Python-announce-list November 2020

python-announce-list@python.org

17 participants
18 discussions

pip 20.3 release (heads-up for potential disruption)

by Sumana Harihareswara

On behalf of the Python Packaging Authority, I am pleased to announce that we have just released pip 20.3, a new version of pip. You can install it by running `python -m pip install --upgrade pip`. This is an important and disruptive release -- we explained why in a blog post last year https://pyfound.blogspot.com/2019/12/moss-czi-support-pip.html . We even made a video about it: https://www.youtube.com/watch?v=B4GQCBBsuNU . Blog post with details: https://pyfound.blogspot.com/2020/11/pip-20-3-new-resolver.html Highlights include: * **DISRUPTION**: Switch to the new dependency resolver by default. Watch out for changes in handling editable installs, constraints files, and more: https://pip.pypa.io/en/latest/user_guide/#changes-to-the-pip-dependency-res… * **DEPRECATION**: Deprecate support for Python 3.5 (to be removed in pip 21.0) * **DEPRECATION**: pip freeze will stop filtering the pip, setuptools, distribute and wheel packages from pip freeze output in a future version. To keep the previous behavior, users should use the new `--exclude` option. * Support for PEP 600: Future ‘manylinux’ Platform Tags for Portable Linux Built Distributions. * Add support for MacOS Big Sur compatibility tags. The new resolver is now *on by default*. It is significantly stricter and more consistent when it receives incompatible instructions, and reduces support for certain kinds of constraints files, so some workarounds and workflows may break. Please see our guide on how to test and migrate, and how to report issues: https://pip.pypa.io/en/latest/user_guide/#changes-to-the-pip-dependency-res… . You can use the deprecated (old) resolver, using the flag `--use-deprecated=legacy-resolver`, until we remove it in the pip 21.0 release in January 2021. In pip 21.0 we will also remove support for Python 2.7. You can find more details (including deprecations and removals) [in the changelog](https://pip.pypa.io/en/stable/news/), and you can find thank-yous and instructions on reporting issues at https://pyfound.blogspot.com/2020/11/pip-20-3-new-resolver.html . Thank you, Sumana Harihareswara pip project manager Changeset Consulting https://changeset.nyc

7 months, 4 weeks

EuroPython videos all on archive.org

by M.-A. Lemburg

Over the last few weeks, Anthon van der Neut, our media work group chair for EP2015 and EP2016, put in a lot of effort into getting all our conference videos on archive.org, the Internet Archive. Archive.org is not meant as an alternative for YouTube to watch the videos, but it allows you to retrieve the original uploads, and as such also functions as a backup location for the us and the community. He first downloaded all videos from our YouTube account (over 2.5TB as of this writing), then enriched the meta data based on the talk information we have on the websites, fixed issues he found, and then uploaded the videos and meta data to our archive.org account. Overall, around 40 hours of work went into this. We now have more than 1000 conference videos available on our archive.org account, covering EP2014 - EP2020. * EuroPython videos on archive.org * https://archive.org/details/@europythonvideos Thanks, Anthon, for making this possible ! Help spread the word -------------------- Please help us spread this message by sharing it on your social networks as widely as possible. Thank you ! Link to the blog post: https://blog.europython.eu/post/636029350382387200/europython-videos-all-on… Tweet: https://twitter.com/europython/status/1332661436632457216 Thanks, -- EuroPython 2021 Team https://www.europython-society.org/

