[Bug 41608] New: Ability to select multiple values of a facet in the professional interface
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Bug ID: 41608 Summary: Ability to select multiple values of a facet in the professional interface Initiative type: --- Sponsorship --- status: Product: Koha Version: Main Hardware: All OS: All Status: NEW Severity: enhancement Priority: P5 - low Component: Searching Assignee: koha-bugs@lists.koha-community.org Reporter: mathsabypro@gmail.com QA Contact: testopia@bugs.koha-community.org In the professional interface, we should be able to select several values from a facet in a single operation (e.g., several libraries, several languages, etc.), so that Koha applies the "OR" operator between each value (e.g., return documents present in library A or library B) -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Mathieu Saby <mathsabypro@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Comma delimited| |Kohala Association list of Sponsors| | Sponsorship status|--- |Sponsored -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #1 from Mathieu Saby <mathsabypro@gmail.com> --- Kohala association is sponsoring this enhancement. -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Arthur Suzuki <arthur.suzuki@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |ASSIGNED -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #2 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 201434 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=201434&action=edit Bug 41608: add support for multiple facets Test plan: - start KTD - on staff interface, search for "complete" - you should get around 10 results - there are checkboxes in front of the facets, check "Austen, Jane" and "Beckett, Samuel" - click on "Refine Search" at the bottom of the facets menu - You should get a refined search with "complete" from both Jane Austen and Samuel Beckett. - If you check "Centerville" the CCL query should also show "holdingbranch:CPL". - If you uncheck "Centerville" the CCL should go back to "kw:complete and (au:Austen, Jane or au:Beckett, Samuel)" - repeat for OPAC. -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Michaela Sieber <michaela.sieber@kit.edu> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |clemens.tubach@kit.edu, | |michaela.sieber@kit.edu -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 David Cook <dcook@prosentient.com.au> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |dcook@prosentient.com.au --- Comment #3 from David Cook <dcook@prosentient.com.au> --- Hmm this is an interesting one. I can 100% understand using checkboxes instead of clicking the link/the little X. I think most modern facetted search engines update the interface immediately after a checkbox check/uncheck, but they tend to be extremely fast at returning results. That said, I had a case recently where my Internet was poor and it was so frustrating because the search results would take a long time to load on a retail site. It made the UI almost unusable... I think we'd want to system preference this one... but it's an interesting idea for sure. I think Koha might benefit from having to click a button to apply the facet filters rather than doing it immediately. It would improve the speed of getting the desired results for users (better UX) and it would mean fewer operations (less load on the server). I don't know about "Refine Search" as the button label though (also it would need to be "Refine search" with a lowercase S to follow Koha rules). I suppose it says "Refine your search" at the top of the facet section but it doesn't seem very modern. Honestly, I think we could probably change "Refine your search" to "Filters" and then have the button be "Apply" or "Apply filters". (Funny enough Ex Libris Primo has a title of "Refine my results" and has two buttons: "Clear" and "Apply Filters". After applying filters, they have "Active filters" shown at the top.) -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 David Cook <dcook@prosentient.com.au> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |Failed QA --- Comment #4 from David Cook <dcook@prosentient.com.au> --- But yeah since this is UX-based I think a system preference would be wise. -- Looking at the code... "koha-tmpl/intranet-tmpl/prog/js/facets.js" and "koha-tmpl/opac-tmpl/bootstrap/js/facets.js" are basically copy/pasted... which I suppose is difficult to avoid without a "shared-tmpl" directory. Marking "Failed QA" though because I think you're building the query wrong. If you look at how the facets currently work, you'll notice they're added as arguments to "limit" parameters. Guessing AI might've proposed the rewriting of the search query? -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Arthur Suzuki <arthur.suzuki@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |arthur.suzuki@biblibre.com --- Comment #5 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Hi David, The purpose to having to click on the button instead of having the search results computed on the fly is actually what we want. So is it for the resulting search query with AND / OR. The problem with the current implementation is : It is impossible to select multiple authors and have results for both. That's why I want to be able to select multiple authors and only then click on the "Refine search". With the current behavior, once I've selected one author, I cannot select another to have results for both in the same list. Hope this gives you a better understanding of the code. Best, Arthur -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #6 from David Cook <dcook@prosentient.com.au> --- (In reply to Arthur Suzuki from comment #5)
Hi David, The purpose to having to click on the button instead of having the search results computed on the fly is actually what we want.
Yep, I understand that.
So is it for the resulting search query with AND / OR. The problem with the current implementation is : It is impossible to select multiple authors and have results for both. That's why I want to be able to select multiple authors and only then click on the "Refine search".
With the current behavior, once I've selected one author, I cannot select another to have results for both in the same list.
