Due to changes in the archive and library catalogues at the Swedish Labour Movement Archives and Library, there is widespread link rot on web pages linking to these catalogues.
This program is intended to be used as a Pandoc filter to update source code which contains such links.
The conversion for the archive catalogue is very simple. For the library
catalogue, post IDs have changed, and the mappings from old ones to new ones
are defined in kataids.json, with the format "oldid":"newid". To build the
filer (assuming the Pandoc libraries and the GHC compiler are available), run
make build. At build, the mapping in kataids.json is coded into the filter,
so if this file is changed, the filter should be rebuilt. The present version
of kataids.json contains mappings for all links to posts in the library
catalogue in the RSS feed for the institution's webpage category with articles
about their
collections. Links to
Kata posts with the prefixes http://kata.arbark.se/wsRecInfo.asp?idno= or
http://kata.arbark.se/wsVolumes.asp?idno= are updated with their matched IDs.
Links to Kata searches with the prefix
http://kata.arbark.se/wsHitList.Asp? are currently just redirected to the
search page in the new system. Links to posts without mapping for the IDs are
also redirected to the search page.
I include the body content for my article about the organization
BiS, as a HTML
file in the examples subdirectory. To run the filter on this file, and write
to the target examples/bisbodynew.html, use
pandoc -o examples/bisbodynew.html examples/bisbody.html --filter ./lablinkfix.
The filter should work with any format supported by Pandoc, not just HTML.