : str
             Name of the desired magic function, without '%' prefix.
         line : str
             The rest of the input line as a single string.
         _stack_depth : int
             If run_line_magic() is called from magic() then _stack_depth=2.
             This is added to ensure backward compatibility for use of 'get_ipython().magic()'
         """
         fn = self._find_with_lazy_load("line", magic_name)
         if fn is None:
             lazy = self.magics_manager.lazy_magics.get(magic_name)
             if lazy:
                 self.run_line_magic("load_ext", lazy)
                 fn = self.find_line_magic(magic_name)
         if fn is None:
             cm = self.find_cell_magic(magic_name)
             etpl = "Line magic function `%%%s` not found%s."
             extra = '' if cm is None else (' (But cell magic `%%%%%s` exists, '
                                     'did you mean that instead?)' % magic_name )
             raise UsageError(etpl % (magic_name, extra))
         else:
             # Note: this is the distance in the stack to the user's frame.
             # This will need to be updated if the internal calling logic gets
             # refactored, or else we'll be expanding the wrong variables.
 
             # Determine stack_depth depending on where run_line_magic() has been called
             stack_depth = _stack_depth
             if getattr(fn, magic.MAGIC_NO_VAR_EXPAND_ATTR, False):
                 # magic has opted out of var_expand
                 magic_arg_s = line
             else:
                 magic_arg_s = self.var_expand(line, stack_depth)
             # Put magic args in a list so we can call with f(*a) syntax
             args = [magic_arg_s]
             kwargs = {}
             # Grab local namespace if we need it:
             if getattr(fn, "needs_local_scope", False):
                 kwargs['local_ns'] = self.get_local_scope(stack_depth)
             with self.builtin_trap:
                 result = fn(*args, **kwargs)
 
             # The code below prevents the output from being displayed
             # when using magics with decodator @output_can_be_silenced
             # when the last Python token in the expression is a ';'.
             if getattr(fn, magic.MAGIC_OUTPUT_CAN_BE_SILENCED, False):
                 if DisplayHook.semicolon_at_end_of_expression(magic_arg_s):
                     return None
 
             return result
 
     def get_local_scope(self, stack_depth):
         """Get local scope at given stack depth.
 
         Parameters
         ----------
         stack_depth : int
             Depth relative to calling frame
         """
         return sys._getframe(stack_depth + 1).f_locals
 
     def run_cell_magic(self, magic_name, line, cell):
         """Execute the given cell magic.
 
         Parameters
         ----------
         magic_name : str
             Name of the desired magic function, without '%' prefix.
         line : str
             The rest of the first input line as a single string.
         cell : str
             The body of the cell as a (possibly multiline) string.
         """
         fn = self._find_with_lazy_load("cell", magic_name)
         if fn is None:
             lm = self.find_line_magic(magic_name)
             etpl = "Cell magic `%%{0}` not found{1}."
             extra = '' if lm is None else (' (But line magic `%{0}` exists, '
                             'did you mean that instead?)'.format(magic_name))
             raise UsageError(etpl.format(magic_name, extra))
         elif cell == '':
             message = '%%{0} is a cell magic, but the cell body is empty.'.format(magic_name)
             if self.find_line_magic(magic_name) is not None:
                 message += ' Did you mean the line magic %{0} (single %)?'.format(magic_name)
             raise UsageError(message)
         else:
             # Note: this is the distance in the stack to the user's frame.
             # This will need to be updated if the internal calling logic gets
             # refactored, or else we'll be expanding the wrong variables.
             stack_depth = 2
             if getattr(fn, magic.MAGIC_NO_VAR_EXPAND_ATTR, False):
                 # magic has opted out of var_expand
                 magic_arg_s = line
             else:
                 magic_arg_s = self.var_expand(line, stack_depth)
             kwargs = {}
             if getattr(fn, "needs_local_scope", False):
                 kwargs['local_ns'] = self.user_ns
 
             with self.builtin_trap:
                 args = (magic_arg_s, cell)
                 result = fn(*args, **kwargs)
 
             # The code below prevents the output from being displayed
             # when using magics with decodator @output_can_be_silenced
             # when the last Python token in the expression is a ';'.
             if getattr(fn, magic.MAGIC_OUTPUT_CAN_BE_SILENCED, False):
                 if DisplayHook.semicolon_at_end_of_expression(cell):
                     return None
 
             return result
 
     def find_line_magic(self, magic_name):
         """Find and return a line magic by name.
 
         Returns None if the magic isn't found."""
         return self.magics_manager.magics['line'].get(magic_name)
 
     def find_cell_magic(self, magic_name):
         """Find and return a cell magic by name.
 
         Returns None if the magic isn't found."""
         return self.magics_manager.magics['cell'].get(magic_name)
 
     def find_magic(self, magic_name, magic_kind='line'):
         """Find and return a magic of the given type by name.
 
         Returns None if the magic isn't found."""
         return self.magics_manager.magics[magic_kind].get(magic_name)
 
     def magic(self, arg_s):
         """
         DEPRECATED
 
         Deprecated since IPython 0.13 (warning added in
         8.1), use run_line_magic(magic_name, parameter_s).
 
         Call a magic function by name.
 
         Input: a string containing the name of the magic function to call and
         any additional arguments to be passed to the magic.
 
         magic('name -opt foo bar') is equivalent to typing at the ipython
         prompt:
 
         In[1]: %name -opt foo bar
 
         To call a magic without arguments, simply use magic('name').
 
         This provides a proper Python function to call IPython's magics in any
         valid Python code you can type at the interpreter, including loops and
         compound statements.
         """
         warnings.warn(
             "`magic(...)` is deprecated since IPython 0.13 (warning added in "
             "8.1), use run_line_magic(magic_name, parameter_s).",
             DeprecationWarning,
             stacklevel=2,
         )
         # TODO: should we issue a loud deprecation warning here?
         magic_name, _, magic_arg_s = arg_s.partition(' ')
         magic_name = magic_name.lstrip(prefilter.ESC_MAGIC)
         return self.run_line_magic(magic_name, magic_arg_s, _stack_depth=2)
 
     #-------------------------------------------------------------------------
     # Things related to macros
     #-------------------------------------------------------------------------
 
     def define_macro(self, name, themacro):
         """Define a new macro
 
         Parameters
         ----------
         name : str
             The name of the macro.
         themacro : str or Macro
             The action to do upon invoking the macro.  If a string, a new
             Macro object is created by passing the string to it.
         """
 
         from IPython.core import macro
 
         if isinstance(themacro, str):
             themacro = macro.Macro(themacro)
         if not isinstance(themacro, macro.Macro):
             raise ValueError('A macro must be a string or a Macro instance.')
         self.user_ns[name] = themacro
 
     #-------------------------------------------------------------------------
     # Things related to the running of system commands
     #-------------------------------------------------------------------------
 
     def system_piped(self, cmd):
         """Call the given cmd in a subprocess, piping stdout/err
 
         Parameters
         ----------
         cmd : str
             Command to execute (can not end in '&', as background processes are
             not supported.  Should not be a command that expects input
             other than simple text.
         """
         if cmd.rstrip().endswith('&'):
             # this is *far* from a rigorous test
             # We do not support backgrounding processes because we either use
             # pexpect or pipes to read from.  Users can always just call
             # os.system() or use ip.system=ip.system_raw
             # if they really want a background process.
             raise OSError("Background processes not supported.")
 
         # we explicitly do NOT return the subprocess status code, because
         # a non-None value would trigger :func:`sys.displayhook` calls.
         # Instead, we store the exit_code in user_ns.
         self.user_ns['_exit_code'] = system(self.var_expand(cmd, depth=1))
 
     def system_raw(self, cmd):
         """Call the given cmd in a subprocess using os.system on Windows or
         subprocess.call using the system shell on other platforms.
 
         Parameters
         ----------
         cmd : str
             Command to execute.
         """
         cmd = self.var_expand(cmd, depth=1)
         # warn if there is an IPython magic alternative.
         main_cmd = cmd.split()[0]
         has_magic_alternatives = ("pip", "conda", "cd")
 
         if main_cmd in has_magic_alternatives:
             warnings.warn(
                 (
                     "You executed the system command !{0} which may not work "
                     "as expected. Try the IPython magic %{0} instead."
                 ).format(main_cmd)
             )
 
         # protect os.system from UNC paths on Windows, which it can't handle:
         if sys.platform == 'win32':
             from IPython.utils._process_win32 import AvoidUNCPath
             with AvoidUNCPath() as path:
                 if path is not None:
                     cmd = '"pushd %s &&"%s' % (path, cmd)
                 try:
                     ec = os.system(cmd)
                 except KeyboardInterrupt:
                     print('\n' + self.get_exception_only(), file=sys.stderr)
                     ec = -2
         else:
             # For posix the result of the subprocess.call() below is an exit
             # code, which by convention is zero for success, positive for
             # program failure.  Exit codes above 128 are reserved for signals,
             # and the formula for converting a signal to an exit code is usually
             # signal_number+128.  To more easily differentiate between exit
             # codes and signals, ipython uses negative numbers.  For instance
             # since control-c is signal 2 but exit code 130, ipython's
             # _exit_code variable will read -2.  Note that some shells like
             # csh and fish don't follow sh/bash conventions for exit codes.
             executable = os.environ.get('SHELL', None)
             try:
                 # Use env shell instead of default /bin/sh
                 ec = subprocess.call(cmd, shell=True, executable=executable)
             except KeyboardInterrupt:
                 # intercept control-C; a long traceback is not useful here
                 print('\n' + self.get_exception_only(), file=sys.stderr)
                 ec = 130
             if ec > 128:
                 ec = -(ec - 128)
 
         # We explicitly do NOT return the subprocess status code, because
         # a non-None value would trigger :func:`sys.displayhook` calls.
         # Instead, we store the exit_code in user_ns.  Note the semantics
         # of _exit_code: for control-c, _exit_code == -signal.SIGNIT,
         # but raising SystemExit(_exit_code) will give status 254!
         self.user_ns['_exit_code'] = ec
 
     # use piped system by default, because it is better behaved
     system = system_piped
 
     def getoutput(self, cmd, split=True, depth=0):
         """Get output (possibly including stderr) from a subprocess.
 
         Parameters
         ----------
         cmd : str
             Command to execute (can not end in '&', as background processes are
             not supported.
         split : bool, optional
             If True, split the output into an IPython SList.  Otherwise, an
             IPython LSString is returned.  These are objects similar to normal
             lists and strings, with a few convenience attributes for easier
             manipulation of line-based output.  You can use '?' on them for
             details.
         depth : int, optional
             How many frames above the caller are the local variables which should
             be expanded in the command string? The default (0) assumes that the
             expansion variables are in the stack frame calling this function.
         """
         if cmd.rstrip().endswith('&'):
             # this is *far* from a rigorous test
             raise OSError("Background processes not supported.")
         out = getoutput(self.var_expand(cmd, depth=depth+1))
         if split:
             out = SList(out.splitlines())
         else:
             out = LSString(out)
         return out
 
     #-------------------------------------------------------------------------
     # Things related to aliases
     #-------------------------------------------------------------------------
 
     def init_alias(self):
         self.alias_manager = AliasManager(shell=self, parent=self)
         self.configurables.append(self.alias_manager)
 
     #-------------------------------------------------------------------------
     # Things related to extensions
     #-------------------------------------------------------------------------
 
     def init_extension_manager(self):
         self.extension_manager = ExtensionManager(shell=self, parent=self)
         self.configurables.append(self.extension_manager)
 
     #-------------------------------------------------------------------------
     # Things related to payloads
     #-------------------------------------------------------------------------
 
     def init_payload(self):
         self.payload_manager = PayloadManager(parent=self)
         self.configurables.append(self.payload_manager)
 
     #-------------------------------------------------------------------------
     # Things related to the prefilter
     #-------------------------------------------------------------------------
 
     def init_prefilter(self):
         self.prefilter_manager = PrefilterManager(shell=self, parent=self)
         self.configurables.append(self.prefilter_manager)
         # Ultimately this will be refactored in the new interpreter code, but
         # for now, we should expose the main prefilter method (there's legacy
         # code out there that may rely on this).
         self.prefilter = self.prefilter_manager.prefilter_lines
 
     def auto_rewrite_input(self, cmd):
         """Print to the screen the rewritten form of the user's command.
 
