##// END OF EJS Templates
Clear filter
Show hidden
Commit Message Age Author Refs
load previous
r28846:0e5bd036
Use environment variable to identify conda / mamba Conda and mamba both set an environment variable which refers to the base environment's executable path, use that in preference to less reliable methods, but fall back on the other approaches if unable to locate the executable this way. Additionally, change the search to look for the bare command name rather than the command within the top level of the active environment, I'm dubious this approach works with any current conda / mamba version which usually place their executable links in a `condabin` directory or elsewhere not at the same level as the Python executable.
Shaun Walbridge
0
r28845:0657e1e1
Get back to "Executing" stable (#14511) It should now support Python 3.13
M Bussonnier
merge
0
r28844:494acd8e
more info about ipdb configuration (#14509)
M Bussonnier
merge
0
r28843:e153b0fb
Get back to "Executing" stable It should now support Python 3.13
M Bussonnier
0
r28842:e5d1a069
Fix use of pyside6 >= 6.7.0 (#14510) Fixes #14463. Using `pyside6 >= 6.7.0` as the `qt6` gui loop gives the following error: ``` In [1]: %gui qt6 In [2]: Traceback (most recent call last): File "/Users/iant/micromamba/envs/temp/bin/ipython", line 8, in <module> sys.exit(start_ipython()) ^^^^^^^^^^^^^^^ File "/Users/iant/github/ipython/IPython/__init__.py", line 130, in start_ipython return launch_new_instance(argv=argv, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/iant/micromamba/envs/temp/lib/python3.12/site-packages/traitlets/config/application.py", line 1075, in launch_instance app.start() File "/Users/iant/github/ipython/IPython/terminal/ipapp.py", line 317, in start self.shell.mainloop() File "/Users/iant/github/ipython/IPython/terminal/interactiveshell.py", line 917, in mainloop self.interact() File "/Users/iant/github/ipython/IPython/terminal/interactiveshell.py", line 902, in interact code = self.prompt_for_code() ^^^^^^^^^^^^^^^^^^^^^^ File "/Users/iant/github/ipython/IPython/terminal/interactiveshell.py", line 845, in prompt_for_code text = self.pt_app.prompt( ^^^^^^^^^^^^^^^^^^^ File "/Users/iant/micromamba/envs/temp/lib/python3.12/site-packages/prompt_toolkit/shortcuts/prompt.py", line 1035, in prompt return self.app.run( ^^^^^^^^^^^^^ File "/Users/iant/micromamba/envs/temp/lib/python3.12/site-packages/prompt_toolkit/application/application.py", line 978, in run result = loop.run_until_complete(coro) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/iant/micromamba/envs/temp/lib/python3.12/asyncio/base_events.py", line 674, in run_until_complete self.run_forever() File "/Users/iant/micromamba/envs/temp/lib/python3.12/asyncio/base_events.py", line 641, in run_forever self._run_once() File "/Users/iant/micromamba/envs/temp/lib/python3.12/asyncio/base_events.py", line 1948, in _run_once event_list = self._selector.select(timeout) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/iant/micromamba/envs/temp/lib/python3.12/site-packages/prompt_toolkit/eventloop/inputhook.py", line 150, in select self.inputhook(InputHookContext(self._r, input_is_ready)) File "/Users/iant/github/ipython/IPython/terminal/pt_inputhooks/qt.py", line 50, in inputhook _appref = app = QtGui.QApplication([" "]) ^^^^^^^^^^^^^^^^^^ AttributeError: module 'PySide6.QtPrintSupport' has no attribute 'QApplication' If you suspect this is an IPython 8.28.0.dev bug, please report it at: https://github.com/ipython/ipython/issues or send an email to the mailing list at ipython-dev@python.org You can print a more detailed traceback right now with "%tb", or use "%debug" to interactively debug it. Extra-detailed tracebacks for bug-reporting purposes can be enabled via: %config Application.verbose_crash=True ``` This is because we use the imported module's `__dict__` to get the classes and functions available in the module here: https://github.com/ipython/ipython/blob/9b8cd4a397e5894ffeadad52477bb53e0fb664fc/IPython/external/qt_loaders.py#L309-L311 This no longer works as not all the classes and functions are in the `__dict__`. The solution in this PR is to use `dir(module)` instead. I have tested this locally using `pyside6` 6.6.3.1, 6.7.0, 6.7.1 and 6.7.2 and it works for me. It also successfully creates Matplotlib plots using for example ``` In [1]: %matplotlib qt6 In [2]: import matplotlib.pyplot as plt In [3]: plt.plot([1,3,2]) ``` It would be good to get independent confirmation that this fixes other downstream libraries as I tend to work directly with IPython and IPyKernel.
