The script extracts fragments references from a catalogue. For each fragment in the catalogue the documents not starting with BC are collected and saved to the database under uncuratedReferences property as a list of objects with "File name" column without extension and quotes as document and the list of pages from the "Page" as integers as pages.

• Node
• Lodash

To run the script, use the following command:

node index.js <path to folder> [-o <output file>] [-local] [-no-ssl] [--uri <MongoDB URI>] [--database <MongoDB database>]

The script takes the following arguments:

• <path to folder> - the path to the folder containing the hits data
• • -o <output file> (optional) - saves the output JSON to a file instead of updating the MongoDB database
• -local (optional) - saves the output JSON to a file and skips updating the MongoDB database
• -no-ssl (optional) - connects to the MongoDB server without SSL
• --uri <MongoDB URI> (optional) - the MongoDB connection string
• --database <MongoDB database> (optional) - the name of the MongoDB database

If -local is specified, the script will save the output JSON to a file with a default filename "data.json". If -o is specified, the script will save the output JSON to the file specified in the argument.

The catalogue directory should contain windows1252 encoded text files with the following format:

"Search Results"

"Summary"
"Saved on";"04.08.2018 17:45:55"
"Searched for";"K 6447"
"In index(es)";"X:\xxx\_OA\Indice\Index.pdx"
"Number of document(s) found";"5"              
"Number of instance(s) found";"5"              

"File name","Title","Page","Search Instance"
"Fincke 2000 Augenleiden nach keilschriftlichen Quellen. Untersuchungen zur altorientalischen Medizin WMF 70.pdf","","185","5 ......�........�.....�..... 178 K. 6447+7086+ ... cf. K. 2354+ ... K. 6461 Z. "

"File name","Title","Page","Search Instance"
"Hee�el 2016 Medizinische Texte aus dem Alten Mesopotamien Translating Writings of Early Scholars.pdf","","14","K. 3237 + K. 6447 + K. 7086 + K. 8356 + K. 8800 "

"File name","Title","Page","Search Instance"
"Borger HKL 1 [1967].pdf","","282","Ebeling KMI 45f.)+K 6447 C~,u~l.)+K 9828+K 11868(AMT 6/"

"File name","Title","Page","Search Instance"
"Borger HKL 2 [1975].pdf","","162","38) auch + K 6447 (von Thompson verwertet) + ~m 637 ("

"File name","Title","Page","Search Instance"
"BC 2.pdf","","392","sick people. [K. 6447] Portion out of the middle, 32in. by 31in., . "

The fragment ID is parsed from the file name:

• <number>.csv => K.<number without leading zeroes>
• Sm <number>.csv => Sm.<number without leading zeroes>
• DT <number>.csv => DT.<number without leading zeroes>
• Rm <number>.csv => Rm.<number without leading zeroes>
• Rm 2 <number>.csv => Rm-II.<number without leading zeroes>
• Rm II <number>.csv => Rm-II.<number without leading zeroes>
• BM <number>.csv => BM.<number without leading zeroes>
• AA-BB-CC 0DD.csv.csv => 18AA,<BB padded to to digits><CC padded to two digits>.<DD without leading zeroes>
• AA-BB-CC,DD.csv => 18AA,<BB padded to to digits><CC padded to two digits>.<DD without leading zeroes>
• CBS <id>.csv => CBS.<id>
• N <id>.csv => N.<id>
• UM <id>.csv => UM.<id>
• Ni <id>.csv => Ni.<id>
• Si <id>.csv => Si.<id>
• HS <id>.csv => HS.<id>
• VAT <id>.csv => VAT.<id>
• ANT <id>.csv => ANT.<id>
• AOS-1 <id>.csv => AOS-1.<id>
• DavidAstr <id>.csv => DavidAstr.<id>
• EAH <id>.csv => EAH.<id>
• GCBC <id>.csv => GCBC.<id>
• Lager <id>.csv => Lager.<id>
• MLC <id>.csv => MLC.<id>
• NCBS <id>.csv => NCBS.<id>
• NCBT <id>.csv => NCBT.<id>
• Newell <id>.csv => Newell.<id>
• Ott-Seal <id>.csv => Ott-Seal.<id>
• Ott-Tablet <id>.csv => Ott-Tablet.<id>
• RBC <id>.csv => RBC.<id>
• YBC <id>.csv => YBC.<id>

References from different files resulting in same fragment ID are combined for the final result.

If the file name does not match any pattern the file is ignored.

If saved locally, the JSON file can be imported into the database with the following command (change as needed):

function doit() {
  references = [
    // ...
  ]
  missing = [];
  references.forEach(({_id, uncuratedReferences}) => {
    if (db.getCollection('fragments').findOne({_id: _id})) {
      db.getCollection('fragments').update(
        {_id: _id},
        {
          $set: {
            uncuratedReferences
          }
        }
      );
    } else {
      missing.push({_id, uncuratedReferences});
    }
  });
  return missing;
}

doit();