8 months

[RELEASE] Python 3.9.1rc1 is now ready for testing

by Łukasz Langa

Python 3.9.1rc1 is the release candidate of the first maintenance release of Python 3.9. Go get it here: https://www.python.org/downloads/release/python-391rc1/ <https://www.python.org/downloads/release/python-391rc1/> Assuming no critical problems are found prior to 2020-12-11, the currently scheduled release date for 3.9.1, no code changes are planned between this release candidate and the final release. That being said, please keep in mind that this is a pre-release of 3.9.1 and as such its main purpose is testing. Maintenance releases for the 3.9 series will continue at regular bi-monthly intervals, with 3.9.2 planned for end of January 2021. Installer news 3.9.1rc1 is the first version of Python to support macOS 11 Big Sur. With Xcode 11 and later it is now possible to build “Universal 2” binaries which work on Apple Silicon. We are providing such an installer as the macosx11.0 variant. This installer can be deployed back to older versions, tested down to OS X 10.9. As we are waiting for an updated version of pip, please consider the macosx11.0 installer experimental. This work would not have been possible without the effort of Ronald Oussoren, Ned Deily, and Lawrence D’Anna from Apple. Thank you! In other news, this is the first version of Python to default to the 64-bit installer on Windows. The installer now also actively disallows installation on Windows 7. Python 3.9 is incompatible with this unsupported version of Windows. What’s new in Python 3.9.1rc1? We’ve made 240 changes since v3.9.0 which is a significant amount. To compare, 3.8.1rc1 only saw 168 commits since 3.8.0. See the full change log at https://docs.python.org/release/3.9.1rc1/whatsnew/changelog.html <https://docs.python.org/release/3.9.1rc1/whatsnew/changelog.html>. For more information about features included in the 3.9 series, see the “What’s New in Python 3.9 <https://docs.python.org/release/3.9.0/whatsnew/3.9.html>” document. What about Python 3.8.7rc1? There’s additional work needed to make this release support macOS 11 Big Sur. This should be ready next week, stay tuned. We hope you enjoy the new releases! Thanks to all of the many volunteers who help make Python Development and these releases possible! Please consider supporting our efforts by volunteering yourself or through organization contributions to the Python Software Foundation. https://www.python.org/psf/ <https://www.python.org/psf/> More resources Online Documentation <https://docs.python.org/3.9/> PEP 596 <https://www.python.org/dev/peps/pep-0596/>, 3.9 Release Schedule PEP 619 <https://www.python.org/dev/peps/pep-0619/>, 3.10 Release Schedule Report bugs at https://bugs.python.org <https://bugs.python.org/>. Help fund Python and its community <https://www.python.org/psf/donations/>. Your friendly release team, Ned Deily @nad <https://discuss.python.org/u/nad> Steve Dower @steve.dower <https://discuss.python.org/u/steve.dower> Łukasz Langa @ambv <https://discuss.python.org/u/ambv>

