andrzej
/
micro365
1
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

rolled back 'form_of'

This commit is contained in:
Andrzej Stepien 2023-08-02 12:37:59 +02:00
parent 106103b9ad
commit 4f1e3228b1
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/data/scripts/pipeline.sh
View File

@ -1,10 +1,10 @@
#!/bin/bash
cd ../processing
jq -r '. | select((.pos=="noun") or (.pos=="verb") or (.pos=="adj") or (.pos=="adv")) | select((.word | test("[^a-z]"))|not) | {word: .word, form_of: .senses[].tags | any(. == "form-of"), type: .pos, pronunciation: .sounds[0].ipa, definitions: (try .senses|map(.glosses|join(" ")))}' \
jq -r '. | select((.pos=="noun") or (.pos=="verb") or (.pos=="adj") or (.pos=="adv")) | select((.word | test("[^a-z]"))|not) | {word: .word, type: .pos, pronunciation: .sounds[0].ipa, definitions: (try .senses|map(.glosses|join(" ")))}' \
wiktionary.json > wiktionary-p1.json
jq --slurp '. | group_by(.word)[] | {word:.[0].word, pronunciation:.[0].pronunciation, meanings:[.[]|{type:.type, definitions:[try .definitions[]]| select(.!=[]) |map({(.):1})|add|keys_unsorted}]}' \
jq --slurp '. | group_by(.word)[] | {word:.[0].word, pronunciation:.[0].pronunciation, meanings:[.[]|{type:.type, form_of:(try .form_of), definitions:[try .definitions[]]| select(.!=[]) |map({(.):1})|add|keys_unsorted}]}' \
wiktionary-p1.json > wiktionary-p2.json
jq --slurp '.' wiktionary-p2.json > wiktionary-p3.json