M Bussonnier
merge
0
r28841:eca32508
Linting
Ian Thomas
0
r28840:02160bc7
Fix use of pyside6 >= 6.7.0
Ian Thomas
0
r28839:49314810
more infor about ipdb configuration
M Bussonnier
0
r28838:9b8cd4a3
`AssertionError`: `assert _xterm_term_title_saved` in WSL (#14480) `AssertionError`: `assert _xterm_term_title_saved` In some (unknown) situation, it is possible that the `_xterm_term_title_saved` is unset, but the code would make a call to `_restore_term_title_xterm`, resulting in `AssertionError`. At least on replicatable reproduction is returning from `ipython` to `pudb` via `^D^D` on an empty cell. See more details in https://github.com/ipython/ipython/pull/14480 Signed-off-by: Stavros Ntentos <133706+stdedos@users.noreply.github.com> <details> <summary><em>was:</em></summary> In some (unknown) situation, it is possible that the `_xterm_term_title_saved` is unset, but the code would make a call to `_restore_term_title_xterm`, resulting in `AssertionError`. As title stacking does not seem to be getting traction (https://github.com/microsoft/terminal/issues/14575), do not set the `xterm` variants of `_set_term_title` / `_restore_term_title`. WSL detection: https://superuser.com/a/1749811/533196 Additionally, almost-`black` the file. </details>
M Bussonnier
merge
0
r28837:59ab7217
`AssertionError`: `assert _xterm_term_title_saved` In some (unknown) situation, it is possible that the `_xterm_term_title_saved` is unset, but the code would make a call to `_restore_term_title_xterm`, resulting in `AssertionError`. At least on replicatable reproduction is returning from `ipython` to `pudb` via `^D^D` on an empty cell. See more details in https://github.com/ipython/ipython/pull/14480 Signed-off-by: Stavros Ntentos <133706+stdedos@users.noreply.github.com>
Stavros Ntentos
0
r28836:b90943a0
back to dev
M Bussonnier
0
r28835:82690a06
release 8.27.0
M Bussonnier
0
r28834:7af1c011
Whats new 8.27 (#14508)
M Bussonnier
merge
0
r28833:ef9d41ea
Whats new 8.27
M Bussonnier
0
r28832:5c8bc514
Fix completion in indented lines dropping prefix when jedi is disabled (#14474) Fixes #14470 I would appreciate someone testing this PR.
M Bussonnier
merge
0
r28831:6bde8f6c
Don't force matplotlib backend names to be lowercase (#14475) This is something I missed as part of moving the backend resolution to Matplotlib. Since version 3.9.0 Matplotlib handles the case-sensitivity of backend names, making them all lowercase for internal use. But since matplotlib/matplotlib#28473 uppercase letters are allowed in backend names of the form `module://soMe_moDule.sOmE_NAme`. Hence we need to not force backend names to be lowercase in IPython when passing them to Matplotlib. It is just a one line change that happens to make the code simpler, plus a new test. The Matplotlib change will be released in 3.9.1 which is due shortly, possibly today. The new test checks the Matplotlib version to know whether uppercase backend module names are allowed or not.
M Bussonnier
merge
0
r28830:d5762c16
Fix showing SystemExit exception raise inside except handler (#14503) Doing something like this: ```python try: 5 / 0 except Exception as e: raise SystemExit ``` was hitting an error inside UltraTB, creating a long traceback of its internals (which, ironically, UltraTB itself then displays correctly :-). `ListTB.get_exception_only()` calls the `ListTB.structured_traceback()` method *specifically* - even if `self` is a subclass, it won't use the subclass's method. However, the exception chaining in that method uses recursion by calling `self.structured_traceback()`, which will use a subclass's method. Tuples were added as an option there to support exception chaining, but not all of the machinery in connected classes expects a tuple. This just skips the exception chaining logic for the `etb=None` case, when we're showing the exception only. I'm not sure this is necessarily the best fix, but I didn't want to spend too much time following code around a module that's old enough to vote. Closes #12104
M Bussonnier
merge
0
r28829:82eba444
Fix some tests on Python 3.13 RC1 (#14504) Random little odds and ends that have changed in 3.13 break the tests. I think this should get the CI working again. :crossed_fingers: Closes #14457 Closes #14458
M Bussonnier
merge
0
r28828:2e34c770
Autoformat with darker
Thomas Kluyver
0
r28827:44bffab7
Relax string checks in test_pinfo_docstring_dynamic Allows this test to pass on Python 3.13
Thomas Kluyver
0
load next
< 1 .. 5 6 7 8 9 .. 1361 >
showing 20 out of 27211 commits