         This shows visual feedback by rewriting input lines that cause
         automatic calling to kick in, like::
 
           /f x
 
         into::
 
           ------> f(x)
 
         after the user's input prompt.  This helps the user understand that the
         input line was transformed automatically by IPython.
         """
         if not self.show_rewritten_input:
             return
 
         # This is overridden in TerminalInteractiveShell to use fancy prompts
         print("------> " + cmd)
 
     #-------------------------------------------------------------------------
     # Things related to extracting values/expressions from kernel and user_ns
     #-------------------------------------------------------------------------
 
     def _user_obj_error(self):
         """return simple exception dict
 
         for use in user_expressions
         """
 
         etype, evalue, tb = self._get_exc_info()
         stb = self.InteractiveTB.get_exception_only(etype, evalue)
 
         exc_info = {
             "status": "error",
             "traceback": stb,
             "ename": etype.__name__,
             "evalue": py3compat.safe_unicode(evalue),
         }
 
         return exc_info
 
     def _format_user_obj(self, obj):
         """format a user object to display dict
 
         for use in user_expressions
         """
 
         data, md = self.display_formatter.format(obj)
         value = {
             'status' : 'ok',
             'data' : data,
             'metadata' : md,
         }
         return value
 
     def user_expressions(self, expressions):
         """Evaluate a dict of expressions in the user's namespace.
 
         Parameters
         ----------
         expressions : dict
             A dict with string keys and string values.  The expression values
             should be valid Python expressions, each of which will be evaluated
             in the user namespace.
 
         Returns
         -------
         A dict, keyed like the input expressions dict, with the rich mime-typed
         display_data of each value.
         """
         out = {}
         user_ns = self.user_ns
         global_ns = self.user_global_ns
 
         for key, expr in expressions.items():
             try:
                 value = self._format_user_obj(eval(expr, global_ns, user_ns))
             except:
                 value = self._user_obj_error()
             out[key] = value
         return out
 
     #-------------------------------------------------------------------------
     # Things related to the running of code
     #-------------------------------------------------------------------------
 
     def ex(self, cmd):
         """Execute a normal python statement in user namespace."""
         with self.builtin_trap:
             exec(cmd, self.user_global_ns, self.user_ns)
 
     def ev(self, expr):
         """Evaluate python expression expr in user namespace.
 
         Returns the result of evaluation
         """
         with self.builtin_trap:
             return eval(expr, self.user_global_ns, self.user_ns)
 
     def safe_execfile(self, fname, *where, exit_ignore=False, raise_exceptions=False, shell_futures=False):
         """A safe version of the builtin execfile().
 
         This version will never throw an exception, but instead print
         helpful error messages to the screen.  This only works on pure
         Python files with the .py extension.
 
         Parameters
         ----------
         fname : string
             The name of the file to be executed.
         *where : tuple
             One or two namespaces, passed to execfile() as (globals,locals).
             If only one is given, it is passed as both.
         exit_ignore : bool (False)
             If True, then silence SystemExit for non-zero status (it is always
             silenced for zero status, as it is so common).
         raise_exceptions : bool (False)
             If True raise exceptions everywhere. Meant for testing.
         shell_futures : bool (False)
             If True, the code will share future statements with the interactive
             shell. It will both be affected by previous __future__ imports, and
             any __future__ imports in the code will affect the shell. If False,
             __future__ imports are not shared in either direction.
 
         """
         fname = Path(fname).expanduser().resolve()
 
         # Make sure we can open the file
         try:
             with fname.open("rb"):
                 pass
         except:
             warn('Could not open file <%s> for safe execution.' % fname)
             return
 
         # Find things also in current directory.  This is needed to mimic the
         # behavior of running a script from the system command line, where
         # Python inserts the script's directory into sys.path
         dname = str(fname.parent)
 
         with prepended_to_syspath(dname), self.builtin_trap:
             try:
                 glob, loc = (where + (None, ))[:2]
                 py3compat.execfile(
                     fname, glob, loc,
                     self.compile if shell_futures else None)
             except SystemExit as status:
                 # If the call was made with 0 or None exit status (sys.exit(0)
                 # or sys.exit() ), don't bother showing a traceback, as both of
                 # these are considered normal by the OS:
                 # > python -c'import sys;sys.exit(0)'; echo $?
                 # 0
                 # > python -c'import sys;sys.exit()'; echo $?
                 # 0
                 # For other exit status, we show the exception unless
                 # explicitly silenced, but only in short form.
                 if status.code:
                     if raise_exceptions:
                         raise
                     if not exit_ignore:
                         self.showtraceback(exception_only=True)
             except:
                 if raise_exceptions:
                     raise
                 # tb offset is 2 because we wrap execfile
                 self.showtraceback(tb_offset=2)
 
     def safe_execfile_ipy(self, fname, shell_futures=False, raise_exceptions=False):
         """Like safe_execfile, but for .ipy or .ipynb files with IPython syntax.
 
         Parameters
         ----------
         fname : str
             The name of the file to execute.  The filename must have a
             .ipy or .ipynb extension.
         shell_futures : bool (False)
             If True, the code will share future statements with the interactive
             shell. It will both be affected by previous __future__ imports, and
             any __future__ imports in the code will affect the shell. If False,
             __future__ imports are not shared in either direction.
         raise_exceptions : bool (False)
             If True raise exceptions everywhere.  Meant for testing.
         """
         fname = Path(fname).expanduser().resolve()
 
         # Make sure we can open the file
         try:
             with fname.open("rb"):
                 pass
         except:
             warn('Could not open file <%s> for safe execution.' % fname)
             return
 
         # Find things also in current directory.  This is needed to mimic the
         # behavior of running a script from the system command line, where
         # Python inserts the script's directory into sys.path
         dname = str(fname.parent)
 
         def get_cells():
             """generator for sequence of code blocks to run"""
             if fname.suffix == ".ipynb":
                 from nbformat import read
                 nb = read(fname, as_version=4)
                 if not nb.cells:
                     return
                 for cell in nb.cells:
                     if cell.cell_type == 'code':
                         yield cell.source
             else:
                 yield fname.read_text(encoding="utf-8")
 
         with prepended_to_syspath(dname):
             try:
                 for cell in get_cells():
                     result = self.run_cell(cell, silent=True, shell_futures=shell_futures)
                     if raise_exceptions:
                         result.raise_error()
                     elif not result.success:
                         break
             except:
                 if raise_exceptions:
                     raise
                 self.showtraceback()
                 warn('Unknown failure executing file: <%s>' % fname)
 
     def safe_run_module(self, mod_name, where):
         """A safe version of runpy.run_module().
 
         This version will never throw an exception, but instead print
         helpful error messages to the screen.
 
         `SystemExit` exceptions with status code 0 or None are ignored.
 
         Parameters
         ----------
         mod_name : string
             The name of the module to be executed.
         where : dict
             The globals namespace.
         """
         try:
             try:
                 where.update(
                     runpy.run_module(str(mod_name), run_name="__main__",
                                      alter_sys=True)
                             )
             except SystemExit as status:
                 if status.code:
                     raise
         except:
             self.showtraceback()
             warn('Unknown failure executing module: <%s>' % mod_name)
 
     def run_cell(
         self,
         raw_cell,
         store_history=False,
         silent=False,
         shell_futures=True,
         cell_id=None,
     ):
         """Run a complete IPython cell.
 
         Parameters
         ----------
         raw_cell : str
             The code (including IPython code such as %magic functions) to run.
         store_history : bool
             If True, the raw and translated cell will be stored in IPython's
             history. For user code calling back into IPython's machinery, this
             should be set to False.
         silent : bool
             If True, avoid side-effects, such as implicit displayhooks and
             and logging.  silent=True forces store_history=False.
         shell_futures : bool
             If True, the code will share future statements with the interactive
             shell. It will both be affected by previous __future__ imports, and
             any __future__ imports in the code will affect the shell. If False,
             __future__ imports are not shared in either direction.
 
         Returns
         -------
         result : :class:`ExecutionResult`
         """
         result = None
         try:
             result = self._run_cell(
                 raw_cell, store_history, silent, shell_futures, cell_id
             )
         finally:
             self.events.trigger('post_execute')
             if not silent:
                 self.events.trigger('post_run_cell', result)
         return result
 
     def _run_cell(
         self,
         raw_cell: str,
         store_history: bool,
         silent: bool,
         shell_futures: bool,
         cell_id: str,
     ) -> ExecutionResult:
         """Internal method to run a complete IPython cell."""
 
         # we need to avoid calling self.transform_cell multiple time on the same thing
         # so we need to store some results:
         preprocessing_exc_tuple = None
         try:
             transformed_cell = self.transform_cell(raw_cell)
         except Exception:
             transformed_cell = raw_cell
             preprocessing_exc_tuple = sys.exc_info()
 
         assert transformed_cell is not None
         coro = self.run_cell_async(
             raw_cell,
             store_history=store_history,
             silent=silent,
             shell_futures=shell_futures,
             transformed_cell=transformed_cell,
             preprocessing_exc_tuple=preprocessing_exc_tuple,
             cell_id=cell_id,
         )
 
         # run_cell_async is async, but may not actually need an eventloop.
         # when this is the case, we want to run it using the pseudo_sync_runner
         # so that code can invoke eventloops (for example via the %run , and
         # `%paste` magic.
         if self.trio_runner:
             runner = self.trio_runner
         elif self.should_run_async(
             raw_cell,
             transformed_cell=transformed_cell,
             preprocessing_exc_tuple=preprocessing_exc_tuple,
         ):
             runner = self.loop_runner
         else:
             runner = _pseudo_sync_runner
 
         try:
             result = runner(coro)
         except BaseException as e:
             info = ExecutionInfo(
                 raw_cell, store_history, silent, shell_futures, cell_id
             )
             result = ExecutionResult(info)
             result.error_in_exec = e
             self.showtraceback(running_compiled_code=True)
         finally:
             return result
 
     def should_run_async(
         self, raw_cell: str, *, transformed_cell=None, preprocessing_exc_tuple=None
     ) -> bool:
         """Return whether a cell should be run asynchronously via a coroutine runner
 
         Parameters
         ----------
         raw_cell : str
             The code to be executed
 
         Returns
         -------
         result: bool
             Whether the code needs to be run with a coroutine runner or not
         .. versionadded:: 7.0
         """
         if not self.autoawait:
             return False
         if preprocessing_exc_tuple is not None:
             return False
         assert preprocessing_exc_tuple is None
         if transformed_cell is None:
             warnings.warn(
                 "`should_run_async` will not call `transform_cell`"
                 " automatically in the future. Please pass the result to"
                 " `transformed_cell` argument and any exception that happen"
                 " during the"
                 "transform in `preprocessing_exc_tuple` in"
                 " IPython 7.17 and above.",
                 DeprecationWarning,
                 stacklevel=2,
             )
             try:
                 cell = self.transform_cell(raw_cell)
             except Exception:
                 # any exception during transform will be raised
                 # prior to execution
                 return False
         else:
             cell = transformed_cell
         return _should_be_async(cell)
 
     async def run_cell_async(
         self,
         raw_cell: str,
         store_history=False,
         silent=False,
         shell_futures=True,
         *,
         transformed_cell: Optional[str] = None,
         preprocessing_exc_tuple: Optional[AnyType] = None,
         cell_id=None,
     ) -> ExecutionResult:
         """Run a complete IPython cell asynchronously.
 
         Parameters
         ----------
         raw_cell : str
           The code (including IPython code such as %magic functions) to run.
         store_history : bool
           If True, the raw and translated cell will be stored in IPython's
           history. For user code calling back into IPython's machinery, this
           should be set to False.
         silent : bool
           If True, avoid side-effects, such as implicit displayhooks and
           and logging.  silent=True forces store_history=False.
         shell_futures : bool
           If True, the code will share future statements with the interactive
           shell. It will both be affected by previous __future__ imports, and
           any __future__ imports in the code will affect the shell. If False,
           __future__ imports are not shared in either direction.
         transformed_cell: str
           cell that was passed through transformers
         preprocessing_exc_tuple:
           trace if the transformation failed.
 
         Returns
         -------
         result : :class:`ExecutionResult`
 
         .. versionadded:: 7.0
         """
         info = ExecutionInfo(raw_cell, store_history, silent, shell_futures, cell_id)
         result = ExecutionResult(info)
 
         if (not raw_cell) or raw_cell.isspace():
             self.last_execution_succeeded = True
             self.last_execution_result = result
             return result
 
         if silent:
             store_history = False
 
         if store_history:
             result.execution_count = self.execution_count
 
         def error_before_exec(value):
             if store_history:
                 self.execution_count += 1
             result.error_before_exec = value
             self.last_execution_succeeded = False
             self.last_execution_result = result
             return result
 
         self.events.trigger('pre_execute')
         if not silent:
             self.events.trigger('pre_run_cell', info)
 
         if transformed_cell is None:
             warnings.warn(
                 "`run_cell_async` will not call `transform_cell`"
                 " automatically in the future. Please pass the result to"
                 " `transformed_cell` argument and any exception that happen"
                 " during the"
                 "transform in `preprocessing_exc_tuple` in"
                 " IPython 7.17 and above.",
                 DeprecationWarning,
                 stacklevel=2,
             )
             # If any of our input transformation (input_transformer_manager or
             # prefilter_manager) raises an exception, we store it in this variable
             # so that we can display the error after logging the input and storing
             # it in the history.
             try:
                 cell = self.transform_cell(raw_cell)
             except Exception:
                 preprocessing_exc_tuple = sys.exc_info()
                 cell = raw_cell  # cell has to exist so it can be stored/logged
             else:
                 preprocessing_exc_tuple = None
         else:
             if preprocessing_exc_tuple is None:
                 cell = transformed_cell
             else:
                 cell = raw_cell
 
         # Do NOT store paste/cpaste magic history
         if "get_ipython().run_line_magic(" in cell and "paste" in cell:
             store_history = False
 
         # Store raw and processed history
         if store_history:
             self.history_manager.store_inputs(self.execution_count, cell, raw_cell)
         if not silent:
             self.logger.log(cell, raw_cell)
 
         # Display the exception if input processing failed.
         if preprocessing_exc_tuple is not None:
             self.showtraceback(preprocessing_exc_tuple)
             if store_history:
                 self.execution_count += 1
             return error_before_exec(preprocessing_exc_tuple[1])
 
         # Our own compiler remembers the __future__ environment. If we want to
         # run code with a separate __future__ environment, use the default
         # compiler
         compiler = self.compile if shell_futures else self.compiler_class()
 
         _run_async = False
 
         with self.builtin_trap:
             cell_name = compiler.cache(cell, self.execution_count, raw_code=raw_cell)
 
             with self.display_trap:
                 # Compile to bytecode
                 try:
                     code_ast = compiler.ast_parse(cell, filename=cell_name)
                 except self.custom_exceptions as e:
                     etype, value, tb = sys.exc_info()
                     self.CustomTB(etype, value, tb)
                     return error_before_exec(e)
                 except IndentationError as e:
                     self.showindentationerror()
                     return error_before_exec(e)
                 except (OverflowError, SyntaxError, ValueError, TypeError,
                         MemoryError) as e:
                     self.showsyntaxerror()
                     return error_before_exec(e)
 
                 # Apply AST transformations
                 try:
                     code_ast = self.transform_ast(code_ast)
                 except InputRejected as e:
                     self.showtraceback()
                     return error_before_exec(e)
 
                 # Give the displayhook a reference to our ExecutionResult so it
                 # can fill in the output value.
                 self.displayhook.exec_result = result
 
                 # Execute the user code
                 interactivity = "none" if silent else self.ast_node_interactivity
 
 
                 has_raised = await self.run_ast_nodes(code_ast.body, cell_name,
                        interactivity=interactivity, compiler=compiler, result=result)
 
                 self.last_execution_succeeded = not has_raised
                 self.last_execution_result = result
 
                 # Reset this so later displayed values do not modify the
                 # ExecutionResult
                 self.displayhook.exec_result = None
 
         if store_history:
             # Write output to the database. Does nothing unless
             # history output logging is enabled.
             self.history_manager.store_output(self.execution_count)
             # Each cell is a *single* input, regardless of how many lines it has
             self.execution_count += 1
 
         return result
 
     def transform_cell(self, raw_cell):
         """Transform an input cell before parsing it.
 
         Static transformations, implemented in IPython.core.inputtransformer2,
         deal with things like ``%magic`` and ``!system`` commands.
         These run on all input.
         Dynamic transformations, for things like unescaped magics and the exit
         autocall, depend on the state of the interpreter.
         These only apply to single line inputs.
 
         These string-based transformations are followed by AST transformations;
         see :meth:`transform_ast`.
         """
         # Static input transformations
         cell = self.input_transformer_manager.transform_cell(raw_cell)
 
         if len(cell.splitlines()) == 1:
             # Dynamic transformations - only applied for single line commands
             with self.builtin_trap:
                 # use prefilter_lines to handle trailing newlines
                 # restore trailing newline for ast.parse
                 cell = self.prefilter_manager.prefilter_lines(cell) + '\n'
 
         lines = cell.splitlines(keepends=True)
         for transform in self.input_transformers_post:
             lines = transform(lines)
         cell = ''.join(lines)
 
         return cell
 
     def transform_ast(self, node):
         """Apply the AST transformations from self.ast_transformers
 
         Parameters
         ----------
         node : ast.Node
             The root node to be transformed. Typically called with the ast.Module
             produced by parsing user input.
 
         Returns
         -------
         An ast.Node corresponding to the node it was called with. Note that it
         may also modify the passed object, so don't rely on references to the
         original AST.
         """
         for transformer in self.ast_transformers:
             try:
                 node = transformer.visit(node)
             except InputRejected:
                 # User-supplied AST transformers can reject an input by raising
                 # an InputRejected.  Short-circuit in this case so that we
                 # don't unregister the transform.
                 raise
-            except Exception:
-                warn("AST transformer %r threw an error. It will be unregistered." % transformer)
+            except Exception as e:
+                warn(
+                    "AST transformer %r threw an error. It will be unregistered. %s"
+                    % (transformer, e)
+                )
                 self.ast_transformers.remove(transformer)
 
         if self.ast_transformers:
             ast.fix_missing_locations(node)
         return node
 
     async def run_ast_nodes(
         self,
         nodelist: ListType[stmt],
         cell_name: str,
         interactivity="last_expr",
         compiler=compile,
         result=None,
     ):
         """Run a sequence of AST nodes. The execution mode depends on the
         interactivity parameter.
 
         Parameters
         ----------
         nodelist : list
           A sequence of AST nodes to run.
         cell_name : str
           Will be passed to the compiler as the filename of the cell. Typically
           the value returned by ip.compile.cache(cell).
         interactivity : str
           'all', 'last', 'last_expr' , 'last_expr_or_assign' or 'none',
           specifying which nodes should be run interactively (displaying output
           from expressions). 'last_expr' will run the last node interactively
           only if it is an expression (i.e. expressions in loops or other blocks
           are not displayed) 'last_expr_or_assign' will run the last expression
           or the last assignment. Other values for this parameter will raise a
           ValueError.
 
         compiler : callable
           A function with the same interface as the built-in compile(), to turn
           the AST nodes into code objects. Default is the built-in compile().
         result : ExecutionResult, optional
           An object to store exceptions that occur during execution.
 
         Returns
         -------
         True if an exception occurred while running code, False if it finished
         running.
         """
         if not nodelist:
             return
 
 
         if interactivity == 'last_expr_or_assign':
             if isinstance(nodelist[-1], _assign_nodes):
                 asg = nodelist[-1]
                 if isinstance(asg, ast.Assign) and len(asg.targets) == 1:
                     target = asg.targets[0]
                 elif isinstance(asg, _single_targets_nodes):
                     target = asg.target
                 else:
                     target = None
                 if isinstance(target, ast.Name):
                     nnode = ast.Expr(ast.Name(target.id, ast.Load()))
                     ast.fix_missing_locations(nnode)
                     nodelist.append(nnode)
             interactivity = 'last_expr'
 
         _async = False
         if interactivity == 'last_expr':
             if isinstance(nodelist[-1], ast.Expr):
                 interactivity = "last"
             else:
                 interactivity = "none"
 
         if interactivity == 'none':
             to_run_exec, to_run_interactive = nodelist, []
         elif interactivity == 'last':
             to_run_exec, to_run_interactive = nodelist[:-1], nodelist[-1:]
         elif interactivity == 'all':
             to_run_exec, to_run_interactive = [], nodelist
         else:
             raise ValueError("Interactivity was %r" % interactivity)
 
         try:
 
             def compare(code):
                 is_async = inspect.CO_COROUTINE & code.co_flags == inspect.CO_COROUTINE
                 return is_async
 
             # refactor that to just change the mod constructor.
             to_run = []
             for node in to_run_exec:
                 to_run.append((node, "exec"))
 
             for node in to_run_interactive:
                 to_run.append((node, "single"))
 
             for node, mode in to_run:
                 if mode == "exec":
                     mod = Module([node], [])
                 elif mode == "single":
                     mod = ast.Interactive([node])  # type: ignore
                 with compiler.extra_flags(
                     getattr(ast, "PyCF_ALLOW_TOP_LEVEL_AWAIT", 0x0)
                     if self.autoawait
                     else 0x0
                 ):
                     code = compiler(mod, cell_name, mode)
                     asy = compare(code)
                 if await self.run_code(code, result, async_=asy):
                     return True
 
             # Flush softspace
             if softspace(sys.stdout, 0):
                 print()
 
         except:
             # It's possible to have exceptions raised here, typically by
             # compilation of odd code (such as a naked 'return' outside a
             # function) that did parse but isn't valid. Typically the exception
             # is a SyntaxError, but it's safest just to catch anything and show
             # the user a traceback.
 
             # We do only one try/except outside the loop to minimize the impact
             # on runtime, and also because if any node in the node list is
             # broken, we should stop execution completely.
             if result:
                 result.error_before_exec = sys.exc_info()[1]
             self.showtraceback()
             return True
 
         return False
 
     async def run_code(self, code_obj, result=None, *, async_=False):
         """Execute a code object.
 
         When an exception occurs, self.showtraceback() is called to display a
         traceback.
 
         Parameters
         ----------
         code_obj : code object
           A compiled code object, to be executed
         result : ExecutionResult, optional
           An object to store exceptions that occur during execution.
         async_ :  Bool (Experimental)
           Attempt to run top-level asynchronous code in a default loop.
 
         Returns
         -------
         False : successful execution.
         True : an error occurred.
         """
         # special value to say that anything above is IPython and should be
         # hidden.
         __tracebackhide__ = "__ipython_bottom__"
         # Set our own excepthook in case the user code tries to call it
         # directly, so that the IPython crash handler doesn't get triggered
         old_excepthook, sys.excepthook = sys.excepthook, self.excepthook
 
         # we save the original sys.excepthook in the instance, in case config
         # code (such as magics) needs access to it.
         self.sys_excepthook = old_excepthook
         outflag = True  # happens in more places, so it's easier as default
         try:
             try:
                 if async_:
                     await eval(code_obj, self.user_global_ns, self.user_ns)
                 else:
                     exec(code_obj, self.user_global_ns, self.user_ns)
             finally:
                 # Reset our crash handler in place
                 sys.excepthook = old_excepthook
         except SystemExit as e:
             if result is not None:
                 result.error_in_exec = e
             self.showtraceback(exception_only=True)
             warn("To exit: use 'exit', 'quit', or Ctrl-D.", stacklevel=1)
         except bdb.BdbQuit:
             etype, value, tb = sys.exc_info()
             if result is not None:
                 result.error_in_exec = value
             # the BdbQuit stops here
         except self.custom_exceptions:
             etype, value, tb = sys.exc_info()
             if result is not None:
                 result.error_in_exec = value
             self.CustomTB(etype, value, tb)
         except:
             if result is not None:
                 result.error_in_exec = sys.exc_info()[1]
             self.showtraceback(running_compiled_code=True)
         else:
             outflag = False
         return outflag
 
     # For backwards compatibility
     runcode = run_code
 
     def check_complete(self, code: str) -> Tuple[str, str]:
         """Return whether a block of code is ready to execute, or should be continued
 
         Parameters
         ----------
         code : string
             Python input code, which can be multiline.
 
         Returns
         -------
         status : str
             One of 'complete', 'incomplete', or 'invalid' if source is not a
             prefix of valid code.
         indent : str
             When status is 'incomplete', this is some whitespace to insert on
             the next line of the prompt.
         """
         status, nspaces = self.input_transformer_manager.check_complete(code)
         return status, ' ' * (nspaces or 0)
 
     #-------------------------------------------------------------------------
     # Things related to GUI support and pylab
     #-------------------------------------------------------------------------
 
     active_eventloop = None
 
     def enable_gui(self, gui=None):
         raise NotImplementedError('Implement enable_gui in a subclass')
 
     def enable_matplotlib(self, gui=None):
         """Enable interactive matplotlib and inline figure support.
 
         This takes the following steps:
 
         1. select the appropriate eventloop and matplotlib backend
         2. set up matplotlib for interactive use with that backend
         3. configure formatters for inline figure display
         4. enable the selected gui eventloop
 
         Parameters
         ----------
         gui : optional, string
             If given, dictates the choice of matplotlib GUI backend to use
             (should be one of IPython's supported backends, 'qt', 'osx', 'tk',
             'gtk', 'wx' or 'inline'), otherwise we use the default chosen by
             matplotlib (as dictated by the matplotlib build-time options plus the
             user's matplotlibrc configuration file).  Note that not all backends
             make sense in all contexts, for example a terminal ipython can't
             display figures inline.
         """
         from matplotlib_inline.backend_inline import configure_inline_support
 
         from IPython.core import pylabtools as pt
         gui, backend = pt.find_gui_and_backend(gui, self.pylab_gui_select)
 
         if gui != 'inline':
             # If we have our first gui selection, store it
             if self.pylab_gui_select is None:
                 self.pylab_gui_select = gui
             # Otherwise if they are different
             elif gui != self.pylab_gui_select:
                 print('Warning: Cannot change to a different GUI toolkit: %s.'
                         ' Using %s instead.' % (gui, self.pylab_gui_select))
                 gui, backend = pt.find_gui_and_backend(self.pylab_gui_select)
 
         pt.activate_matplotlib(backend)
         configure_inline_support(self, backend)
 
         # Now we must activate the gui pylab wants to use, and fix %run to take
         # plot updates into account
         self.enable_gui(gui)
         self.magics_manager.registry['ExecutionMagics'].default_runner = \
             pt.mpl_runner(self.safe_execfile)
 
         return gui, backend
 
     def enable_pylab(self, gui=None, import_all=True, welcome_message=False):
         """Activate pylab support at runtime.
 
