Ideally csv2srophe can both create new records and add data to existing records. Here are a few notes on what some of those functionalities might look like and how they could be implemented. This is not a priority, however
titleStmt: would append funder, sponsor, creator editor, respStmts based on a selected template file (deduplicating as needed)
publicationStmt would only update the date value
seriesStmts should be updated based on the selected template file (i.e., run the template.xqm function to get the seriesStmts and then add and deduplicate)
prepend a revisionDesc/change for the appending of data
For both persons and places:
having a bibl off-set would be useful. Create the new bibls as usual, but with the attributes include the offset (by default for new records this is 0)
similar to existing name offsets that are used after headword creation, but with an override based on existing highest xml:id of the bibl or name
for names, separate out the matches (where you just append the source id, or ignore if it was 'resp') from the non-matches (where you create the el and add at the end of the sequence)
the 'headwords' for appending data should be instead treated as regular names
for abstracts, if one exists to be appended, append and add the URIs of seriesStmts that do not already exist in the existing abstract to it. Otherwise (if no abstract exists to append), take the existing abstract and update the corresp attribute with all the seriesStmt URIs
add all new idnos -- might need, for example with JEG, to specify if one should be added under the syriaca one or all appended to end
relations can be appended as usual, though possible deduplicate by matching ref and uris involved first; can be a 'if match, include the source instead'
for persons:
dates: exact matches update the sources; else add in correct order (can have multiple of each)
gender: exact matches update the sources; else add after
for places:
location: exact matches update the sources; else add after
Ideally csv2srophe can both create new records and add data to existing records. Here are a few notes on what some of those functionalities might look like and how they could be implemented. This is not a priority, however
For both persons and places:
for persons:
for places: