alignscript

#!/bin/bash

rm wikidata-names.tsv

curl "https://tools.wmflabs.org/mix-n-match/api.php?query=download2&catalogs=1793&columns=%7B%22exturl%22%3A0%2C%22username%22%3A0%2C%22aux%22%3A0%2C%22dates%22%3A0%2C%22location%22%3A0%2C%22multimatch%22%3A0%7D&hidden=%7B%22any_matched%22%3A0%2C%22firmly_matched%22%3A0%2C%22user_matched%22%3A0%2C%22unmatched%22%3A1%2C%22automatched%22%3A0%2C%22name_date_matched%22%3A0%2C%22aux_matched%22%3A0%2C%22no_multiple%22%3A0%7D&format=tab&as_file=1" > mixandmatch.tsv

for i in `seq 1 3641` ; do

echo -e `grep -P "^$i\t" names-numbers.tsv | cut -f 5`"\t"`grep -P "\t$i\tQ" mixandmatch.tsv | cut -f 4` >> wikidata-names.tsv

# this is the full-name string and then Qid if known

done

####

# 

rm working/all-data.tsv

for j in `seq 0 12253` ; do

echo -e `grep -P "^$j\t" names-postings.tsv | cut -f 4` > working/slug

echo -e  `grep -P "^$j\t" names-postings.tsv | cut -f 1`"\t"`grep -P "^$j\t" names-postings.tsv | cut -f 4`"\t"`grep -P "^$j\t" names-postings.tsv | cut -f 2`"\t"`grep -P "^$j\t" names-postings.tsv | cut -f 3`"\t"`grep -P "^$j\t" names-postings.tsv | cut -f 5`"\t"` grep -f working/slug wikidata-names.tsv | cut -f 2` >> working/all-data.tsv

#nb number at front is for alignment

done

cp working/all-data.tsv all-data.tsv

grep -P "\tQ\d" all-data.tsv > matched.tsv
# only those matched to Wikidata

cut -f 1 matched.tsv > working/matchlist
grep -vxFf uploadedids working/matchlist > working/to-upload

# uploadedids has a tracker for all those I've put up, taken from the spreadsheet
# this now has the line IDs for everything that can be uploaded

rm matched-to-upload.tsv

for i in `cat working/to-upload` ; do grep -P "^$i\t" matched.tsv >> matched-to-upload.tsv ; done

# this finds everything which is matched and ready to upload

exit