         This turns on support for matplotlib, preloads into the interactive
         namespace all of numpy and pylab, and configures IPython to correctly
         interact with the GUI event loop.  The GUI backend to be used can be
         optionally selected with the optional ``gui`` argument.
 
         This method only adds preloading the namespace to InteractiveShell.enable_matplotlib.
 
         Parameters
         ----------
         gui : optional, string
             If given, dictates the choice of matplotlib GUI backend to use
             (should be one of IPython's supported backends, 'qt', 'osx', 'tk',
             'gtk', 'wx' or 'inline'), otherwise we use the default chosen by
             matplotlib (as dictated by the matplotlib build-time options plus the
             user's matplotlibrc configuration file).  Note that not all backends
             make sense in all contexts, for example a terminal ipython can't
             display figures inline.
         import_all : optional, bool, default: True
             Whether to do `from numpy import *` and `from pylab import *`
             in addition to module imports.
         welcome_message : deprecated
             This argument is ignored, no welcome message will be displayed.
         """
         from IPython.core.pylabtools import import_pylab
 
         gui, backend = self.enable_matplotlib(gui)
 
         # We want to prevent the loading of pylab to pollute the user's
         # namespace as shown by the %who* magics, so we execute the activation
         # code in an empty namespace, and we update *both* user_ns and
         # user_ns_hidden with this information.
         ns = {}
         import_pylab(ns, import_all)
         # warn about clobbered names
         ignored = {"__builtins__"}
         both = set(ns).intersection(self.user_ns).difference(ignored)
         clobbered = [ name for name in both if self.user_ns[name] is not ns[name] ]
         self.user_ns.update(ns)
         self.user_ns_hidden.update(ns)
         return gui, backend, clobbered
 
     #-------------------------------------------------------------------------
     # Utilities
     #-------------------------------------------------------------------------
 
     def var_expand(self, cmd, depth=0, formatter=DollarFormatter()):
         """Expand python variables in a string.
 
         The depth argument indicates how many frames above the caller should
         be walked to look for the local namespace where to expand variables.
 
         The global namespace for expansion is always the user's interactive
         namespace.
         """
         ns = self.user_ns.copy()
         try:
             frame = sys._getframe(depth+1)
         except ValueError:
             # This is thrown if there aren't that many frames on the stack,
             # e.g. if a script called run_line_magic() directly.
             pass
         else:
             ns.update(frame.f_locals)
 
         try:
             # We have to use .vformat() here, because 'self' is a valid and common
             # name, and expanding **ns for .format() would make it collide with
             # the 'self' argument of the method.
             cmd = formatter.vformat(cmd, args=[], kwargs=ns)
         except Exception:
             # if formatter couldn't format, just let it go untransformed
             pass
         return cmd
 
     def mktempfile(self, data=None, prefix='ipython_edit_'):
         """Make a new tempfile and return its filename.
 
         This makes a call to tempfile.mkstemp (created in a tempfile.mkdtemp),
         but it registers the created filename internally so ipython cleans it up
         at exit time.
 
         Optional inputs:
 
           - data(None): if data is given, it gets written out to the temp file
             immediately, and the file is closed again."""
 
         dir_path = Path(tempfile.mkdtemp(prefix=prefix))
         self.tempdirs.append(dir_path)
 
         handle, filename = tempfile.mkstemp(".py", prefix, dir=str(dir_path))
         os.close(handle)  # On Windows, there can only be one open handle on a file
 
         file_path = Path(filename)
         self.tempfiles.append(file_path)
 
         if data:
             file_path.write_text(data, encoding="utf-8")
         return filename
 
     def ask_yes_no(self, prompt, default=None, interrupt=None):
         if self.quiet:
             return True
         return ask_yes_no(prompt,default,interrupt)
 
     def show_usage(self):
         """Show a usage message"""
         page.page(IPython.core.usage.interactive_usage)
 
     def extract_input_lines(self, range_str, raw=False):
         """Return as a string a set of input history slices.
 
         Parameters
         ----------
         range_str : str
             The set of slices is given as a string, like "~5/6-~4/2 4:8 9",
             since this function is for use by magic functions which get their
             arguments as strings. The number before the / is the session
             number: ~n goes n back from the current session.
 
             If empty string is given, returns history of current session
             without the last input.
 
         raw : bool, optional
             By default, the processed input is used.  If this is true, the raw
             input history is used instead.
 
         Notes
         -----
         Slices can be described with two notations:
 
         * ``N:M`` -> standard python form, means including items N...(M-1).
         * ``N-M`` -> include items N..M (closed endpoint).
         """
         lines = self.history_manager.get_range_by_str(range_str, raw=raw)
         text = "\n".join(x for _, _, x in lines)
 
         # Skip the last line, as it's probably the magic that called this
         if not range_str:
             if "\n" not in text:
                 text = ""
             else:
                 text = text[: text.rfind("\n")]
 
         return text
 
     def find_user_code(self, target, raw=True, py_only=False, skip_encoding_cookie=True, search_ns=False):
         """Get a code string from history, file, url, or a string or macro.
 
         This is mainly used by magic functions.
 
         Parameters
         ----------
         target : str
             A string specifying code to retrieve. This will be tried respectively
             as: ranges of input history (see %history for syntax), url,
             corresponding .py file, filename, or an expression evaluating to a
             string or Macro in the user namespace.
 
             If empty string is given, returns complete history of current
             session, without the last line.
 
         raw : bool
             If true (default), retrieve raw history. Has no effect on the other
             retrieval mechanisms.
 
         py_only : bool (default False)
             Only try to fetch python code, do not try alternative methods to decode file
             if unicode fails.
 
         Returns
         -------
         A string of code.
         ValueError is raised if nothing is found, and TypeError if it evaluates
         to an object of another type. In each case, .args[0] is a printable
         message.
         """
         code = self.extract_input_lines(target, raw=raw)  # Grab history
         if code:
             return code
         try:
             if target.startswith(('http://', 'https://')):
                 return openpy.read_py_url(target, skip_encoding_cookie=skip_encoding_cookie)
         except UnicodeDecodeError as e:
             if not py_only :
                 # Deferred import
                 from urllib.request import urlopen
                 response = urlopen(target)
                 return response.read().decode('latin1')
             raise ValueError(("'%s' seem to be unreadable.") % target) from e
 
         potential_target = [target]
         try :
             potential_target.insert(0,get_py_filename(target))
         except IOError:
             pass
 
         for tgt in potential_target :
             if os.path.isfile(tgt):                        # Read file
                 try :
                     return openpy.read_py_file(tgt, skip_encoding_cookie=skip_encoding_cookie)
                 except UnicodeDecodeError as e:
                     if not py_only :
                         with io_open(tgt,'r', encoding='latin1') as f :
                             return f.read()
                     raise ValueError(("'%s' seem to be unreadable.") % target) from e
             elif os.path.isdir(os.path.expanduser(tgt)):
                 raise ValueError("'%s' is a directory, not a regular file." % target)
 
         if search_ns:
             # Inspect namespace to load object source
             object_info = self.object_inspect(target, detail_level=1)
             if object_info['found'] and object_info['source']:
                 return object_info['source']
 
         try:                                              # User namespace
             codeobj = eval(target, self.user_ns)
         except Exception as e:
             raise ValueError(("'%s' was not found in history, as a file, url, "
                                 "nor in the user namespace.") % target) from e
 
         if isinstance(codeobj, str):
             return codeobj
         elif isinstance(codeobj, Macro):
             return codeobj.value
 
         raise TypeError("%s is neither a string nor a macro." % target,
                         codeobj)
 
     def _atexit_once(self):
         """
         At exist operation that need to be called at most once.
         Second call to this function per instance will do nothing.
         """
 
         if not getattr(self, "_atexit_once_called", False):
             self._atexit_once_called = True
             # Clear all user namespaces to release all references cleanly.
             self.reset(new_session=False)
             # Close the history session (this stores the end time and line count)
             # this must be *before* the tempfile cleanup, in case of temporary
             # history db
             self.history_manager.end_session()
             self.history_manager = None
 
     #-------------------------------------------------------------------------
     # Things related to IPython exiting
     #-------------------------------------------------------------------------
     def atexit_operations(self):
         """This will be executed at the time of exit.
 
         Cleanup operations and saving of persistent data that is done
         unconditionally by IPython should be performed here.
 
         For things that may depend on startup flags or platform specifics (such
         as having readline or not), register a separate atexit function in the
         code that has the appropriate information, rather than trying to
         clutter
         """
         self._atexit_once()
 
         # Cleanup all tempfiles and folders left around
         for tfile in self.tempfiles:
             try:
                 tfile.unlink()
                 self.tempfiles.remove(tfile)
             except FileNotFoundError:
                 pass
         del self.tempfiles
         for tdir in self.tempdirs:
             try:
                 tdir.rmdir()
                 self.tempdirs.remove(tdir)
             except FileNotFoundError:
                 pass
         del self.tempdirs
 
         # Restore user's cursor
         if hasattr(self, "editing_mode") and self.editing_mode == "vi":
             sys.stdout.write("\x1b[0 q")
             sys.stdout.flush()
 
     def cleanup(self):
         self.restore_sys_module_state()
 
 
     # Overridden in terminal subclass to change prompts
     def switch_doctest_mode(self, mode):
         pass
 
 
 class InteractiveShellABC(metaclass=abc.ABCMeta):
     """An abstract base class for InteractiveShell."""
 
 InteractiveShellABC.register(InteractiveShell)

IPython/core/magics/execution.py

@@ -1,1522 +1,1614 b''
 # -*- coding: utf-8 -*-
 """Implementation of execution-related magic functions."""
 
 # Copyright (c) IPython Development Team.
 # Distributed under the terms of the Modified BSD License.
 
 
 import ast
 import bdb
 import builtins as builtin_mod
+import copy
 import cProfile as profile
 import gc
 import itertools
 import math
 import os
 import pstats
 import re
 import shlex
 import sys
 import time
 import timeit
-from ast import Module
+from typing import Dict, Any
+from ast import (
+    Assign,
+    Call,
+    Expr,
+    Load,
+    Module,
+    Name,
+    NodeTransformer,
+    Store,
+    parse,
+    unparse,
+)
 from io import StringIO
 from logging import error
 from pathlib import Path
 from pdb import Restart
+from textwrap import dedent, indent
 from warnings import warn
 
 from IPython.core import magic_arguments, oinspect, page
+from IPython.core.displayhook import DisplayHook
 from IPython.core.error import UsageError
 from IPython.core.macro import Macro
 from IPython.core.magic import (
     Magics,
     cell_magic,
     line_cell_magic,
     line_magic,
     magics_class,
     needs_local_scope,
     no_var_expand,
-    output_can_be_silenced,
     on_off,
+    output_can_be_silenced,
 )
 from IPython.testing.skipdoctest import skip_doctest
 from IPython.utils.capture import capture_output
 from IPython.utils.contexts import preserve_keys
 from IPython.utils.ipstruct import Struct
 from IPython.utils.module_paths import find_mod
 from IPython.utils.path import get_py_filename, shellglob
 from IPython.utils.timing import clock, clock2
-from IPython.core.displayhook import DisplayHook
+from IPython.core.magics.ast_mod import ReplaceCodeTransformer
 
 #-----------------------------------------------------------------------------
 # Magic implementation classes
 #-----------------------------------------------------------------------------
 
 
 class TimeitResult(object):
     """
     Object returned by the timeit magic with info about the run.
 
     Contains the following attributes :
 
     loops: (int) number of loops done per measurement
     repeat: (int) number of times the measurement has been repeated
     best: (float) best execution time / number
     all_runs: (list of float) execution time of each run (in s)
     compile_time: (float) time of statement compilation (s)
 
     """
     def __init__(self, loops, repeat, best, worst, all_runs, compile_time, precision):
         self.loops = loops
         self.repeat = repeat
         self.best = best
         self.worst = worst
         self.all_runs = all_runs
         self.compile_time = compile_time
         self._precision = precision
         self.timings = [ dt / self.loops for dt in all_runs]
 
     @property
     def average(self):
         return math.fsum(self.timings) / len(self.timings)
 
     @property
     def stdev(self):
         mean = self.average
         return (math.fsum([(x - mean) ** 2 for x in self.timings]) / len(self.timings)) ** 0.5
 
     def __str__(self):
         pm = '+-'
         if hasattr(sys.stdout, 'encoding') and sys.stdout.encoding:
             try:
                 u'\xb1'.encode(sys.stdout.encoding)
                 pm = u'\xb1'
             except:
                 pass
         return "{mean} {pm} {std} per loop (mean {pm} std. dev. of {runs} run{run_plural}, {loops:,} loop{loop_plural} each)".format(
             pm=pm,
             runs=self.repeat,
             loops=self.loops,
             loop_plural="" if self.loops == 1 else "s",
             run_plural="" if self.repeat == 1 else "s",
             mean=_format_time(self.average, self._precision),
             std=_format_time(self.stdev, self._precision),
         )
 
     def _repr_pretty_(self, p , cycle):
         unic = self.__str__()
         p.text(u'<TimeitResult : '+unic+u'>')
 
 
 class TimeitTemplateFiller(ast.NodeTransformer):
     """Fill in the AST template for timing execution.
 
