<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: Pawamoy</title><link>https://news.ycombinator.com/user?id=Pawamoy</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sat, 20 Jun 2026 09:50:16 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=Pawamoy" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by Pawamoy in "How uv got so fast"]]></title><description><![CDATA[
<p>No. When such upper bounds are respected, they contaminate other packages, because you have to add them yourself to be compatible with your dependencies. Then your dependents must add them too, etc. This brings only pain. Python 4 is not even a thing, core developers say there won't ever be a Python 4.h</p>
]]></description><pubDate>Sat, 27 Dec 2025 08:14:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=46400133</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=46400133</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46400133</guid></item><item><title><![CDATA[New comment by Pawamoy in "How the brain's activity, energy use and blood flow change as people fall asleep"]]></title><description><![CDATA[
<p>A friend shared his technique with me, and after adapting it to my needs, it works well. None of the other commonly mentioned techniques ever worked for me, because they ask for focus, which is the opposite of what my brain needs to fall asleep.<p>Here's his technique: pick a letter of the alphabet, and find as much words that start with this letter as you can. Once you can't find words anymore, pick the next letter. Doesn't work for me, my brain won't ever stop.<p>I noticed I have to visualize stuff in my head to fall asleep so my adaptation is to pick a single letter and a single word, and visualize it in my head, using it, manipulating it, experiencing it, whatever. For example: letter P, word Pineapple, imagine you're holding a pineapple, you feel the roughness of it's skin in your hands, you throw it in the air and catch it again, you take a knife and slice it on a wooden table on the beach, etc.. The dream kicks in seconds. Without external interruptions, after a few minutes I'm asleep (instead of rummaging for hours).<p>If you notice you're stuck in a loop/pattern (for me anything about text, like reading or writing, and voice, like listening and speaking, or stressful scenarios), just pick a new letter, pick a new word, visualize it.</p>
]]></description><pubDate>Tue, 28 Oct 2025 09:30:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=45730784</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=45730784</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45730784</guid></item><item><title><![CDATA[New comment by Pawamoy in "Show HN: Timep – A next-gen profiler and flamegraph-generator for bash code"]]></title><description><![CDATA[
<p>I'll get a closer look at the code tomorrow, but maybe you can provide an answer in the meantime: how do you handle line numbers? Bash's LINENO is unreliable, contrary to Zsh's, see <a href="https://gist.github.com/pawamoy/cca35f0f5ccd56665d6421e9b2d2d4bc" rel="nofollow">https://gist.github.com/pawamoy/cca35f0f5ccd56665d6421e9b2d2...</a>. That's why I eventually gave up writing any profiling/tracing/debugging/coverage tool for Bash and moved to support these in Zsh instead. Unfortunately Zsh is missing a ZSH_XTRACEFD like Bash, but that's another story (never finished contributing one). Here's my own attempt at a shell profiler: <a href="https://github.com/shellm-org/profiler" rel="nofollow">https://github.com/shellm-org/profiler</a> :)</p>
]]></description><pubDate>Mon, 25 Aug 2025 23:53:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=45020614</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=45020614</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45020614</guid></item><item><title><![CDATA[New comment by Pawamoy in "Ask HN: How do you monetize personal code if it's not an "app"?"]]></title><description><![CDATA[
<p>I follow a sponsorware strategy: public version with basic features, paid version (monthly subscription) with more features. When a funding goal is reached (dollars per month), a subset of paid features become available to everyone. Paying users essentially fund the development of new features. I don't have any "app", only tools and libraries :)</p>
]]></description><pubDate>Sat, 12 Apr 2025 23:21:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=43668720</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=43668720</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43668720</guid></item><item><title><![CDATA[New comment by Pawamoy in "Show HN: Transform your codebase into a single Markdown doc for feeding into AI"]]></title><description><![CDATA[
<p>Following the /llms.txt standard proposition, I create a MkDocs plugin that generates an /llms.txt file at the root of your site. So, same thing, but generates the Markdown document from your docs (possibly containing API reference) instead of your code.</p>
]]></description><pubDate>Sat, 15 Feb 2025 02:17:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=43055233</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=43055233</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43055233</guid></item><item><title><![CDATA[Show HN: Griffe, load Python APIs, find breaking changes]]></title><description><![CDATA[
