Feature/#56 section wise collation

Feature

Summary

This MR provides

• a new module, collate.xqm, which takes care of the collation process. Only passages following tei:milestone are considered for extracting text nodes.
• a new module, commons.xqm, which holds variables and functions used in several modules.
• a refactored version of the RESTAPI endpoint returning *.txt files.

Compliance to “Definition of Done”

• Unit tests passed
• Code reviewed
• Acceptance criteria met
• Functional tests passed
• Non-Functional requirements met
• Product Owner accepts the User Story

Use Cases

If you can, please provide use cases for this feature.

Documentation

• I updated the README (if applicable)
• I provided my functions with appropriate documentation
• I updated existing documentation

Tests

Are we able to test this new feature?

• Yes, everything can be done via unit tests.

Changelog

• I added a statement to the CHANGELOG.

Version number

• I bumped the version number in build.properties.

Related Tickets

Closes

#56 (closed)

/cc Mathias Göbel, Frank Schneider, Michelle Weidling

exist-app/modules/collate.xqm

@@ -5,9 +5,6 @@ xquery version "3.1";
  : the TEI files and uses them as an input for the plain text creation. These
  : serve as a basis for the collation with CollateX in the project's CI/CD
  : pipelines.
- : 
- : The chunks are defined in the file "xpaths_for_collation.xml" which has the
- : TextGrid URI textgrid:3rzkb.
  :)
 
 module namespace coll="http://ahikar.sub.uni-goettingen.de/ns/collate";
@@ -19,14 +16,13 @@ import module namespace fragment="https://wiki.tei-c.org/index.php?title=Milesto
 
 declare variable $coll:textgrid := "/db/apps/sade/textgrid";
 declare variable $coll:data := $coll:textgrid || "/data";
-declare variable $coll:xpaths := doc($coll:data || "/3rzkb.xml");
 declare variable $coll:txt := $coll:textgrid || "/txt";
 
 
 declare function coll:main() 
 as xs:string+ {
     coll:create-txt-collection-if-not-available(),
-    for $text in coll:get-transcriptions-and-transliterations()[descendant::tei:milestone] return
+    for $text in coll:get-transcriptions-and-transliterations() return
         let $relevant-text := coll:get-relevant-text($text)
         let $file-name := coll:make-file-name($text)
         return
@@ -55,13 +51,17 @@ as xs:boolean {
         false()
 };
 
+(:~
+ : An example for the file name is 
+ : syriac-Brit_Lib_Add_7200-3r131-transcription.txt
+ :)
 declare function coll:make-file-name($text as element(tei:text))
 as xs:string {
     let $lang-prefix := coll:get-language-prefix($text)
     let $title-from-metadata := coll:create-metadata-title-for-file-name($text)
-    let $file-name-plus-text-type := coll:make-file-name-suffix($text)
+    let $uri-plus-text-type := coll:make-file-name-suffix($text)
     return
-        $lang-prefix || "-" || $title-from-metadata || "-" || $file-name-plus-text-type
+        $lang-prefix || "-" || $title-from-metadata || "-" || $uri-plus-text-type
 };
 
 declare function coll:get-language-prefix($text as element(tei:text))
@@ -77,8 +77,7 @@ as xs:string? {
         (: although the transliteration may have another language than
         the transcription, the latter remains decisive for the prefix :)
         case "transliteration" return
-            (: transliterations are always encoded before transcriptions :)
-            switch ($text/following-sibling::tei:text[@type = "transcription"]/@xml:lang)
+            switch ($text/root()//tei:text[@type = "transcription"]/@xml:lang)
                 case "ara" return "arabic"
                 case "karshuni" return "karshuni"
                 case "syc" return "syriac"
@@ -188,7 +187,7 @@ as xs:string {
         for $text in $texts return
             coll:prepare-plain-text-creation($text)
     return
-        coll:format-string($prepared-texts)
+        coll:format-and-normalize-string($prepared-texts)
 };
 
 (:~ 
@@ -223,7 +222,7 @@ as xs:string {
         $text
 };
 
-declare function coll:format-string($strings as xs:string+)
+declare function coll:format-and-normalize-string($strings as xs:string+)
 as xs:string {
     string-join($strings, " ")
     => replace(" @", "")