     This is quite closely tied to the template definition, which is in
     :meth:`ExecutionMagics.timeit`.
     """
     def __init__(self, ast_setup, ast_stmt):
         self.ast_setup = ast_setup
         self.ast_stmt = ast_stmt
 
     def visit_FunctionDef(self, node):
         "Fill in the setup statement"
         self.generic_visit(node)
         if node.name == "inner":
             node.body[:1] = self.ast_setup.body
 
         return node
 
     def visit_For(self, node):
         "Fill in the statement to be timed"
         if getattr(getattr(node.body[0], 'value', None), 'id', None) == 'stmt':
             node.body = self.ast_stmt.body
         return node
 
 
 class Timer(timeit.Timer):
     """Timer class that explicitly uses self.inner
     
     which is an undocumented implementation detail of CPython,
     not shared by PyPy.
     """
     # Timer.timeit copied from CPython 3.4.2
     def timeit(self, number=timeit.default_number):
         """Time 'number' executions of the main statement.
 
         To be precise, this executes the setup statement once, and
         then returns the time it takes to execute the main statement
         a number of times, as a float measured in seconds.  The
         argument is the number of times through the loop, defaulting
         to one million.  The main statement, the setup statement and
         the timer function to be used are passed to the constructor.
         """
         it = itertools.repeat(None, number)
         gcold = gc.isenabled()
         gc.disable()
         try:
             timing = self.inner(it, self.timer)
         finally:
             if gcold:
                 gc.enable()
         return timing
 
 
 @magics_class
 class ExecutionMagics(Magics):
-    """Magics related to code execution, debugging, profiling, etc.
+    """Magics related to code execution, debugging, profiling, etc."""
 
-    """
+    _transformers: Dict[str, Any] = {}
 
     def __init__(self, shell):
         super(ExecutionMagics, self).__init__(shell)
         # Default execution function used to actually run user code.
         self.default_runner = None
 
     @skip_doctest
     @no_var_expand
     @line_cell_magic
     def prun(self, parameter_s='', cell=None):
 
         """Run a statement through the python code profiler.
 
         Usage, in line mode:
           %prun [options] statement
 
         Usage, in cell mode:
           %%prun [options] [statement]
           code...
           code...
 
         In cell mode, the additional code lines are appended to the (possibly
         empty) statement in the first line.  Cell mode allows you to easily
         profile multiline blocks without having to put them in a separate
         function.
 
         The given statement (which doesn't require quote marks) is run via the
         python profiler in a manner similar to the profile.run() function.
         Namespaces are internally managed to work correctly; profile.run
         cannot be used in IPython because it makes certain assumptions about
         namespaces which do not hold under IPython.
 
         Options:
 
         -l <limit>
           you can place restrictions on what or how much of the
           profile gets printed. The limit value can be:
 
              * A string: only information for function names containing this string
                is printed.
 
              * An integer: only these many lines are printed.
 
              * A float (between 0 and 1): this fraction of the report is printed
                (for example, use a limit of 0.4 to see the topmost 40% only).
 
           You can combine several limits with repeated use of the option. For
           example, ``-l __init__ -l 5`` will print only the topmost 5 lines of
           information about class constructors.
 
         -r
           return the pstats.Stats object generated by the profiling. This
           object has all the information about the profile in it, and you can
           later use it for further analysis or in other functions.
 
         -s <key>
           sort profile by given key. You can provide more than one key
           by using the option several times: '-s key1 -s key2 -s key3...'. The
           default sorting key is 'time'.
 
           The following is copied verbatim from the profile documentation
           referenced below:
 
           When more than one key is provided, additional keys are used as
           secondary criteria when the there is equality in all keys selected
           before them.
 
           Abbreviations can be used for any key names, as long as the
           abbreviation is unambiguous.  The following are the keys currently
           defined:
 
           ============  =====================
           Valid Arg     Meaning
           ============  =====================
           "calls"       call count
           "cumulative"  cumulative time
           "file"        file name
           "module"      file name
           "pcalls"      primitive call count
           "line"        line number
           "name"        function name
           "nfl"         name/file/line
           "stdname"     standard name
           "time"        internal time
           ============  =====================
 
           Note that all sorts on statistics are in descending order (placing
           most time consuming items first), where as name, file, and line number
           searches are in ascending order (i.e., alphabetical). The subtle
           distinction between "nfl" and "stdname" is that the standard name is a
           sort of the name as printed, which means that the embedded line
           numbers get compared in an odd way.  For example, lines 3, 20, and 40
           would (if the file names were the same) appear in the string order
           "20" "3" and "40".  In contrast, "nfl" does a numeric compare of the
           line numbers.  In fact, sort_stats("nfl") is the same as
           sort_stats("name", "file", "line").
 
         -T <filename>
           save profile results as shown on screen to a text
           file. The profile is still shown on screen.
 
         -D <filename>
           save (via dump_stats) profile statistics to given
           filename. This data is in a format understood by the pstats module, and
           is generated by a call to the dump_stats() method of profile
           objects. The profile is still shown on screen.
 
         -q
           suppress output to the pager.  Best used with -T and/or -D above.
 
         If you want to run complete programs under the profiler's control, use
         ``%run -p [prof_opts] filename.py [args to program]`` where prof_opts
         contains profiler specific options as described here.
 
         You can read the complete documentation for the profile module with::
 
           In [1]: import profile; profile.help()
 
         .. versionchanged:: 7.3
             User variables are no longer expanded,
             the magic line is always left unmodified.
 
         """
         opts, arg_str = self.parse_options(parameter_s, 'D:l:rs:T:q',
                                            list_all=True, posix=False)
         if cell is not None:
             arg_str += '\n' + cell
         arg_str = self.shell.transform_cell(arg_str)
         return self._run_with_profiler(arg_str, opts, self.shell.user_ns)
 
     def _run_with_profiler(self, code, opts, namespace):
         """
         Run `code` with profiler.  Used by ``%prun`` and ``%run -p``.
 
         Parameters
         ----------
         code : str
             Code to be executed.
         opts : Struct
             Options parsed by `self.parse_options`.
         namespace : dict
             A dictionary for Python namespace (e.g., `self.shell.user_ns`).
 
         """
 
         # Fill default values for unspecified options:
         opts.merge(Struct(D=[''], l=[], s=['time'], T=['']))
 
         prof = profile.Profile()
         try:
             prof = prof.runctx(code, namespace, namespace)
             sys_exit = ''
         except SystemExit:
             sys_exit = """*** SystemExit exception caught in code being profiled."""
 
         stats = pstats.Stats(prof).strip_dirs().sort_stats(*opts.s)
 
         lims = opts.l
         if lims:
             lims = []  # rebuild lims with ints/floats/strings
             for lim in opts.l:
                 try:
                     lims.append(int(lim))
                 except ValueError:
                     try:
                         lims.append(float(lim))
                     except ValueError:
                         lims.append(lim)
 
         # Trap output.
         stdout_trap = StringIO()
         stats_stream = stats.stream
         try:
             stats.stream = stdout_trap
             stats.print_stats(*lims)
         finally:
             stats.stream = stats_stream
 
         output = stdout_trap.getvalue()
         output = output.rstrip()
 
         if 'q' not in opts:
             page.page(output)
         print(sys_exit, end=' ')
 
         dump_file = opts.D[0]
         text_file = opts.T[0]
         if dump_file:
             prof.dump_stats(dump_file)
             print(
                 f"\n*** Profile stats marshalled to file {repr(dump_file)}.{sys_exit}"
             )
         if text_file:
             pfile = Path(text_file)
             pfile.touch(exist_ok=True)
             pfile.write_text(output, encoding="utf-8")
 
             print(
                 f"\n*** Profile printout saved to text file {repr(text_file)}.{sys_exit}"
             )
 
         if 'r' in opts:
             return stats
 
         return None
 
     @line_magic
     def pdb(self, parameter_s=''):
         """Control the automatic calling of the pdb interactive debugger.
 
         Call as '%pdb on', '%pdb 1', '%pdb off' or '%pdb 0'. If called without
         argument it works as a toggle.
 
         When an exception is triggered, IPython can optionally call the
         interactive pdb debugger after the traceback printout. %pdb toggles
         this feature on and off.
 
         The initial state of this feature is set in your configuration
         file (the option is ``InteractiveShell.pdb``).
 
         If you want to just activate the debugger AFTER an exception has fired,
         without having to type '%pdb on' and rerunning your code, you can use
         the %debug magic."""
 
         par = parameter_s.strip().lower()
 
         if par:
             try:
                 new_pdb = {'off':0,'0':0,'on':1,'1':1}[par]
             except KeyError:
                 print ('Incorrect argument. Use on/1, off/0, '
                        'or nothing for a toggle.')
                 return
         else:
             # toggle
             new_pdb = not self.shell.call_pdb
 
         # set on the shell
         self.shell.call_pdb = new_pdb
         print('Automatic pdb calling has been turned',on_off(new_pdb))
 
     @magic_arguments.magic_arguments()
     @magic_arguments.argument('--breakpoint', '-b', metavar='FILE:LINE',
         help="""
         Set break point at LINE in FILE.
         """
     )
     @magic_arguments.argument('statement', nargs='*',
         help="""
         Code to run in debugger.
         You can omit this in cell magic mode.
         """
     )
     @no_var_expand
     @line_cell_magic
     @needs_local_scope
     def debug(self, line="", cell=None, local_ns=None):
         """Activate the interactive debugger.
 
         This magic command support two ways of activating debugger.
         One is to activate debugger before executing code.  This way, you
         can set a break point, to step through the code from the point.
         You can use this mode by giving statements to execute and optionally
         a breakpoint.
 
         The other one is to activate debugger in post-mortem mode.  You can
         activate this mode simply running %debug without any argument.
         If an exception has just occurred, this lets you inspect its stack
         frames interactively.  Note that this will always work only on the last
         traceback that occurred, so you must call this quickly after an
         exception that you wish to inspect has fired, because if another one
         occurs, it clobbers the previous one.
 
         If you want IPython to automatically do this on every exception, see
         the %pdb magic for more details.
 
         .. versionchanged:: 7.3
             When running code, user variables are no longer expanded,
             the magic line is always left unmodified.
 
         """
         args = magic_arguments.parse_argstring(self.debug, line)
 
         if not (args.breakpoint or args.statement or cell):
             self._debug_post_mortem()
         elif not (args.breakpoint or cell):
             # If there is no breakpoints, the line is just code to execute
             self._debug_exec(line, None, local_ns)
         else:
             # Here we try to reconstruct the code from the output of
             # parse_argstring. This might not work if the code has spaces
             # For example this fails for `print("a b")`
             code = "\n".join(args.statement)
             if cell:
                 code += "\n" + cell
             self._debug_exec(code, args.breakpoint, local_ns)
 
     def _debug_post_mortem(self):
         self.shell.debugger(force=True)
 
     def _debug_exec(self, code, breakpoint, local_ns=None):
         if breakpoint:
             (filename, bp_line) = breakpoint.rsplit(':', 1)
             bp_line = int(bp_line)
         else:
             (filename, bp_line) = (None, None)
         self._run_with_debugger(
             code, self.shell.user_ns, filename, bp_line, local_ns=local_ns
         )
 
     @line_magic
     def tb(self, s):
         """Print the last traceback.
 
         Optionally, specify an exception reporting mode, tuning the
         verbosity of the traceback. By default the currently-active exception
         mode is used. See %xmode for changing exception reporting modes.
 
         Valid modes: Plain, Context, Verbose, and Minimal.
         """
         interactive_tb = self.shell.InteractiveTB
         if s:
             # Switch exception reporting mode for this one call.
             # Ensure it is switched back.
             def xmode_switch_err(name):
                 warn('Error changing %s exception modes.\n%s' %
                     (name,sys.exc_info()[1]))
 
             new_mode = s.strip().capitalize()
             original_mode = interactive_tb.mode
             try:
                 try:
                     interactive_tb.set_mode(mode=new_mode)
                 except Exception:
                     xmode_switch_err('user')
                 else:
                     self.shell.showtraceback()
             finally:
                 interactive_tb.set_mode(mode=original_mode)
         else:
             self.shell.showtraceback()
 
     @skip_doctest
     @line_magic
     def run(self, parameter_s='', runner=None,
                   file_finder=get_py_filename):
         """Run the named file inside IPython as a program.
 
