brobertson / lace2 Goto Github PK

If a verified word corresponds to a page number zone, then add a @data-page-number attribute to

dc:identifier is not a good key, since it isn't present in other packages, requiring a non-optimized search through them. Instead, use code like:

xquery version "3.1";
import module namespace apputil="http://exist-db.org/xquery/apps";
import module namespace repo="http://exist-db.org/xquery/repo";
let $root := repo:get-root()
let $uri := "http://heml.mta.ca/Lace/Images/actaphilippietac00bonnuoft"
return $root || apputil:resolve($uri)

Now that eXist-db corrects this, we have double '%' signs.

Right now, this error is thrown:
exerr:ERROR XPTY0004: The actual cardinality for parameter 1 does not match the cardinality declared in the function's signature: app:sortCollection($collectionUri as xs:string) item(). Expected cardinality: exactly one, got 0. [at line 563, column 37, source: /db/apps/lace/modules/app.xql]
In function:
app:sortCollection(xs:string) [563:17:/db/apps/lace/modules/app.xql]
app:sidebyside(node(), map(), xs:string, xs:integer?) [-1:-1:/db/apps/lace/modules/app.xql]

If I sideload an updated app xar, will it overwite the old one? If I put it on the public repo, will it ask for people to update?

Provide a switch to search only hocr or only tei, or both.

The volume-wide replacement is very slow
for these two searches:
let $search_results := doc($documentPath)//xh:span[@Class='ocr_word' and text()=$query]/@data-manually-confirmed
let $search_results2 := doc($documentPath)//xh:span[@Class='ocr_word' and text()=$query]/text()
and these queries:
let $foo := update value $search_results with 'true'
let $foo2 := update replace $search_results2 with $correctedForm
the time is 350 seconds. Most of the drag is in the second of update of the text().
I think this is an eXist-db bug.
Check against future versions of eXist.

The raw content links don't give raw data. "raw" and "selected" seem to do the same thing.

If I sideload a text fill, will it automatically upload the image file from the public repo? Will it clobber the same id if it's already installed?

These would be <div @data-start-primary-text="true" or @data-start-zone="primary-text" or just @id="primary_text_zone1">. Gotta think that through. Is there any advantage to doing this dynamically as we draw the rectangles? Currently, I think not and would do this just before compiling to the unedited TEI. Well, the obvious case is that we could use this to inspect the page#s.

Because of #13 , and because volume-wide is pretty serious, make ctrl-return only change the same spans in the same page.

This means we don't have to rely on relative document-order to determine the beginning of the pertinent span of elements for a given urn.

It's also applying to svg elements, which is stupid.