Update token annotations from gregorycrane/gAGDT

[jacobwegner]

gregorycrane/gAGDT

TODOs

• [x] Greg will review his local edits
• [x] Jake will expand Farnoosh's glosses to Odyssey
• [x] Bring newest edits online

[jacobwegner]

@gregorycrane @farnoosh-shamsian:

I've refreshed the treebanks as we'd discussed on the call today (including trying to map additional Persian glosses throughout).

@gregorycrane I have a PR up for you that fixed a couple of issues with the XML files:

https://github.com/gregorycrane/gAGDT/pull/4

I also updated the Gists I shared on the call with the missing glosses / errors:

• Treebank ingestion errors
• Missing glosses

There is a new type of error; I'm currently just skipping these words rather than trying to assume a headword. Once we have a headword attrib set, I can re-run the ingestion:

# From tlg0012.tlg002.perseus-grc2
No @head found [form=ἔλθοι sentence_id=2189859 ref="16.101"]
No @head found [form=ἀλητεύων sentence_id=2189859 ref="16.101"]
No @head found [form=ἔτι sentence_id=2189859 ref="16.101"]
No @head found [form=γὰρ sentence_id=2189859 ref="16.101"]
No @head found [form=καὶ sentence_id=2189859 ref="16.101"]
No @head found [form=ἐλπίδος sentence_id=2189859 ref="16.101"]
No @head found [form=αἶσα sentence_id=2189859 ref="16.101"]

[jacobwegner]

https://beyond-translation-gagdt-dev.herokuapp.com/reader/urn:cts:greekLit:tlg0012.tlg002.perseus-grc2:1.1?mode=syntax-trees

[jacobwegner]

64 will be used to improve our next iteration