summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* default to level=0 if tag not presentstuebinm2023-11-141-1/+1
| | | | seems to give slightly better results
* re-add help textstuebinm2023-11-141-1/+7
| | | | whoops, forgot this bit
* sort platforms by level/layerstuebinm2023-11-141-3/+11
|
* rewrite it in Haskellstuebinm2023-11-1411-7016/+609
|
* platforms: update (also use local_ref)stuebinm2023-11-061-0/+498
|
* platform data bundlestuebinm2023-11-056-28/+29676
| | | | this is pretty low-effort, but it seems to work fine?
* update gleam & depsstuebinm2023-10-302-116/+151
|
* allow _ for spacesstuebinm2023-07-091-0/+2
| | | | | (easier to use in browsers/shells where spaces lead to unwanted behaviour, e.g. starting a search or ending an argument)
* separate domains, fuzzy matching all-roundstuebinm2023-07-091-104/+138
|
* add LeitpunkteJohann1502023-06-183-13/+696
| | | | (by johann150, with some modification by stuebinm)
* refactor lookups to separate functionsJohann1502023-06-171-25/+38
| | | | This allows them to be used separately.
* drop CSV headerJohann1502023-06-171-0/+2
|
* blackhole favicon.ico requestsJohann1502023-06-171-0/+2
| | | | | It is not necessary to do anything for these kinds of requests, so instead of making the fuzzy search do stuff, just blackhole these requests.
* broken encoding in some stationsstuebinm2023-05-111-1/+2
| | | | with thanks to lis: https://mk.catgirlsfor.science/notes/9en6jb2ig8
* use haystack for searchstuebinm2023-05-113-85/+136
| | | | for some reason it doesn't find Karlsruhe, but basically everything else??
* url percent encoding & (naïve) fuzzy matchingstuebinm2023-04-264-6/+170
|
* initial commitstuebinm2023-04-222-0/+99