urgent documentation updates

[kmcdono2]

Requests for documentation:

• which id and code fields in the lwmdb map onto pre-existing ids from METS/ALTO from newspaper collections (e.g. is newspaper_code = NLP #?)

Would be ideal to include this in db schema somehow.

[griff-rees]

Quoting https://github.com/Living-with-machines/data-culture-newspapers/issues/25 (thanks @mcollardanuy):

An example in a table provided to link records with lwmdb Newspaper records:

NLP,issue,art_num
3406,1208,art0039

In this case

• The NLP column -> Newspaper.publication_code field. The import process adds 000 in front -> 0003406
• The issues column -> Issue.issue_code field (an Issue of a Newspaper). In the example. 1208 is mmdd (and querying indicates the year in that example is 1888). This fits the issue_code field in Issue.issue_code -> 0003406-18881208 which matches 000{NLP}-{YYYY}{MM}{DD}
• The art_num column -> Item.item_code field (an Item of an Issue, such as an article or add). Here art0039, would be stored as Item.item_code 003406-18881208-art0039.

[kmcdono2]

@griff-rees - fantastic!

Q: is the way issues & art_num expressed the same way/format across providers (e.g. BNA/JISC/HMD/LWM)?

[griff-rees]

Thanks @kmcdono2: will hopefully be easier to test this now #121 is closed (all of those fields have now been indexed).