8 months

ANN: wxPython 4.1.1

by Robin Dunn

Announcing wxPython 4.1.1 ========================= PyPI: https://pypi.org/project/wxPython/4.1.1 Extras: https://extras.wxPython.org/wxPython4/extras/ Pip: ``pip install wxPython==4.1.1`` New and improved in this release: * This should have been mentioned in the notes for the last release, but alas, it wandered away and got lost. wxWidgets is now validating the flags passed when adding items to a sizer, to ensure that they are the correct flags for the type of the sizer. If the given flags do not make sense, for example using horizontal alignment flags in a horizontal box sizer, then a wxAssertionError error is raised. * Fixed missing binder for wxEVT_STC_AUTOCOMP_SELECTION_CHANGE. (#1613) * DataViewModel.HasValue can be overridden and will inform the DataViewCtrl whether or not an item and column has data. If HasValue returns False, then GetValue for that item/col will not be called. This allows a distinction between a truly empty cell, and one that has a value even if it is an empty string. (#1600) * Added flag that allows blocking of item dragging in the UltimateListControl class. (PR#1620) * Add the column index to notification events in UltimateListControl (PR#1630) * Added orientation parameter to UltimateListControl.GetScrollPos. (PR#1632) * wx.lib.agw.aui.AuiNotebook RemovePage() now hides the removed page, so it needs to be shown again if it is reused in another place. (PR#1668) * Fixed issue that could modify `bytes` objects under Python. (PR#1680) * Added wx.lib.agw.aui.EVT_AUI_PANE_CLOSE event which is sent when a AUI (the agw version) Pane has been closed (after it has been closed, not when it is about to be closed, which is when EVT_AUI_PANE_CLOSE is sent.) (PR#1628) * Exposed the wx.DC methods GetGraphicsContext and SetGraphicsContext. Depending on the platform and the type of the DC, there may be a wx.GraphicsContext used for the implementation of the DC. If so, the GetGraphicsContext method enables access to it. Be sure to check that the return value is not None before trying to use it. * Simplified the implementation of the wx.App.InitLocale method. See the MigrationGuide for more information. * Added wx.lib.agw.aui.AUI_DOCKART_HINT_WINDOW_BORDER_COLOUR constant so the hint window border color can be themed as well. * The wx.lib.mixins.listCtrl.CheckListCtrlMixin is now obsolete because wx.ListCtrl has new functionality which does pretty much the same thing. In fact there is some overlap in method names which may trip up some use cases. It is advised to drop the use of CheckListCtrlMixin and just use the wx.ListBox functionality. You will need to call EnableCheckBoxes to turn it on, and you may need to change some event handlers or overloaded methods. * wx.html2.WebView is now able to use Microsoft's Edge browser component as its backend renderer. This should improve the capabilities of the WebView widget on Windows, and be more consistent with the WebViews on the other platforms, compared to the original IE 11 backend. Using this backed requires that a new-ish version of the Edge browser is installed on the end user's computer. * Added the wx.Image.ConvertToRegion method. This lets you create a wx.Region from an image and a specified color or the mask if the image has one. This was done to workaround a bug in wxMac, but it seems worthwhile enough to keep it around even after the bug was fixed. * Added the missing context manager methods for wx.LogNull. (#1842) * Refactored ScrolledThumbnail out of agw.ThumbnailCtrl so as to be usable outside of ThumbnailCtrl. What is wxPython? ----------------- wxPython is a cross-platform GUI toolkit for the Python programming language. It allows Python programmers to create programs with a robust, highly functional graphical user interface, simply and easily. It is implemented as a set of Python extension modules that wrap the GUI components of the popular wxWidgets cross platform library, which is written in C++. Supported platforms are Microsoft Windows, Mac OS X and macOS, and Linux or other unix-like systems with GTK2 or GTK3 libraries. In most cases the native widgets are used on each platform to provide a 100% native look and feel for the application. What is wxPython Phoenix? ------------------------- wxPython's Project Phoenix is a new from-the-ground-up implementation of wxPython, created with the intent of making wxPython “better, stronger, faster than he was before.” In other words, this new implementation is focused on improving speed, maintainability and extensibility of wxPython, as well as removing most of the cruft that had accumulated over the long life of Classic wxPython. The project has been in development off and on, mostly behind the scenes, for many years. For the past few years automated snapshot builds have been available for those adventurous enough to try it, and many people eventually started using the snapshots in their projects, even for production releases. While there are still some things on the periphery that need to be completed, the core of the new wxPython extension modules which wrap the wxWidgets code has been stable for a long time now. Due to some things being cleaned up, reorganized, simplified and dehackified wxPython Phoenix is not completely backwards compatible with wxPython Classic. This is intended. In general, however, the API differences tend to be minor and some applications can use Phoenix with slight, or even with no modifications. In some other cases the correct way to do things was also available in Classic and it's only the wrong way that has been removed from Phoenix. For more information there is a Migration Guide document available at: https://docs.wxpython.org/MigrationGuide.html The new wxPython API reference documentation, including all Python-specific additions and customizations, and docs for the wx.lib package, is located at: https://docs.wxpython.org/

8 months

Fil v0.11.0, a memory profiler for scientists and data scientists

by Itamar Turner-Trauring

Your code reads some data, processes it, and uses too much memory. In order to reduce memory usage, you need to figure out: 1. Where peak memory usage is, also known as the high-water mark. 2. What code was responsible for allocating the memory that was present at that peak moment. That's exactly what Fil will help you find. Fil an open source memory profiler designed for data processing applications written in Python, and includes native support for Jupyter. It is designed to be high-performance and easy to use. At the moment it only runs on Linux and macOS. You can learn more about Fil at https://pythonspeed.com/fil or on GitHub at https://github.com/pythonspeed/filprofiler/. v0.11 includes performance improvements and less intrusive behavior under Jupyter. Fil vs. other Python memory tools There are two distinct patterns of Python usage, each with its own source of memory problems. In a long-running server, memory usage can grow indefinitely due to memory leaks. That is, some memory is not being freed. * If the issue is in Python code, tools like `tracemalloc` <https://docs.python.org/3/library/tracemalloc.html> and Pympler <https://pypi.org/project/Pympler/> can tell you which objects are leaking and what is preventing them from being leaked. * If you're leaking memory in C code, you can use tools like Valgrind <https://valgrind.org/>. Fil, however, is not aimed at memory leaks, but at the other use case: data processing applications. These applications load in data, process it somehow, and then finish running. The problem with these applications is that they can, on purpose or by mistake, allocate huge amounts of memory. It might get freed soon after, but if you allocate 16GB RAM and only have 8GB in your computer, the lack of leaks doesn't help you. Fil will therefore tell you, in an easy to understand way: 1. Where peak memory usage is, also known as the high-water mark. 2. What code was responsible for allocating the memory that was present at that peak moment. 3. This includes C/Fortran/C++/whatever extensions that don't use Python's memory allocation API (`tracemalloc` only does Python memory APIs).