         Usage::
 
           %run [-n -i -e -G]
                [( -t [-N<N>] | -d [-b<N>] | -p [profile options] )]
                ( -m mod | filename ) [args]
 
         The filename argument should be either a pure Python script (with
         extension ``.py``), or a file with custom IPython syntax (such as
         magics). If the latter, the file can be either a script with ``.ipy``
         extension, or a Jupyter notebook with ``.ipynb`` extension. When running
         a Jupyter notebook, the output from print statements and other
         displayed objects will appear in the terminal (even matplotlib figures
         will open, if a terminal-compliant backend is being used). Note that,
         at the system command line, the ``jupyter run`` command offers similar
         functionality for executing notebooks (albeit currently with some
         differences in supported options).
 
         Parameters after the filename are passed as command-line arguments to
         the program (put in sys.argv). Then, control returns to IPython's
         prompt.
 
         This is similar to running at a system prompt ``python file args``,
         but with the advantage of giving you IPython's tracebacks, and of
         loading all variables into your interactive namespace for further use
         (unless -p is used, see below).
 
         The file is executed in a namespace initially consisting only of
         ``__name__=='__main__'`` and sys.argv constructed as indicated. It thus
         sees its environment as if it were being run as a stand-alone program
         (except for sharing global objects such as previously imported
         modules). But after execution, the IPython interactive namespace gets
         updated with all variables defined in the program (except for __name__
         and sys.argv). This allows for very convenient loading of code for
         interactive work, while giving each program a 'clean sheet' to run in.
 
         Arguments are expanded using shell-like glob match.  Patterns
         '*', '?', '[seq]' and '[!seq]' can be used.  Additionally,
         tilde '~' will be expanded into user's home directory.  Unlike
         real shells, quotation does not suppress expansions.  Use
         *two* back slashes (e.g. ``\\\\*``) to suppress expansions.
         To completely disable these expansions, you can use -G flag.
 
         On Windows systems, the use of single quotes `'` when specifying
         a file is not supported. Use double quotes `"`.
 
         Options:
 
         -n
           __name__ is NOT set to '__main__', but to the running file's name
           without extension (as python does under import).  This allows running
           scripts and reloading the definitions in them without calling code
           protected by an ``if __name__ == "__main__"`` clause.
 
         -i
           run the file in IPython's namespace instead of an empty one. This
           is useful if you are experimenting with code written in a text editor
           which depends on variables defined interactively.
 
         -e
           ignore sys.exit() calls or SystemExit exceptions in the script
           being run.  This is particularly useful if IPython is being used to
           run unittests, which always exit with a sys.exit() call.  In such
           cases you are interested in the output of the test results, not in
           seeing a traceback of the unittest module.
 
         -t
           print timing information at the end of the run.  IPython will give
           you an estimated CPU time consumption for your script, which under
           Unix uses the resource module to avoid the wraparound problems of
           time.clock().  Under Unix, an estimate of time spent on system tasks
           is also given (for Windows platforms this is reported as 0.0).
 
         If -t is given, an additional ``-N<N>`` option can be given, where <N>
         must be an integer indicating how many times you want the script to
         run.  The final timing report will include total and per run results.
 
         For example (testing the script uniq_stable.py)::
 
             In [1]: run -t uniq_stable
 
             IPython CPU timings (estimated):
               User  :    0.19597 s.
               System:        0.0 s.
 
             In [2]: run -t -N5 uniq_stable
 
             IPython CPU timings (estimated):
             Total runs performed: 5
               Times :      Total       Per run
               User  :   0.910862 s,  0.1821724 s.
               System:        0.0 s,        0.0 s.
 
         -d
           run your program under the control of pdb, the Python debugger.
           This allows you to execute your program step by step, watch variables,
           etc.  Internally, what IPython does is similar to calling::
 
               pdb.run('execfile("YOURFILENAME")')
 
           with a breakpoint set on line 1 of your file.  You can change the line
           number for this automatic breakpoint to be <N> by using the -bN option
           (where N must be an integer). For example::
 
               %run -d -b40 myscript
 
           will set the first breakpoint at line 40 in myscript.py.  Note that
           the first breakpoint must be set on a line which actually does
           something (not a comment or docstring) for it to stop execution.
 
           Or you can specify a breakpoint in a different file::
 
               %run -d -b myotherfile.py:20 myscript
 
           When the pdb debugger starts, you will see a (Pdb) prompt.  You must
           first enter 'c' (without quotes) to start execution up to the first
           breakpoint.
 
           Entering 'help' gives information about the use of the debugger.  You
           can easily see pdb's full documentation with "import pdb;pdb.help()"
           at a prompt.
 
         -p
           run program under the control of the Python profiler module (which
           prints a detailed report of execution times, function calls, etc).
 
           You can pass other options after -p which affect the behavior of the
           profiler itself. See the docs for %prun for details.
 
           In this mode, the program's variables do NOT propagate back to the
           IPython interactive namespace (because they remain in the namespace
           where the profiler executes them).
 
           Internally this triggers a call to %prun, see its documentation for
           details on the options available specifically for profiling.
 
         There is one special usage for which the text above doesn't apply:
         if the filename ends with .ipy[nb], the file is run as ipython script,
         just as if the commands were written on IPython prompt.
 
         -m
           specify module name to load instead of script path. Similar to
           the -m option for the python interpreter. Use this option last if you
           want to combine with other %run options. Unlike the python interpreter
           only source modules are allowed no .pyc or .pyo files.
           For example::
 
               %run -m example
 
           will run the example module.
 
         -G
           disable shell-like glob expansion of arguments.
 
         """
 
         # Logic to handle issue #3664
         # Add '--' after '-m <module_name>' to ignore additional args passed to a module.
         if '-m' in parameter_s and '--' not in parameter_s:
             argv = shlex.split(parameter_s, posix=(os.name == 'posix'))
             for idx, arg in enumerate(argv):
                 if arg and arg.startswith('-') and arg != '-':
                     if arg == '-m':
                         argv.insert(idx + 2, '--')
                         break
                 else:
                     # Positional arg, break
                     break
             parameter_s = ' '.join(shlex.quote(arg) for arg in argv)
 
         # get arguments and set sys.argv for program to be run.
         opts, arg_lst = self.parse_options(parameter_s,
                                            'nidtN:b:pD:l:rs:T:em:G',
                                            mode='list', list_all=1)
         if "m" in opts:
             modulename = opts["m"][0]
             modpath = find_mod(modulename)
             if modpath is None:
                 msg = '%r is not a valid modulename on sys.path'%modulename
                 raise Exception(msg)
             arg_lst = [modpath] + arg_lst
         try:
             fpath = None # initialize to make sure fpath is in scope later
             fpath = arg_lst[0]
             filename = file_finder(fpath)
         except IndexError as e:
             msg = 'you must provide at least a filename.'
             raise Exception(msg) from e
         except IOError as e:
             try:
                 msg = str(e)
             except UnicodeError:
                 msg = e.message
             if os.name == 'nt' and re.match(r"^'.*'$",fpath):
                 warn('For Windows, use double quotes to wrap a filename: %run "mypath\\myfile.py"')
             raise Exception(msg) from e
         except TypeError:
             if fpath in sys.meta_path:
                 filename = ""
             else:
                 raise
 
         if filename.lower().endswith(('.ipy', '.ipynb')):
             with preserve_keys(self.shell.user_ns, '__file__'):
                 self.shell.user_ns['__file__'] = filename
                 self.shell.safe_execfile_ipy(filename, raise_exceptions=True)
             return
 
         # Control the response to exit() calls made by the script being run
         exit_ignore = 'e' in opts
 
         # Make sure that the running script gets a proper sys.argv as if it
         # were run from a system shell.
         save_argv = sys.argv # save it for later restoring
 
         if 'G' in opts:
             args = arg_lst[1:]
         else:
             # tilde and glob expansion
             args = shellglob(map(os.path.expanduser,  arg_lst[1:]))
 
         sys.argv = [filename] + args  # put in the proper filename
 
         if 'n' in opts:
             name = Path(filename).stem
         else:
             name = '__main__'
 
         if 'i' in opts:
             # Run in user's interactive namespace
             prog_ns = self.shell.user_ns
             __name__save = self.shell.user_ns['__name__']
             prog_ns['__name__'] = name
             main_mod = self.shell.user_module
 
             # Since '%run foo' emulates 'python foo.py' at the cmd line, we must
             # set the __file__ global in the script's namespace
             # TK: Is this necessary in interactive mode?
             prog_ns['__file__'] = filename
         else:
             # Run in a fresh, empty namespace
 
             # The shell MUST hold a reference to prog_ns so after %run
             # exits, the python deletion mechanism doesn't zero it out
             # (leaving dangling references). See interactiveshell for details
             main_mod = self.shell.new_main_mod(filename, name)
             prog_ns = main_mod.__dict__
 
         # pickle fix.  See interactiveshell for an explanation.  But we need to
         # make sure that, if we overwrite __main__, we replace it at the end
         main_mod_name = prog_ns['__name__']
 
         if main_mod_name == '__main__':
             restore_main = sys.modules['__main__']
         else:
             restore_main = False
 
         # This needs to be undone at the end to prevent holding references to
         # every single object ever created.
         sys.modules[main_mod_name] = main_mod
 
         if 'p' in opts or 'd' in opts:
             if 'm' in opts:
                 code = 'run_module(modulename, prog_ns)'
                 code_ns = {
                     'run_module': self.shell.safe_run_module,
                     'prog_ns': prog_ns,
                     'modulename': modulename,
                 }
             else:
                 if 'd' in opts:
                     # allow exceptions to raise in debug mode
                     code = 'execfile(filename, prog_ns, raise_exceptions=True)'
                 else:
                     code = 'execfile(filename, prog_ns)'
                 code_ns = {
                     'execfile': self.shell.safe_execfile,
                     'prog_ns': prog_ns,
                     'filename': get_py_filename(filename),
                 }
 
         try:
             stats = None
             if 'p' in opts:
                 stats = self._run_with_profiler(code, opts, code_ns)
             else:
                 if 'd' in opts:
                     bp_file, bp_line = parse_breakpoint(
                         opts.get('b', ['1'])[0], filename)
                     self._run_with_debugger(
                         code, code_ns, filename, bp_line, bp_file)
                 else:
                     if 'm' in opts:
                         def run():
                             self.shell.safe_run_module(modulename, prog_ns)
                     else:
                         if runner is None:
                             runner = self.default_runner
                         if runner is None:
                             runner = self.shell.safe_execfile
 
                         def run():
                             runner(filename, prog_ns, prog_ns,
                                     exit_ignore=exit_ignore)
 
                     if 't' in opts:
                         # timed execution
                         try:
                             nruns = int(opts['N'][0])
                             if nruns < 1:
                                 error('Number of runs must be >=1')
                                 return
                         except (KeyError):
                             nruns = 1
                         self._run_with_timing(run, nruns)
                     else:
                         # regular execution
                         run()
 
             if 'i' in opts:
                 self.shell.user_ns['__name__'] = __name__save
             else:
                 # update IPython interactive namespace
 
                 # Some forms of read errors on the file may mean the
                 # __name__ key was never set; using pop we don't have to
                 # worry about a possible KeyError.
                 prog_ns.pop('__name__', None)
 
                 with preserve_keys(self.shell.user_ns, '__file__'):
                     self.shell.user_ns.update(prog_ns)
         finally:
             # It's a bit of a mystery why, but __builtins__ can change from
             # being a module to becoming a dict missing some key data after
             # %run.  As best I can see, this is NOT something IPython is doing
             # at all, and similar problems have been reported before:
             # http://coding.derkeiler.com/Archive/Python/comp.lang.python/2004-10/0188.html
             # Since this seems to be done by the interpreter itself, the best
             # we can do is to at least restore __builtins__ for the user on
             # exit.
             self.shell.user_ns['__builtins__'] = builtin_mod
 
             # Ensure key global structures are restored
             sys.argv = save_argv
             if restore_main:
                 sys.modules['__main__'] = restore_main
                 if '__mp_main__' in sys.modules:
                     sys.modules['__mp_main__'] = restore_main
             else:
                 # Remove from sys.modules the reference to main_mod we'd
                 # added.  Otherwise it will trap references to objects
                 # contained therein.
                 del sys.modules[main_mod_name]
 
         return stats
 
     def _run_with_debugger(
         self, code, code_ns, filename=None, bp_line=None, bp_file=None, local_ns=None
     ):
         """
         Run `code` in debugger with a break point.
 
         Parameters
         ----------
         code : str
             Code to execute.
         code_ns : dict
             A namespace in which `code` is executed.
         filename : str
             `code` is ran as if it is in `filename`.
         bp_line : int, optional
             Line number of the break point.
         bp_file : str, optional
             Path to the file in which break point is specified.
             `filename` is used if not given.
         local_ns : dict, optional
             A local namespace in which `code` is executed.
 
         Raises
         ------
         UsageError
             If the break point given by `bp_line` is not valid.
 
