Created attachment 107630 [details] uk-UA fixed queries for default authority Upon fresh install when I choose to add authorities, the following error shows near every authorities file install.pl: DBD::mysql::st execute failed: Incorrect integer value: '' for column `koha_sql`.`auth_tag_structure`.`repeatable` at row 1 After completing install, I tried executing those SQL commands in phpmyadmin and it seems that in mandatory and repeatable columns have values '' or 1. I tried replacing those empty strings with zero and it worked. Seems like each SQL query in those files should be checked and columns mandatory, repeatable should only have values 0 or 1 as it is type tinyint(4).
Created attachment 107637 [details] [review] Patch for uk-UA fix of authority SQL queries Could not upload unzipped, got 413 nginx error for some reason
Created attachment 107638 [details] Patch for uk-UA fix of authority SQL queries Could not upload as patch, got 413 error
I have uploaded a signed off patch to a branch here: https://github.com/mrenvoize/Koha/tree/bug_26112 I'm afraid I don't speak the language, but the files apply and run so I'm happy to signoff. Thankyou for the contribution.
Replacing Martin's with my local branch - includes a tiny follow-up :) https://gitlab.com/kfischer1/Koha/-/tree/bug_26112
I have tried a new install using uk-UA but I am getting tons of errors, attaching the output logs. Can you have a look Bohdan?
Created attachment 111377 [details] installer_uk-UA_output.txt
Created attachment 112637 [details] [review] Fix frameworks inserts
Created attachment 112638 [details] [review] Remove duplicate values
Bohdan, I don't manage to apply your patch. In the first zip file, there are some files that don't exist in our code base, like installer/data/mysql/uk-UA/marcflavour/marc21/optional/marc21_bibliographic_BKS_local.sql Can you please provide a remote branch that I could check out locally?
(In reply to Jonathan Druart from comment #9) > Bohdan, I don't manage to apply your patch. > In the first zip file, there are some files that don't exist in our code > base, like > installer/data/mysql/uk-UA/marcflavour/marc21/optional/ > marc21_bibliographic_BKS_local.sql > > Can you please provide a remote branch that I could check out locally? Wait, that's not true. This file exists. I am retrying.
The second patch does not apply, maybe because of the [] in the patch? I have no idea why but I've started to apply the changes manually and found: 149 ('LANG', 'ang', 'давньоанглійська (староанглійська) (450-1100)'), 150 - ('LANG', 'ang', 'англо-саксонська'), 151 - ('LANG', 'ang', 'західносаксонсь'), I am not sure we want to keep the first one. I haven't check the others, but I don't think it makes sense to keep only the first one. Can you please provide some changes to make things consistent in the second patch and attach them on this bug report on a git format-patch format, or all the patches on a remote branch.
(In reply to Jonathan Druart from comment #11) > The second patch does not apply, maybe because of the [] in the patch? > I have no idea why but I've started to apply the changes manually and found: > > > 149 ('LANG', 'ang', 'давньоанглійська (староанглійська) (450-1100)'), > 150 - ('LANG', 'ang', 'англо-саксонська'), > 151 - ('LANG', 'ang', 'західносаксонсь'), > > I am not sure we want to keep the first one. > I haven't check the others, but I don't think it makes sense to keep only > the first one. > > Can you please provide some changes to make things consistent in the second > patch and attach them on this bug report on a git format-patch format, or > all the patches on a remote branch. From what I understand, only first value in authorised_values is shown in dropdown when cataloguing. Therefore I removed all duplicates and left those values that suit better for those codes.
Using google translate: давньоанглійська (староанглійська) (450-1100) => Old English (Old English) (450-1100) англо-саксонська => Anglo-Saxon західносаксонсь => West Saxons Why did you keep the first one? I was trying to understand if you kept the most relevant one or just the first one without judgement.
(In reply to Jonathan Druart from comment #13) > Using google translate: > давньоанглійська (староанглійська) (450-1100) => Old English (Old English) > (450-1100) > англо-саксонська => Anglo-Saxon > західносаксонсь => West Saxons > > Why did you keep the first one? I was trying to understand if you kept the > most relevant one or just the first one without judgement. In this case it was just the first one.
Could you rework the last patch (to keep the most relevant entries) and make the different patches available on a remote branch?
This is invalid now that we have removed the language specific installer files for yaml/po files.