8 months, 1 week

Microdict : A high performance python hash table library that is generally faster and consumes significantly less memory than Python Dictionaries.

by Touqir Sajed

Microdict v0.1 is released : https://github.com/touqir14/Microdict What is Microdict? A high performance python hash table library that is generally faster and consumes significantly less memory than Python Dictionaries. Benchmarks show that it can be upto 7 times more memory efficient in comparison to Python Dictionaries. Microdict is a typed python dictionary that heavily relies on C extensions. Moreover, benchmarks show that its underlying C implementation can outperform both Google's Swiss Table and Facebook's F14 hash tables. Currently it supports the following 5 hash table types: "i32:i32" -> 32 bit signed keys and 32 bit signed values "i32:i64" -> 32 bit signed keys and 64 bit signed values "i64:i32" -> 64 bit signed keys and 32 bit signed values "i64:i64" -> 64 bit signed keys and 64 bit signed values "str:str" -> string keys and string values. To learn more: <P><A HREF="https://github.com/touqir14/Microdict">Microdict v0.1</A> A high performance python hash table library that is generally faster and consumes significantly less memory than Python Dictionaries. (19-11-2020). Contact me at shuhash6(a)gmail.com. Cheers, Touqir

8 months, 1 week

EuroPython "Ask me Anything" - Nov Edition

by M.-A. Lemburg

Dear Community, after our first successful “Ask me Anything” (AMA) session in September https://blog.europython.eu/post/629951901661790208/europython-ask-me-anythi… we want to run another meeting this Thursday to answer any questions you may have, share our knowledge or help you in planning your online event. Some of the topics we can cover: - our tools research and findings - our concepts for running an online conference or event - our experience with running EuroPython online - what we could do to support you - how our Zoom license sharing works - how you can apply for a grant - how the EuroPython Society works - how we run the EuroPython organization with fully remote work groups and, of course, anything else :-) Join us on Thursday ------------------- If you’re interested in joining the call, please send your name and email to helpdesk(a)europython.eu and we’ll send you the invite for the call on Thursday, Nov 19, starting at 19:30 CET. When requesting the invite, please also consider adding more context, in case we’d have to prepare for the answers, so we can better prepare and make the meeting more effective. Please note that we will be recording the meeting and put it up on our YouTube account after the call, just like we did last time. BTW: We are making our conference resources available on our website, in case you have missed our blog post earlier this year. Many thanks. Help spread the word -------------------- Please help us spread this message by sharing it on your social networks as widely as possible. Thank you ! Link to the blog post: https://blog.europython.eu/post/634934022827622400/europython-ask-me-anythi… Tweet: https://twitter.com/europython/status/1328274983743877121 Thanks, -- EuroPython 2020 Team https://ep2020.europython.eu/ https://www.europython-society.org/

8 months, 1 week

pymemtrace - Tracing Python Memory Usage

by Paul Ross

Welcome to *pymemtrace*, a collection of tools for monitoring and analysing Python's memory usage including that of Python Extensions. These tools work at various levels, at different granularities and with different runtime costs so you can choose one appropriate to the problem at hand. They range from periodically measuring total memory usage down to reporting every malloc and free call per Python line or C function call. The tools are: * *process* is a very lightweight way of logging the total memory usage at regular time intervals. It can plot memory over time with plotting programs such as gnuplot. Some examples https://pymemtrace.readthedocs.io/en/latest/examples/process.html * *cPyMemTrace* is a memory tracer written in C that can report total memory usage for every function call/return for both C and Python sections. Some examples: https://pymemtrace.readthedocs.io/en/latest/examples/c_py_mem_trace.html and a technical note on cPyMemTrace: https://pymemtrace.readthedocs.io/en/latest/tech_notes/cPyMemTrace.html. * *DTrace*: There are a number of D scripts that can trace the low level ``malloc()`` and ``free()`` system calls and report how much memory was allocated and by whom. Some DTrace examples: https://pymemtrace.readthedocs.io/en/latest/examples/dtrace.html and a technical note on DTrace https://pymemtrace.readthedocs.io/en/latest/tech_notes/dtrace.html * *trace_malloc* is a convenience wrapper around the Python standard library tracemalloc module. This can take memory snapshots before and after code blocks and show the change in memory caused by that code by module and line compensating for the cost of tracemalloc itself. Some trace_malloc examples https://pymemtrace.readthedocs.io/en/latest/examples/trace_malloc.html * *debug_malloc_stats* is a wrapper around the sys._debugmallocstats function that can take snapshots of memory before and after code execution and report the significant differences in the Python small object allocator. Some debug_malloc_stats examples https://pymemtrace.readthedocs.io/en/latest/examples/debug_malloc_stats.html Within the documentation is also some (rather limited, for the moment) advice on tracking down memory leaks with Python extensions. On PyPi: https://pypi.org/project/pymemtrace/ Project: https://github.com/paulross/pymemtrace Documentation: https://pymemtrace.readthedocs.io/en/latest/index.html I hope that you find this useful as I do! Paul Ross.

