virtualtime

Settings | Report Duplicate

0

I Use This!

Inactive

Commits : Listings

Analyzed about 15 hours ago. based on code collected about 15 hours ago.

Commit Message	Contributor	Files Modified	Lines Added	Lines Removed	Code Location	Date
Jan 19, 2025 — Jan 19, 2026 Showing page 1 of 5 Search / Filter on:
Merge pull request #14 from davidfraser/really-trap-import-locks	rorygrandin	More...				over 5 years ago
Rename the `date` wrapper class to `date_no_importerror` to match what we did with `time` Note in both cases that we don't actually patch this, but it exists as an implementation people can use / we can patch in the future if necessary	David Fraser	More...				over 5 years ago
Add support for datetime extensions to strftime: %f, %z, %Z This is done by implementing the same algorithm to adjust the format string in Python as is done in datetimemodule.c Add unit tests for the datetime extensions to strftime Handle the fact that importing _strptime doesn't actually raise an error - we've got enough testing with the things that do, and so this isn't a problem	David Fraser	More...				over 5 years ago
Don't raise an error if the call to import _strptime doesn't raise an error - it doesn't seem to, but this isn't a problem	David Fraser	More...				over 5 years ago
Fix up derived time class to derive from time, not date Added strftime fallback for date class	David Fraser	More...				over 5 years ago
Make a time class so we can test the time class's issues and their solution	David Fraser	More...				over 5 years ago
Correct Linux implementation to use byref for structs, and access the value of the timezone reference	David Fraser	More...				over 5 years ago
Patch datetime.strftime to handle ImportErrors as well This doesn't yet handle %z, %Z and %f, all of which are implemented by the datetimemodule.c code for wrap_strftime	David Fraser	More...				over 5 years ago
Implement a "virtual" date class - this doesn't actually adjust time, but does prevent ImportErrors	David Fraser	More...				over 5 years ago
Remove experimental test code that snuck into commit, and patch the virtual_datetime class as well Rather than testing patching, simply test the classes directly (both virtualtime's base datetime and its datetime classes)	David Fraser	More...				over 5 years ago
Add overridden methods for timetuple, that can fall back if it gets an ImportError This is necessary since both now() and utcnow() call timetuple	David Fraser	More...				over 5 years ago
Fix the function name we're showing the attributes of Add safety fallbacks for datetime.datetime.now() and datetime.datetime.utcnow() Rearrange tests to be understandable and consistent, and test virtualtime as well (Currently disabled tests that we don't pass on virtualtime) Also increased time of wait for background thread so it doesn't time out while debugging Ensure _strptime is missing from sys.modules (but this still works even when it needs to import it in my environment)	David Fraser	More...				over 5 years ago
Added alternative implementations of now() and utcnow() using win32api/libc, that we can use to check (We will also use these as a fallback if now/utcnow fail with this error)	David Fraser	More...				over 5 years ago
Added some checks on function results	David Fraser	More...				over 5 years ago
Add unit tests that can reproduce what happens when the import lock error occurs in datetime module functions	David Fraser	More...				over 5 years ago
Merge pull request #13 from matthewhampton/datetime_tz_hashable	yaseen-mowzer-hexagon	More...				over 5 years ago
The whole purpose of this set of extensions to the base datetime_tz was to allow comparison with naive instances of the base datetime classes (by just hoping that they refer to local times). We believe that this was misguided, and likely to cause trouble. It also put us in a slightly untenable situation in trying to figure out how to implement the __hash__ function in a way that would not break some key invariant on the comparitors. For those two reasons, we've decided to simply remove this capability on PY3. Python3 is a lot stricter about comparing mixed types, so I think this is in the spirit of the activities required for update anyway.	Matt Hampton	More...				over 5 years ago
Revert "If we do not override this method, we end up with the virtualtime subclass being unhashable."	Matt Hampton	More...				over 5 years ago
If we do not override this method, we end up with the virtualtime subclass being unhashable.	Matt Hampton	More...				over 5 years ago
Merge pull request #12 from yaseen-mowzer-hexagon/fix-astimezone	Matthew Hampton	More...				almost 6 years ago
Make tzinfo parameter of astimezone optional	Yaseen Mowzer	More...				almost 6 years ago
Merge pull request #11 from dee42/py3	David Fraser	More...				almost 7 years ago
Command string needs to be different on py2 and py3, because on py2 there's no buffer on stdin/out	Dee Moore	More...				about 7 years ago
Fix formats that python3 actually gets correct and our patch does not	Dee Moore	More...				about 7 years ago
More Python3 fixes	Dee Moore	More...				about 7 years ago
to read and write bytes from py3 stdin/stdout we need to use their buffer	Dee Moore	More...				about 7 years ago
Avoid checking functions that are not there on Python 3	Dee Moore	More...				about 7 years ago
Merge pull request #8 from matthewhampton/import_pandas_numpy_first	David Fraser	More...				over 8 years ago
It is pandas actually that matters - not numpy	Matt Hampton	More...				over 8 years ago
Try and import numpy and pandas, if present, before patching datetime, else they get upset.	Matt Hampton	More...				over 8 years ago

←
1
2
3
4
5
→