<p>Hey all, Griffe was already submitted two years ago on HN, but it nicely evolved since then so I figured I'd submit it again myself, especially since I just finished rewriting its documentation and reworking its API :)<p>Griffe is a Python tool and library to load Python APIs as whole program signatures, kinda. With this data, static site generators can generate API docs. Griffe itself can use the data to compare two versions of your API and find breaking changes. No excuses for not bumping your major semver version now!<p>Griffe is used by mkdocstrings (I'm the author too), and therefore indirectly used by FastAPI, Pydantic, and many other cool projects :)<p>The docs provide a lot of hints, advice, recommendations for public APIs, deprecations, and Python code generally. They somewhat come from my own experience, and not everyone might share my opinions: happy to discuss about all this!</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=40967013">https://news.ycombinator.com/item?id=40967013</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 15 Jul 2024 11:55:00 +0000</pubDate><link>https://mkdocstrings.github.io/griffe/</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=40967013</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40967013</guid></item><item><title><![CDATA[New comment by Pawamoy in "Git-cliff – Generate changelog from the Git history"]]></title><description><![CDATA[
<p>I maintain git-changelog, which is a similar implementation in Python (started two years before git-cliff). When I discovered git-cliff a few months ago, I was very impressed by the number of things they support. Also, being written in Rust, it must be much more performant than my Python implementation (which indeed has trouble with huge Git histories). I have started recommending git-cliff to some of my users who request features I don't yet support :)<p>Great work, git-cliff devs!</p>
]]></description><pubDate>Fri, 28 Jun 2024 12:55:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=40820162</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=40820162</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40820162</guid></item><item><title><![CDATA[Show HN: Python) Markdown Exec, execute code blocks and render their output]]></title><description><![CDATA[
<p>Hey everyone, here's an extension I made for Python-Markdown (<a href="https://github.com/Python-Markdown/markdown">https://github.com/Python-Markdown/markdown</a>). It builds on top of PyMDown Extensions' SuperFences (<a href="https://facelessuser.github.io/pymdown-extensions/extensions/superfences/" rel="nofollow">https://facelessuser.github.io/pymdown-extensions/extensions...</a>), and allows Markdown writers to execute their Markdown code blocks to render the execution output in place of / in addition to the code blocks.<p>Languages supported: python/pycon; sh/bash/console; md (literate programming!)<p>Special fences: tree, to easily render file trees; pyodide, to render an interactive Ace.js editor that runs Python code thanks to Pyodide<p>Adding examples to the gallery is quite fun :) Don't hesitate to create an issue if you'd like me to add examples for another tool or Python library!</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=40692991">https://news.ycombinator.com/item?id=40692991</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Sat, 15 Jun 2024 21:57:19 +0000</pubDate><link>https://pawamoy.github.io/markdown-exec/gallery/</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=40692991</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40692991</guid></item><item><title><![CDATA[New comment by Pawamoy in "Textual TUI framework for Python adds CSS renderer"]]></title><description><![CDATA[
<p>> But again they don't export a Sphinx inventory so I cannot cross-reference their units<p>But they do! So you can!<p>The inventory is right here: <a href="https://textual.textualize.io/objects.inv" rel="nofollow">https://textual.textualize.io/objects.inv</a>. It's generated by mkdocstrings, the "autodoc" plugin for MkDocs. It's Sphinx compatible.</p>
]]></description><pubDate>Mon, 31 Oct 2022 09:48:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=33403690</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=33403690</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=33403690</guid></item><item><title><![CDATA[New comment by Pawamoy in "Griffe – analyse Python code to find breaking API changes or generate API docs"]]></title><description><![CDATA[
<p>Author here (I'm a bit late!): indeed Griffe's goal is to be able to support any Python code, natively or thanks to extensions. Thanks for sharing by the way!</p>
]]></description><pubDate>Sun, 05 Jun 2022 12:48:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=31630514</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=31630514</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=31630514</guid></item><item><title><![CDATA[New comment by Pawamoy in "Ask HN: How to raise funds for rare disease research?"]]></title><description><![CDATA[
<p>This is probably a bit off-topic, but I wonder if you know about the GenIDA project: <a href="https://genida.unistra.fr/" rel="nofollow">https://genida.unistra.fr/</a>. It's basically a research website where families can register and answer questionnaires about the everyday life of patients with rare diseases. The data can then be used to discover new things about these rare diseases. They have a Kleefstra cohort if I recall correctly.
