##// END OF EJS Templates
Fix #13654, improve performance of auto match for quotes...
Fix #13654, improve performance of auto match for quotes As pointed out in #13654, auto matching of quotes may take a long time if the prefix is long. To be more precise, the longer the text before the first quote, the slower it is. This is all caused by the regex pattern used: `r'^([^"]+|"[^"]*")*$'`, which I suspect is O(2^N) slow. ```python In [1]: text = "function_with_long_nameeee('arg" In [2]: import re In [3]: pattern = re.compile(r"^([^']+|'[^']*')*$") In [4]: %timeit pattern.match(text) 10.3 s ± 67.2 ms per loop (mean ± std. dev. of 7 runs, 1 loop each) In [5]: %timeit pattern.match("1'") 312 ns ± 0.775 ns per loop (mean ± std. dev. of 7 runs, 1,000,000 loops each) In [6]: %timeit pattern.match("12'") 462 ns ± 1.95 ns per loop (mean ± std. dev. of 7 runs, 1,000,000 loops each) In [7]: %timeit pattern.match("123'") 766 ns ± 6.32 ns per loop (mean ± std. dev. of 7 runs, 1,000,000 loops each) In [8]: %timeit pattern.match("1234'") 1.59 µs ± 20.9 ns per loop (mean ± std. dev. of 7 runs, 1,000,000 loops each) ``` But the pattern we want here can actually be detected with a Python implemention in O(N) time.

File last commit:

r27087:0e59ae55
r27762:c179c2a5
Show More
test_compilerop.py
71 lines | 2.2 KiB | text/x-python | PythonLexer
# coding: utf-8
"""Tests for the compilerop module.
"""
#-----------------------------------------------------------------------------
# Copyright (C) 2010-2011 The IPython Development Team.
#
# Distributed under the terms of the BSD License.
#
# The full license is in the file COPYING.txt, distributed with this software.
#-----------------------------------------------------------------------------
#-----------------------------------------------------------------------------
# Imports
#-----------------------------------------------------------------------------
# Stdlib imports
import linecache
import sys
# Third-party imports
import pytest
# Our own imports
from IPython.core import compilerop
#-----------------------------------------------------------------------------
# Test functions
#-----------------------------------------------------------------------------
def test_code_name():
code = 'x=1'
name = compilerop.code_name(code)
assert name.startswith("<ipython-input-0")
def test_code_name2():
code = 'x=1'
name = compilerop.code_name(code, 9)
assert name.startswith("<ipython-input-9")
def test_cache():
"""Test the compiler correctly compiles and caches inputs
"""
cp = compilerop.CachingCompiler()
ncache = len(linecache.cache)
cp.cache('x=1')
assert len(linecache.cache) > ncache
def test_proper_default_encoding():
# Check we're in a proper Python 2 environment (some imports, such
# as GTK, can change the default encoding, which can hide bugs.)
assert sys.getdefaultencoding() == "utf-8"
def test_cache_unicode():
cp = compilerop.CachingCompiler()
ncache = len(linecache.cache)
cp.cache(u"t = 'žćčšđ'")
assert len(linecache.cache) > ncache
def test_compiler_check_cache():
"""Test the compiler properly manages the cache.
"""
# Rather simple-minded tests that just exercise the API
cp = compilerop.CachingCompiler()
cp.cache('x=1', 99)
# Ensure now that after clearing the cache, our entries survive
linecache.checkcache()
assert any(
k.startswith("<ipython-input-99") for k in linecache.cache
), "Entry for input-99 missing from linecache"