8 months, 1 week

Wing Python IDE 7.2.7 has been released

by Wingware

Wing 7.2.7 improves running unit tests for Django and other testing frameworks, fixes Command-Alt-Click to add multiple selections on macOS, changes Goto Source in the Testing tool to display the innermost project file stack frame, and makes a number of other usability improvements. Details: https://wingware.com/news/2020-11-13 Downloads: https://wingware.com/downloads == About Wing == Wing is a light-weight but full-featured Python IDE designed specifically for Python, with powerful editing, code inspection, testing, and debugging capabilities. Wing's deep code analysis provides auto-completion, auto-editing, and refactoring that speed up development. Its top notch debugger works with any Python code, locally or on a remote host. Wing also supports test-driven development, version control, UI color and layout customization, and includes extensive documentation and support. Wing is available in three product levels: Wing Pro is the full-featured Python IDE for professional developers, Wing Personal is a free Python IDE for students and hobbyists (omits some features), and Wing 101 is a very simplified free Python IDE for beginners (omits many features). Learn more at https://wingware.com/

8 months, 1 week

Nikola v8.1.2 is out!

by Chris Warrick

On behalf of the Nikola team, I am pleased to announce the immediate availability of Nikola v8.1.2. This release has a bunch of new features, fixes some bugs, and improves compatibility with some dependencies and Windows. What is Nikola? =============== Nikola is a static site and blog generator, written in Python. It can use Mako and Jinja2 templates, and input in many popular markup formats, such as reStructuredText and Markdown — and can even turn Jupyter Notebooks into blog posts! It also supports image galleries, and is multilingual. Nikola is flexible, and page builds are extremely fast, courtesy of doit (which is rebuilding only what has been changed). Find out more at the website: https://getnikola.com/ Downloads ========= Install using `pip install Nikola`. Changes ======= Features -------- * Support for multiple authors per post — comma-separated, enabled by `MULTIPLE_AUTHORS_PER_POST` setting (Issue #3252) * Add `navbar_custom_bg` theme option to `bootstrap4` and document options for `bootstrap4` better (Issue #3443) * Add Marathi translation * Restore `caps` typogrify filter (wraps strings of capital letters with `<span class="caps">` (via Issue #3405) * Improve `nikola auto` performance with rate-limiting, support `-n` argument to pass to `nikola build` (Issue #3401) * Add support for the Utterances comment system Bugfixes -------- * Fix previewimages (post- and root-relative) in bootblog4/galleries featured posts for non-server-root sites (Issue #3471) * Windows: Also fix symlinks when installing from source with pip; if possible, enable Developer Mode and run `git config --global core.symlinks true` before cloning the Nikola repo * Fix clicking on links to image files. baguetteBox now requires a link (`<a>`) to have an `<img>` as a child in order to make the link activate the lightbox. (Issue #3472) * Fix `LINK_CHECK_WHITELIST` having issues due to mixing Unicode and bytestrings (Issue #3466) * Add support for `nbconvert>=6.0.0` (Issue #3457) * Read files as utf-8-sig to allow BOM in input files * Don’t break slugs with slashes in `doc` directive (Issue #3450) * Avoid warnings from type annotations in `auto` caused by missing `aiohttp` (Issue #3451) * Ensure query strings and fragments are kept with `URL_TYPE = "full_path"` (Issue #3448) * Don’t run typogrify filters on `<title>` tag to avoid adding extra tags (Issue #3405) * Fix handling of duplicate plugins on Windows * Allow else clause in post-list plugin. (Issue #3436) * Ensure `type` metadata value from plugins is preserved (Issue 3445) * Fix path handlers with slashes on Windows -- Chris Warrick <https://chriswarrick.com/> PGP: 5EAAEA16

8 months, 1 week

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

Python-announce-list November 2020