Wonderful idea for next time :-) On Thu, Sep 11, 2025 at 3:45 AM Adrien Grand <jpou...@gmail.com> wrote:
> Hey David, > > Thank you for doing this! I don't have good ideas for new good first > issues. > > However, something that we need to periodically do and hasn't been done in > a long time now is taking a look at the slower Lucene tests and checking if > they are slow for a good reason or not. Some tests are slow for a good > reason (e.g. PostingsFormatTestCase and DocValuesFormatTestCase sub classes > usually), but often, they are not slow for a good reason but because they > do too much merging or index many documents with the SimpleText codec. For > instance I've seen TestPriorityQueue and TestLateInteractionRescorer often > take several seconds recently, I'm not sure if this is needed. There are > likely many other tests in this case. This could be a good opportunity to > get familiar with Lucene's tooling (tests.slowestTests, tests.profile, > etc.), how Lucene works (text analysis, indexing, merging, searching), and > getting a first change in. I wouldn't mind separate PRs for each slow test > that gets a speedup to help share this task across multiple participants to > the Hackathon. > > On Wed, Sep 10, 2025 at 6:16 PM David Smiley <dsmi...@apache.org> wrote: > >> I'll be running a hackathon tomorrow at Community-over-Code with >> Nazerke. I'll be sharing the "good first issue" issues as a potential list >> of items for new contributors: >> https://github.com/apache/lucene/issues?q=is%3Aissue%20state%3Aopen%20label%3A%22good%20first%20issue%22 >> (only 9 issues as I write this) >> >> You can help our project & contributors & hackathon simply by doing a >> little label grooming. If you created an issue recently and think it's a >> "good first issue" for a contributor then add it! Come to think of it, >> maybe even some bugs might be such. Or maybe something on your mind hasn't >> been filed yet; please do! >> >> ~ David Smiley >> Apache Lucene/Solr Search Developer >> http://www.linkedin.com/in/davidwsmiley >> > > > -- > Adrien >