         """
         deb = self.shell.InteractiveTB.pdb
         if not deb:
             self.shell.InteractiveTB.pdb = self.shell.InteractiveTB.debugger_cls()
             deb = self.shell.InteractiveTB.pdb
 
         # deb.checkline() fails if deb.curframe exists but is None; it can
         # handle it not existing. https://github.com/ipython/ipython/issues/10028
         if hasattr(deb, 'curframe'):
             del deb.curframe
 
         # reset Breakpoint state, which is moronically kept
         # in a class
         bdb.Breakpoint.next = 1
         bdb.Breakpoint.bplist = {}
         bdb.Breakpoint.bpbynumber = [None]
         deb.clear_all_breaks()
         if bp_line is not None:
             # Set an initial breakpoint to stop execution
             maxtries = 10
             bp_file = bp_file or filename
             checkline = deb.checkline(bp_file, bp_line)
             if not checkline:
                 for bp in range(bp_line + 1, bp_line + maxtries + 1):
                     if deb.checkline(bp_file, bp):
                         break
                 else:
                     msg = ("\nI failed to find a valid line to set "
                            "a breakpoint\n"
                            "after trying up to line: %s.\n"
                            "Please set a valid breakpoint manually "
                            "with the -b option." % bp)
                     raise UsageError(msg)
             # if we find a good linenumber, set the breakpoint
             deb.do_break('%s:%s' % (bp_file, bp_line))
 
         if filename:
             # Mimic Pdb._runscript(...)
             deb._wait_for_mainpyfile = True
             deb.mainpyfile = deb.canonic(filename)
 
         # Start file run
         print("NOTE: Enter 'c' at the %s prompt to continue execution." % deb.prompt)
         try:
             if filename:
                 # save filename so it can be used by methods on the deb object
                 deb._exec_filename = filename
             while True:
                 try:
                     trace = sys.gettrace()
                     deb.run(code, code_ns, local_ns)
                 except Restart:
                     print("Restarting")
                     if filename:
                         deb._wait_for_mainpyfile = True
                         deb.mainpyfile = deb.canonic(filename)
                     continue
                 else:
                     break
                 finally:
                     sys.settrace(trace)
             
 
         except:
             etype, value, tb = sys.exc_info()
             # Skip three frames in the traceback: the %run one,
             # one inside bdb.py, and the command-line typed by the
             # user (run by exec in pdb itself).
             self.shell.InteractiveTB(etype, value, tb, tb_offset=3)
 
     @staticmethod
     def _run_with_timing(run, nruns):
         """
         Run function `run` and print timing information.
 
         Parameters
         ----------
         run : callable
             Any callable object which takes no argument.
         nruns : int
             Number of times to execute `run`.
 
         """
         twall0 = time.perf_counter()
         if nruns == 1:
             t0 = clock2()
             run()
             t1 = clock2()
             t_usr = t1[0] - t0[0]
             t_sys = t1[1] - t0[1]
             print("\nIPython CPU timings (estimated):")
             print("  User   : %10.2f s." % t_usr)
             print("  System : %10.2f s." % t_sys)
         else:
             runs = range(nruns)
             t0 = clock2()
             for nr in runs:
                 run()
             t1 = clock2()
             t_usr = t1[0] - t0[0]
             t_sys = t1[1] - t0[1]
             print("\nIPython CPU timings (estimated):")
             print("Total runs performed:", nruns)
             print("  Times  : %10s   %10s" % ('Total', 'Per run'))
             print("  User   : %10.2f s, %10.2f s." % (t_usr, t_usr / nruns))
             print("  System : %10.2f s, %10.2f s." % (t_sys, t_sys / nruns))
         twall1 = time.perf_counter()
         print("Wall time: %10.2f s." % (twall1 - twall0))
 
     @skip_doctest
     @no_var_expand
     @line_cell_magic
     @needs_local_scope
     def timeit(self, line='', cell=None, local_ns=None):
         """Time execution of a Python statement or expression
 
         Usage, in line mode:
           %timeit [-n<N> -r<R> [-t|-c] -q -p<P> -o] statement
         or in cell mode:
           %%timeit [-n<N> -r<R> [-t|-c] -q -p<P> -o] setup_code
           code
           code...
 
         Time execution of a Python statement or expression using the timeit
         module.  This function can be used both as a line and cell magic:
 
         - In line mode you can time a single-line statement (though multiple
           ones can be chained with using semicolons).
 
         - In cell mode, the statement in the first line is used as setup code
           (executed but not timed) and the body of the cell is timed.  The cell
           body has access to any variables created in the setup code.
 
         Options:
         -n<N>: execute the given statement <N> times in a loop. If <N> is not
         provided, <N> is determined so as to get sufficient accuracy.
 
         -r<R>: number of repeats <R>, each consisting of <N> loops, and take the
         best result.
         Default: 7
 
         -t: use time.time to measure the time, which is the default on Unix.
         This function measures wall time.
 
         -c: use time.clock to measure the time, which is the default on
         Windows and measures wall time. On Unix, resource.getrusage is used
         instead and returns the CPU user time.
 
         -p<P>: use a precision of <P> digits to display the timing result.
         Default: 3
 
         -q: Quiet, do not print result.
 
         -o: return a TimeitResult that can be stored in a variable to inspect
             the result in more details.
 
         .. versionchanged:: 7.3
             User variables are no longer expanded,
             the magic line is always left unmodified.
 
         Examples
         --------
         ::
 
           In [1]: %timeit pass
           8.26 ns ± 0.12 ns per loop (mean ± std. dev. of 7 runs, 100000000 loops each)
 
           In [2]: u = None
 
           In [3]: %timeit u is None
           29.9 ns ± 0.643 ns per loop (mean ± std. dev. of 7 runs, 10000000 loops each)
 
           In [4]: %timeit -r 4 u == None
 
           In [5]: import time
 
           In [6]: %timeit -n1 time.sleep(2)
 
         The times reported by %timeit will be slightly higher than those
         reported by the timeit.py script when variables are accessed. This is
         due to the fact that %timeit executes the statement in the namespace
         of the shell, compared with timeit.py, which uses a single setup
         statement to import function or create variables. Generally, the bias
         does not matter as long as results from timeit.py are not mixed with
         those from %timeit."""
 
         opts, stmt = self.parse_options(
             line, "n:r:tcp:qo", posix=False, strict=False, preserve_non_opts=True
         )
         if stmt == "" and cell is None:
             return
         
         timefunc = timeit.default_timer
         number = int(getattr(opts, "n", 0))
         default_repeat = 7 if timeit.default_repeat < 7 else timeit.default_repeat
         repeat = int(getattr(opts, "r", default_repeat))
         precision = int(getattr(opts, "p", 3))
         quiet = 'q' in opts
         return_result = 'o' in opts
         if hasattr(opts, "t"):
             timefunc = time.time
         if hasattr(opts, "c"):
             timefunc = clock
 
         timer = Timer(timer=timefunc)
         # this code has tight coupling to the inner workings of timeit.Timer,
         # but is there a better way to achieve that the code stmt has access
         # to the shell namespace?
         transform  = self.shell.transform_cell
 
         if cell is None:
             # called as line magic
             ast_setup = self.shell.compile.ast_parse("pass")
             ast_stmt = self.shell.compile.ast_parse(transform(stmt))
         else:
             ast_setup = self.shell.compile.ast_parse(transform(stmt))
             ast_stmt = self.shell.compile.ast_parse(transform(cell))
 
         ast_setup = self.shell.transform_ast(ast_setup)
         ast_stmt = self.shell.transform_ast(ast_stmt)
 
         # Check that these compile to valid Python code *outside* the timer func
         # Invalid code may become valid when put inside the function & loop,
         # which messes up error messages.
         # https://github.com/ipython/ipython/issues/10636
         self.shell.compile(ast_setup, "<magic-timeit-setup>", "exec")
         self.shell.compile(ast_stmt, "<magic-timeit-stmt>", "exec")
 
         # This codestring is taken from timeit.template - we fill it in as an
         # AST, so that we can apply our AST transformations to the user code
         # without affecting the timing code.
         timeit_ast_template = ast.parse('def inner(_it, _timer):\n'
                                         '    setup\n'
                                         '    _t0 = _timer()\n'
                                         '    for _i in _it:\n'
                                         '        stmt\n'
                                         '    _t1 = _timer()\n'
                                         '    return _t1 - _t0\n')
 
         timeit_ast = TimeitTemplateFiller(ast_setup, ast_stmt).visit(timeit_ast_template)
         timeit_ast = ast.fix_missing_locations(timeit_ast)
 
         # Track compilation time so it can be reported if too long
         # Minimum time above which compilation time will be reported
         tc_min = 0.1
 
         t0 = clock()
         code = self.shell.compile(timeit_ast, "<magic-timeit>", "exec")
         tc = clock()-t0
 
         ns = {}
         glob = self.shell.user_ns
         # handles global vars with same name as local vars. We store them in conflict_globs.
         conflict_globs = {}
         if local_ns and cell is None:
             for var_name, var_val in glob.items():
                 if var_name in local_ns:
                     conflict_globs[var_name] = var_val
             glob.update(local_ns)
             
         exec(code, glob, ns)
         timer.inner = ns["inner"]
 
         # This is used to check if there is a huge difference between the
         # best and worst timings.
         # Issue: https://github.com/ipython/ipython/issues/6471
         if number == 0:
             # determine number so that 0.2 <= total time < 2.0
             for index in range(0, 10):
                 number = 10 ** index
                 time_number = timer.timeit(number)
                 if time_number >= 0.2:
                     break
 
         all_runs = timer.repeat(repeat, number)
         best = min(all_runs) / number
         worst = max(all_runs) / number
         timeit_result = TimeitResult(number, repeat, best, worst, all_runs, tc, precision)
 
         # Restore global vars from conflict_globs
         if conflict_globs:
            glob.update(conflict_globs)
                 
         if not quiet :
             # Check best timing is greater than zero to avoid a
             # ZeroDivisionError.
             # In cases where the slowest timing is lesser than a microsecond
             # we assume that it does not really matter if the fastest
             # timing is 4 times faster than the slowest timing or not.
             if worst > 4 * best and best > 0 and worst > 1e-6:
                 print("The slowest run took %0.2f times longer than the "
                       "fastest. This could mean that an intermediate result "
                       "is being cached." % (worst / best))
            
             print( timeit_result )
 
             if tc > tc_min:
                 print("Compiler time: %.2f s" % tc)
         if return_result:
             return timeit_result
 
     @skip_doctest
     @no_var_expand
     @needs_local_scope
     @line_cell_magic
     @output_can_be_silenced
     def time(self,line='', cell=None, local_ns=None):
         """Time execution of a Python statement or expression.
 
         The CPU and wall clock times are printed, and the value of the
         expression (if any) is returned.  Note that under Win32, system time
         is always reported as 0, since it can not be measured.
 
         This function can be used both as a line and cell magic:
 
         - In line mode you can time a single-line statement (though multiple
           ones can be chained with using semicolons).
 
         - In cell mode, you can time the cell body (a directly
           following statement raises an error).
 
         This function provides very basic timing functionality.  Use the timeit
         magic for more control over the measurement.
 
         .. versionchanged:: 7.3
             User variables are no longer expanded,
             the magic line is always left unmodified.
 
         Examples
         --------
         ::
 
           In [1]: %time 2**128
           CPU times: user 0.00 s, sys: 0.00 s, total: 0.00 s
           Wall time: 0.00
           Out[1]: 340282366920938463463374607431768211456L
 
           In [2]: n = 1000000
 
           In [3]: %time sum(range(n))
           CPU times: user 1.20 s, sys: 0.05 s, total: 1.25 s
           Wall time: 1.37
           Out[3]: 499999500000L
 
           In [4]: %time print 'hello world'
           hello world
           CPU times: user 0.00 s, sys: 0.00 s, total: 0.00 s
           Wall time: 0.00
 
         .. note::
             The time needed by Python to compile the given expression will be
             reported if it is more than 0.1s.
 
