Clean up submissions for FRUS 1934-39, etc.; place into frus-tags.xml

[joewiz]

frus-tags-todo.zip

cleanup-scripts.zip

[joewiz]

@awmarrs I found them!

[awmarrs]

Huzzah!

[joewiz]

Closed with https://github.com/HistoryAtState/tags/commit/7cad575052dfddbe8bfcc51888684b8b929a447b.

The procedure:

1. Open .xlsx submission in Excel, save to .xls
2. Import .xls in oXygen (File > Import > Excel)
3. Rename root/row as studies/study, rename columns as title, link, tags, unofficial-tags-tbd
4. Save to eXist, in /db/apps/tags/tagged-resources/frus-tags-todo

5. Convert exact matches to final format:

   xquery version "3.1";
   
   let $studies := collection('/db/apps/tags/tagged-resources/frus-tags-todo')//study
   for $tag in $studies//tag[not(@id)]
   let $match := doc('/db/apps/tags/taxonomy/taxonomy.xml')//label[. = $tag]/../id
   return
   if ($match) then
   (
   update insert attribute id {$match} into $tag,
   update value $tag with ''
   )
   else ()

6. Merge these into existing frus-tags document (and move non-matches into unofficial-tags-tbd):

   xquery version "3.1";
   
   let $frus-tags := doc('/db/apps/tags/tagged-resources/frus/frus-tags.xml')//study
   let $new-studies := 
      for $study in collection('/db/apps/tags/tagged-resources/frus-tags-todo')//study
      let $link := $study/link
      let $vol-id := substring-after($link, 'historicaldocuments/')
      return 
          element study {
              element title { normalize-space($study/title) },
              element link { "https://history.state.gov/historicaldocuments/" || $vol-id },
              element tags { 
                  for $tag in $study/tags/tag[@id]
                  order by $tag/@id
                  return 
                      element tag { attribute id { $tag/@id } }
              },
              element unofficial-tags-tbd {
                  for $tag in distinct-values(($study/tags/tag[not(@id)], $study/unofficial-tags-tbd/tag))
                  order by $tag
                  return
                      element tag { $tag }
              }
          }
   let $re-sorted-studies := 
      element studies {
          for $study in ($frus-tags, $new-studies)
          let $link := $study/link
          let $vol-id := substring-after($link, 'historicaldocuments/')
          order by $vol-id
          return $study
      }
   return 
      xmldb:store('/db/apps/tags/tagged-resources/frus', 'frus-tags.xml', $re-sorted-studies)

From the commit message:

This leaves only 8 published volumes without tags:

• frus1889
• frus1890
• frus1891
• frus1892
• frus1893
• frus1902app3 (intern noted this wasn’t available to him/her)
• frus1969-76ve14p2
• frus1977-80v30

(As discussed before, we intentionally did not submit frus1861-99Index & frus1900-18Index for tagging.)

As I did with policy studies tags, I moved all tags that did not have an exact match with entries in the taxonomy into the “unofficial-tags-tbd.” I didn’t proof these for typos / slight differences from the controlled values in the taxonomy. With further proofreading / fuzzy searching, some of these tags could be added and thus available to users of our website.

[joewiz]

Revised submissions as of https://github.com/HistoryAtState/tags/commit/4ff8e0a75f4b68b67082a947e170a2ff09cc6ad8: frus-tags-imported-vsfs-submissions-2015-16.zip

[github-actions[bot]]

:tada: This issue has been resolved in version 1.0.0 :tada:

The release is available on GitHub release

Your semantic-release bot :package::rocket: