geminispace.info

gemini search engine
git clone https://git.clttr.info/geminispace.info.git
Log (Feed) | Files | Refs (Tags) | README | LICENSE

commit 14aa70b3c65c25224721478364331ad468e10f39
parent 88fa8d44136444f0f24e556da07275b5d50bfd8e
Author: Natalie Pendragon <natpen@natpen.net>
Date:   Sat, 29 Feb 2020 08:33:12 -0500

Improve README readability

Diffstat:
MREADME.md | 12++++++------
1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/README.md b/README.md @@ -27,28 +27,28 @@ as this guide to [mailing list etiquette](https://man.sr.ht/lists.sr.ht/etiquett # Roadmap / TODOs -- *general code cleanup*: most notably crawl.py. There are a lot +- **general code cleanup**: most notably crawl.py. There are a lot of hacks in there that I put in for expediency, but haven't taken the time to address. -- *improve the indexing*: currently, the url is prepended to +- **improve the indexing**: currently, the url is prepended to the page content, and everything is simply indexed with the default indexer. I think a better solution would be to have urls indexed with a url-specific indexer that doesn't do things like, e.g., porter-stemming, which I assume the default indexer is doing. -- *extend the index to handle binary links in Geminispace*: +- **extend the index to handle binary links in Geminispace**: currently, there's a hack in the code to simply skip anything that looks like a binary link. I think with the above improvement to how indexing works, they could be made very effectively searchable. Also in this vein, binary links should be identified via their mime types probably, instead of the suffix hack used now. -- *add tests*: there aren't any yet! -- *add functionality to create a mock index*: this would +- **add tests**: there aren't any yet! +- **add functionality to create a mock index**: this would be useful for local hacking on serve.py, so one does not need to perform a real scrape of Geminispace to do said hacking. -- *add paging functionality*: haven't fully thought through +- **add paging functionality**: haven't fully thought through how this would work, but I think it would be nice? Also open to feedback though if anyone thinks a better UI would be to simply present as many results as exist (I think that