             In the example below, the actual exponentiation is done by Python
             at compilation time, so while the expression can take a noticeable
             amount of time to compute, that time is purely due to the
             compilation::
 
                 In [5]: %time 3**9999;
                 CPU times: user 0.00 s, sys: 0.00 s, total: 0.00 s
                 Wall time: 0.00 s
 
                 In [6]: %time 3**999999;
                 CPU times: user 0.00 s, sys: 0.00 s, total: 0.00 s
                 Wall time: 0.00 s
                 Compiler : 0.78 s
         """
         # fail immediately if the given expression can't be compiled
         
         if line and cell:
             raise UsageError("Can't use statement directly after '%%time'!")
         
         if cell:
             expr = self.shell.transform_cell(cell)
         else:
             expr = self.shell.transform_cell(line)
 
         # Minimum time above which parse time will be reported
         tp_min = 0.1
 
         t0 = clock()
         expr_ast = self.shell.compile.ast_parse(expr)
         tp = clock()-t0
 
         # Apply AST transformations
         expr_ast = self.shell.transform_ast(expr_ast)
 
         # Minimum time above which compilation time will be reported
         tc_min = 0.1
 
         expr_val=None
         if len(expr_ast.body)==1 and isinstance(expr_ast.body[0], ast.Expr):
             mode = 'eval'
             source = '<timed eval>'
             expr_ast = ast.Expression(expr_ast.body[0].value)
         else:
             mode = 'exec'
             source = '<timed exec>'
             # multi-line %%time case
             if len(expr_ast.body) > 1 and isinstance(expr_ast.body[-1], ast.Expr):
                 expr_val= expr_ast.body[-1]
                 expr_ast = expr_ast.body[:-1]
                 expr_ast = Module(expr_ast, [])
                 expr_val = ast.Expression(expr_val.value)
 
         t0 = clock()
         code = self.shell.compile(expr_ast, source, mode)
         tc = clock()-t0
 
         # skew measurement as little as possible
         glob = self.shell.user_ns
         wtime = time.time
         # time execution
         wall_st = wtime()
         if mode=='eval':
             st = clock2()
             try:
                 out = eval(code, glob, local_ns)
             except:
                 self.shell.showtraceback()
                 return
             end = clock2()
         else:
             st = clock2()
             try:
                 exec(code, glob, local_ns)
                 out=None
                 # multi-line %%time case
                 if expr_val is not None:
                     code_2 = self.shell.compile(expr_val, source, 'eval')
                     out = eval(code_2, glob, local_ns)
             except:
                 self.shell.showtraceback()
                 return
             end = clock2()
 
         wall_end = wtime()
         # Compute actual times and report
         wall_time = wall_end - wall_st
         cpu_user = end[0] - st[0]
         cpu_sys = end[1] - st[1]
         cpu_tot = cpu_user + cpu_sys
         # On windows cpu_sys is always zero, so only total is displayed
         if sys.platform != "win32":
             print(
                 f"CPU times: user {_format_time(cpu_user)}, sys: {_format_time(cpu_sys)}, total: {_format_time(cpu_tot)}"
             )
         else:
             print(f"CPU times: total: {_format_time(cpu_tot)}")
         print(f"Wall time: {_format_time(wall_time)}")
         if tc > tc_min:
             print(f"Compiler : {_format_time(tc)}")
         if tp > tp_min:
             print(f"Parser   : {_format_time(tp)}")
         return out
 
     @skip_doctest
     @line_magic
     def macro(self, parameter_s=''):
         """Define a macro for future re-execution. It accepts ranges of history,
         filenames or string objects.
 
         Usage:\\
           %macro [options] name n1-n2 n3-n4 ... n5 .. n6 ...
 
         Options:
 
           -r: use 'raw' input.  By default, the 'processed' history is used,
           so that magics are loaded in their transformed version to valid
           Python.  If this option is given, the raw input as typed at the
           command line is used instead.
           
           -q: quiet macro definition.  By default, a tag line is printed 
           to indicate the macro has been created, and then the contents of 
           the macro are printed.  If this option is given, then no printout
           is produced once the macro is created.
 
         This will define a global variable called `name` which is a string
         made of joining the slices and lines you specify (n1,n2,... numbers
         above) from your input history into a single string. This variable
         acts like an automatic function which re-executes those lines as if
         you had typed them. You just type 'name' at the prompt and the code
         executes.
 
         The syntax for indicating input ranges is described in %history.
 
         Note: as a 'hidden' feature, you can also use traditional python slice
         notation, where N:M means numbers N through M-1.
 
         For example, if your history contains (print using %hist -n )::
 
           44: x=1
           45: y=3
           46: z=x+y
           47: print x
           48: a=5
           49: print 'x',x,'y',y
 
         you can create a macro with lines 44 through 47 (included) and line 49
         called my_macro with::
 
           In [55]: %macro my_macro 44-47 49
 
         Now, typing `my_macro` (without quotes) will re-execute all this code
         in one pass.
 
         You don't need to give the line-numbers in order, and any given line
         number can appear multiple times. You can assemble macros with any
         lines from your input history in any order.
 
         The macro is a simple object which holds its value in an attribute,
         but IPython's display system checks for macros and executes them as
         code instead of printing them when you type their name.
 
         You can view a macro's contents by explicitly printing it with::
 
           print macro_name
 
         """
         opts,args = self.parse_options(parameter_s,'rq',mode='list')
         if not args:   # List existing macros
             return sorted(k for k,v in self.shell.user_ns.items() if isinstance(v, Macro))
         if len(args) == 1:
             raise UsageError(
                 "%macro insufficient args; usage '%macro name n1-n2 n3-4...")
         name, codefrom = args[0], " ".join(args[1:])
 
         #print 'rng',ranges  # dbg
         try:
             lines = self.shell.find_user_code(codefrom, 'r' in opts)
         except (ValueError, TypeError) as e:
             print(e.args[0])
             return
         macro = Macro(lines)
         self.shell.define_macro(name, macro)
         if not ( 'q' in opts) : 
             print('Macro `%s` created. To execute, type its name (without quotes).' % name)
             print('=== Macro contents: ===')
             print(macro, end=' ')
 
     @magic_arguments.magic_arguments()
     @magic_arguments.argument('output', type=str, default='', nargs='?',
         help="""The name of the variable in which to store output.
         This is a utils.io.CapturedIO object with stdout/err attributes
         for the text of the captured output.
 
         CapturedOutput also has a show() method for displaying the output,
         and __call__ as well, so you can use that to quickly display the
         output.
 
         If unspecified, captured output is discarded.
         """
     )
     @magic_arguments.argument('--no-stderr', action="store_true",
         help="""Don't capture stderr."""
     )
     @magic_arguments.argument('--no-stdout', action="store_true",
         help="""Don't capture stdout."""
     )
     @magic_arguments.argument('--no-display', action="store_true",
         help="""Don't capture IPython's rich display."""
     )
     @cell_magic
     def capture(self, line, cell):
         """run the cell, capturing stdout, stderr, and IPython's rich display() calls."""
         args = magic_arguments.parse_argstring(self.capture, line)
         out = not args.no_stdout
         err = not args.no_stderr
         disp = not args.no_display
         with capture_output(out, err, disp) as io:
             self.shell.run_cell(cell)
         if DisplayHook.semicolon_at_end_of_expression(cell):
             if args.output in self.shell.user_ns:
                 del self.shell.user_ns[args.output]
         elif args.output:
             self.shell.user_ns[args.output] = io
 
+    @skip_doctest
+    @magic_arguments.magic_arguments()
+    @magic_arguments.argument("name", type=str, default="default", nargs="?")
+    @magic_arguments.argument(
+        "--remove", action="store_true", help="remove the current transformer"
+    )
+    @magic_arguments.argument(
+        "--list", action="store_true", help="list existing transformers name"
+    )
+    @magic_arguments.argument(
+        "--list-all",
+        action="store_true",
+        help="list existing transformers name and code template",
+    )
+    @line_cell_magic
+    def code_wrap(self, line, cell=None):
+        """
+        Simple magic to quickly define a code transformer for all IPython's future imput.
+
+        ``__code__`` and ``__ret__`` are special variable that represent the code to run
+        and the value of the last expression of ``__code__`` respectively.
+
+        Examples
+        --------
+
+        .. ipython::
+
+            In [1]: %%code_wrap before_after
+               ...: print('before')
+               ...: __code__
+               ...: print('after')
+               ...: __ret__
+
+
+            In [2]: 1
+            before
+            after
+            Out[2]: 1
+
+            In [3]: %code_wrap --list
+            before_after
+
+            In [4]: %code_wrap --list-all
+            before_after :
+                print('before')
+                __code__
+                print('after')
+                __ret__
+
+            In [5]: %code_wrap --remove before_after
+
+        """
+        args = magic_arguments.parse_argstring(self.code_wrap, line)
+
+        if args.list:
+            for name in self._transformers.keys():
+                print(name)
+            return
+        if args.list_all:
+            for name, _t in self._transformers.items():
+                print(name, ":")
+                print(indent(ast.unparse(_t.template), "    "))
+            print()
+            return
+
+        to_remove = self._transformers.pop(args.name, None)
+        if to_remove in self.shell.ast_transformers:
+            self.shell.ast_transformers.remove(to_remove)
+        if cell is None or args.remove:
+            return
+
+        _trs = ReplaceCodeTransformer(ast.parse(cell))
+
+        self._transformers[args.name] = _trs
+        self.shell.ast_transformers.append(_trs)
+
+
 def parse_breakpoint(text, current_file):
     '''Returns (file, line) for file:line and (current_file, line) for line'''
     colon = text.find(':')
     if colon == -1:
         return current_file, int(text)
     else:
         return text[:colon], int(text[colon+1:])
     
 def _format_time(timespan, precision=3):
     """Formats the timespan in a human readable form"""
 
     if timespan >= 60.0:
         # we have more than a minute, format that in a human readable form
         # Idea from http://snipplr.com/view/5713/
         parts = [("d", 60*60*24),("h", 60*60),("min", 60), ("s", 1)]
         time = []
         leftover = timespan
         for suffix, length in parts:
             value = int(leftover / length)
             if value > 0:
                 leftover = leftover % length
                 time.append(u'%s%s' % (str(value), suffix))
             if leftover < 1:
                 break
         return " ".join(time)
 
     
     # Unfortunately the unicode 'micro' symbol can cause problems in
     # certain terminals.  
     # See bug: https://bugs.launchpad.net/ipython/+bug/348466
     # Try to prevent crashes by being more secure than it needs to
     # E.g. eclipse is able to print a µ, but has no sys.stdout.encoding set.
     units = [u"s", u"ms",u'us',"ns"] # the save value   
     if hasattr(sys.stdout, 'encoding') and sys.stdout.encoding:
         try:
             u'\xb5'.encode(sys.stdout.encoding)
             units = [u"s", u"ms",u'\xb5s',"ns"]
         except:
             pass
     scaling = [1, 1e3, 1e6, 1e9]
         
     if timespan > 0.0:
         order = min(-int(math.floor(math.log10(timespan)) // 3), 3)
     else:
         order = 3
-    return u"%.*g %s" % (precision, timespan * scaling[order], units[order])
+    return "%.*g %s" % (precision, timespan * scaling[order], units[order])

docs/autogen_api.py

@@ -1,83 +1,81 b''
 #!/usr/bin/env python
 """Script to auto-generate our API docs.
 """
 
 import os
 import sys
 
 pjoin = os.path.join
 
 here = os.path.abspath(os.path.dirname(__file__))
 sys.path.append(pjoin(os.path.abspath(here), 'sphinxext'))
 
 from apigen import ApiDocWriter
 
 source = pjoin(here, 'source')
 
 #*****************************************************************************
 if __name__ == '__main__':
     package = 'IPython'
     outdir = pjoin(source, 'api', 'generated')
     docwriter = ApiDocWriter(package,rst_extension='.rst')
     # You have to escape the . here because . is a special char for regexps.
     # You must do make clean if you change this!
     docwriter.package_skip_patterns += [r'\.external$',
                                         # Extensions are documented elsewhere.
                                         r'\.extensions',
-                                        # Magics are documented separately
-                                        r'\.core\.magics',
                                         # This isn't API
                                         r'\.sphinxext',
                                         # Shims
                                         r'\.kernel',
                                         r'\.terminal\.pt_inputhooks',
                                         ]
 
     # The inputhook* modules often cause problems on import, such as trying to
     # load incompatible Qt bindings. It's easiest to leave them all out. The
     docwriter.module_skip_patterns += [
         r"\.lib\.inputhook.+",
         r"\.ipdoctest",
         r"\.testing\.plugin",
         # Backwards compat import for lib.lexers
         r"\.nbconvert\.utils\.lexers",
         # We document this manually.
         r"\.utils\.py3compat",
         # These are exposed in display
         r"\.core\.display",
         r"\.lib\.display",
         # Shims
         r"\.config",
         r"\.consoleapp",
         r"\.frontend$",
         r"\.html",
         r"\.nbconvert",
         r"\.nbformat",
         r"\.parallel",
         r"\.qt",
         # this is deprecated.
         r"\.utils\.version",
         # Private APIs (there should be a lot more here)
         r"\.terminal\.ptutils",
     ]
     # main API is in the inputhook module, which is documented.
 
     # These modules import functions and classes from other places to expose
     # them as part of the public API. They must have __all__ defined. The
     # non-API modules they import from should be excluded by the skip patterns
     # above.
     docwriter.names_from__all__.update(
         {
             "IPython",
             "IPython.display",
         }
     )
 
     # Now, generate the outputs
     docwriter.write_api_docs(outdir)
     # Write index with .txt extension - we can include it, but Sphinx won't try
     # to compile it
     docwriter.write_index(outdir, 'gen.txt',
                           relative_to = pjoin(source, 'api')
                           )
     print ('%d files written' % len(docwriter.written_modules))