https://github.com/mediawiki-utilities/python-mwrefs is a handy tool to grab all the refs from a wiki. How can we best proceed to analyze non-ISBN (either old, or badly cited) books on svwp?
The raw refdump will be of use for T175348, so make sure to share it before processing!
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | Jopparn | T205384 Analyze the most common sources on svwp | |||
| Resolved | Alicia_Fagerving_WMSE | T202541 Examine non-ISBN refs on svwp |
The output of the tool looks like:
page_id page_title rev_id rev_timestamp reference 6097238 Phoebe Point 41704536 20171025051715 <ref>…ref content…</ref>
Thus it will have to be processed further by extracting only the last column of the TSV file.
cut -f 5 svwp.tsv > svwp_refs_only.tsv
Then: find all instances of {{Bokref}} → create author : title pairs -> create a frequency list of those.
Note that author can be encoded in different ways, either as efternamn= |förnamn= or författare=; can be normalized by joining the separate names.
The sorted book refs will live here: https://gist.github.com/Vesihiisi/400529978ad1757db9b096ff1ee6545a