Maybe you could get in touch with the people in charge, like Jean-Louis Mandel.</p>
]]></description><pubDate>Tue, 08 Feb 2022 12:41:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=30257714</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=30257714</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=30257714</guid></item><item><title><![CDATA[New comment by Pawamoy in "It’s OK for your open source library to be a bit shitty (2015)"]]></title><description><![CDATA[
<p>Why not checking the contributors instead of the forks? Just pick the most regular ones, or the ones that contributed the most (in number of PRs).</p>
]]></description><pubDate>Sun, 17 May 2020 08:20:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=23210552</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=23210552</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=23210552</guid></item><item><title><![CDATA[New comment by Pawamoy in "Ask HN: What are your favorite developer-efficiency tips?"]]></title><description><![CDATA[
<p>My favorite as well.<p>I think Terminator gave me a real boost on productivity. The split feature is great. And it has a lot more good things, like increase/decrease/reset text size, resize/rotate/expand or even drag-and-drop splits, grouping (to write in multiple splits at once), and the scrolling is done very well: scroll up a bit to stop following the output, scroll to bottom to follow again.</p>
]]></description><pubDate>Sat, 09 May 2020 09:39:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=23123329</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=23123329</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=23123329</guid></item><item><title><![CDATA[New comment by Pawamoy in "Take care editing bash scripts (2019)"]]></title><description><![CDATA[
<p>To expand on this, yes, you can force Bash to read more lines and keep them in memory before executing them by wrapping them in blocks. These blocks can be (), {}, functions, and maybe other structures.<p>In addition to this, you can make sure no appended code is ever executed by explicitly running "exit" at the end of your block. I actually used this trick in a self git-updating script (the updated version could contain more lines at the end).</p>
]]></description><pubDate>Wed, 06 May 2020 09:41:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=23089283</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=23089283</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=23089283</guid></item><item><title><![CDATA[New comment by Pawamoy in "How do we stop people from blinding other drivers with aftermarket LEDs?"]]></title><description><![CDATA[
<p>I even have this problem with bikes. Not motor bikes, just bikes. Some of them have a front light that just blinds me. I can't determine their speed anymore, not the direction they're moving in, nor if there is something else between them and me. Once I even confused someone on their bike with a random street light, and almost crashed into them, because the angle and speed at which we were moving made it appear immobile. I often feel like a powerful light is far more dangerous than no light at all.</p>
]]></description><pubDate>Mon, 16 Mar 2020 20:49:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=22597555</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=22597555</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=22597555</guid></item><item><title><![CDATA[New comment by Pawamoy in "Bash-lib: Library for bash utility methods and tools"]]></title><description><![CDATA[
<p>Check out Basher: <a href="https://github.com/basherpm/basher" rel="nofollow">https://github.com/basherpm/basher</a></p>
]]></description><pubDate>Sun, 10 Nov 2019 18:11:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=21499245</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=21499245</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=21499245</guid></item><item><title><![CDATA[New comment by Pawamoy in "Htop Explained"]]></title><description><![CDATA[
<p>Shameless plug: aria2p. I built an interactive interface very similar to htop to see your aria2 downloads progress.<p><a href="https://github.com/pawamoy/aria2p" rel="nofollow">https://github.com/pawamoy/aria2p</a></p>
]]></description><pubDate>Sat, 02 Nov 2019 11:23:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=21427037</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=21427037</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=21427037</guid></item><item><title><![CDATA[New comment by Pawamoy in "Ask HN: Website / GitHub for shaming websites that do not work in Firefox?"]]></title><description><![CDATA[
<p>Woops, "didn't read", will delete later</p>
]]></description><pubDate>Tue, 01 Oct 2019 21:28:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=21130895</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=21130895</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=21130895</guid></item><item><title><![CDATA[New comment by Pawamoy in "Ask HN: Website / GitHub for shaming websites that do not work in Firefox?"]]></title><description><![CDATA[
<p>Similar to <a href="https://github.com/dumb-password-rules/dumb-password-rules" rel="nofollow">https://github.com/dumb-password-rules/dumb-password-rules</a></p>
]]></description><pubDate>Tue, 01 Oct 2019 21:26:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=21130870</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=21130870</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=21130870</guid></item><item><title><![CDATA[Ask HN: How do you find developers with same interests, for side projects?]]></title><description><![CDATA[
<p>I have a lot of side projects on which I try to work a bit everytime I have the opportunity, but honestly sometimes it feels like I will never finish all the work I'm giving to myself.<p>Contributing on GitHub and other platforms is nice, but it's often one way contributions: when you contribute to a project, there are few chances the author will contribute back to one of yours. Of course I'm not asking for the authors of projects I contributed to to contribute back, as it's not something I expect from people or myself. But I would love to know some developers with projects at the same level of popularity as mine, who want/need help and are willing to contribute back, and with who I could discuss, share ideas, plan sprints, etc.<p>In short: I would like to create a friendly group of developers who share the same interests. For example, for me it would be around CLI tools, but more specifically, CLI tools that help managing repositories, shell libraries (Bash/Zsh), and tools that help improving your comfort/productivity in the terminal.<p>The question is: how to find them? How do YOU find them, if you do? I don't know any platform serving this purpose. Of course I can just go on GitHub and contact people with projects similar to mine, but it seems a bit random and I think I would spend a lot of time sending emails for small results.</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=20780676">https://news.ycombinator.com/item?id=20780676</a></p>
<p>Points: 2</p>
<p># Comments: 2</p>
]]></description><pubDate>Fri, 23 Aug 2019 18:16:08 +0000</pubDate><link>https://news.ycombinator.com/item?id=20780676</link><dc:creator>Pawamoy</dc:creator><comments>https://news.ycombinator.com/item?id=20780676</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=20780676</guid></item></channel></rss>