Thesaurus label matching hotfix

huridocs / uwazi

Uwazi is a web-based, open-source solution for building and sharing document collections

http://www.uwazi.io

MIT License

225 stars 80 forks source link

Thesaurus label matching hotfix #6943

Closed LaszloKecskes closed 3 days ago

LaszloKecskes commented 3 days ago

fixes #6942

PR checklist:

[ ] Update READ.me ?
[ ] Update API documentation ?

QA checklist:

[ ] Smoke test the functionality described in the issue
[ ] Test for side effects
[ ] UI responsiveness
[ ] Cross browser testing
[ ] Code review

LaszloKecskes commented 3 days ago

The code looks good and it works. What I'm wondering is: creating thesaurus values through the thesaurus ui allows for labels that only differ in casing. e.g. "a" and "A" can co-exists. Is this "case insensitivity" feature desired in the csv import?

As far as I can remember the discussion, yes. We are explicitly testing for it, we've just missed this case on the csv side when added the option to import grouped values. So I've added some more thorough testing on the thesaurus side.

fnocetti commented 3 days ago

The code looks good and it works. What I'm wondering is: creating thesaurus values through the thesaurus ui allows for labels that only differ in casing. e.g. "a" and "A" can co-exists. Is this "case insensitivity" feature desired in the csv import?

As far as I can remember the discussion, yes. We are explicitly testing for it, we've just missed this case on the csv side when added the option to import grouped values. So I've added some more thorough testing on the thesaurus side.

Makes sense!