Thanks for that explanation. That's an interesting point. I think that should probably be a separate bug report to be honest as that's a separate (yet related) issue. -- It's interesting because... I might want to choose several different "Item types" for my results. Maybe I want both "book" and "ebook". That's something that you'd expect to see on a retail site. But then with "Authors" maybe I'm trying to refine my search to find particular authors that co-wrote resources together, so doing the "AND" rather than the "OR" might make sense... but then if you were trying to do that it would be more appropriate to use the Advanced Search rather than the facets/filters. So yeah I think maybe it's a bug. It probably makes sense to AND different types of facets/filters, but when the type is the same it should probably be ORed. I suspect that people will be upset either way, so it might be necessary to have a system preference for it... -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #7 from David Cook <dcook@prosentient.com.au> --- Consider in KTD the following: /cgi-bin/koha/opac-search.pl?idx=kw&q=e&limit=itype%3ABK&limit=itype%3AVM&sort_by=relevance&count=20 This only shows records that have both "Books" and "Visual Materials" indexed item types. That's probably not what we want most of the time. /cgi-bin/koha/opac-search.pl?advsearch=1&q=e&weight_search_submitted=1&weight_search=1&do=Search&limit=mc-itype%2Cphr%3ABK&limit=mc-itype%2Cphr%3AVM&sort_by=relevance This works more like what we'd expect... is that what the "mc-" prefix is all about? Multiple? Something to look at... -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #8 from Mathieu Saby <mathsabypro@gmail.com> --- Here are some remarks: 1/ In a sandbox I tried to search "the*" (229 results) then filter the results with 1 or more values in a facet Oddly it display an error message : No results match your search for 'kw,wrdl: "ccl=(kw:the*) AND (ccode:FIC)" Same issue with "book" and 2 values: No results match your search for 'kw,wrdl: "ccl=(kw:book) AND ((au:Alatriste, Gustavo) OR (au:Brook, Claudio))"'. But the search for "complete" is working well, I don't understand how... 2/ the "Refine" button should be always visible. It is currently at the bottom of the facets, on my little laptop it is not visible, I need to scroll to find it. It should be always visible. Maybe on top of the facets block, and sticky? 3/ The ideal would be to find a user-friendly way for the user to choose whether to combine values using “OR” or “AND”. Is your patch working that way? I am not sure of that, due to the odd results for my queries. In Primo, for example you can - click on the label of 1 value (ex: Library A), get the filtered results, then click on a 2d value of the same facet (ex : Library B), get the filtered-filtered results => in that case the system combine the facets with a "AND" - check the box on the left of 2 values of the same facets, then click on "Refine" button. The system combine the facets with a "OR" -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #9 from Mathieu Saby <mathsabypro@gmail.com> --- The sandbox I used to test : https://staff-bz41608.sandboxes.ptfs-europe.co.uk/cgi-bin/koha/mainpage.pl -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #10 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Maybe check the values of those syspref? QueryAutoTruncate QueryFuzzy QueryStemming(In reply to Mathieu Saby from comment #8)
Here are some remarks:
1/ In a sandbox I tried to search "the*" (229 results) then filter the results with 1 or more values in a facet
Oddly it display an error message : No results match your search for 'kw,wrdl: "ccl=(kw:the*) AND (ccode:FIC)"
Same issue with "book" and 2 values: No results match your search for 'kw,wrdl: "ccl=(kw:book) AND ((au:Alatriste, Gustavo) OR (au:Brook, Claudio))"'.
But the search for "complete" is working well, I don't understand how...
2/ the "Refine" button should be always visible. It is currently at the bottom of the facets, on my little laptop it is not visible, I need to scroll to find it. It should be always visible. Maybe on top of the facets block, and sticky?
3/ The ideal would be to find a user-friendly way for the user to choose whether to combine values using “OR” or “AND”. Is your patch working that way? I am not sure of that, due to the odd results for my queries.
In Primo, for example you can - click on the label of 1 value (ex: Library A), get the filtered results, then click on a 2d value of the same facet (ex : Library B), get the filtered-filtered results => in that case the system combine the facets with a "AND" - check the box on the left of 2 values of the same facets, then click on "Refine" button. The system combine the facets with a "OR"
Check some syspref related to search behaviour, maybe try to disable QueryFuzzy, QueryAutotruncate and QueryStemming? -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #11 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- (In reply to David Cook from comment #7)
Consider in KTD the following:
/cgi-bin/koha/opac-search. pl?idx=kw&q=e&limit=itype%3ABK&limit=itype%3AVM&sort_by=relevance&count=20
This only shows records that have both "Books" and "Visual Materials" indexed item types. That's probably not what we want most of the time.
/cgi-bin/koha/opac-search. pl?advsearch=1&q=e&weight_search_submitted=1&weight_search=1&do=Search&limit= mc-itype%2Cphr%3ABK&limit=mc-itype%2Cphr%3AVM&sort_by=relevance
This works more like what we'd expect... is that what the "mc-" prefix is all about? Multiple? Something to look at...
I should try this, thanks for the hint! -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #12 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- (In reply to David Cook from comment #6)
It's interesting because... I might want to choose several different "Item types" for my results. Maybe I want both "book" and "ebook". That's something that you'd expect to see on a retail site.
Yes
But then with "Authors" maybe I'm trying to refine my search to find particular authors that co-wrote resources together, so doing the "AND" rather than the "OR" might make sense... but then if you were trying to do that it would be more appropriate to use the Advanced Search rather than the facets/filters.
You could already do that with the current implementation of the facets by refining search I think! (click on the two authors successively?)
So yeah I think maybe it's a bug. It probably makes sense to AND different types of facets/filters, but when the type is the same it should probably be ORed.
That is what we want to propose with this bug :) -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #13 from Mathieu Saby <mathsabypro@gmail.com> --- (In reply to Arthur Suzuki from comment #10)
Maybe check the values of those syspref? QueryAutoTruncate QueryFuzzy QueryStemming(In reply to Mathieu Saby from comment #8)
Here are some remarks:
1/ In a sandbox I tried to search "the*" (229 results) then filter the results with 1 or more values in a facet
Oddly it display an error message : No results match your search for 'kw,wrdl: "ccl=(kw:the*) AND (ccode:FIC)"
Same issue with "book" and 2 values: No results match your search for 'kw,wrdl: "ccl=(kw:book) AND ((au:Alatriste, Gustavo) OR (au:Brook, Claudio))"'.
But the search for "complete" is working well, I don't understand how...
The sandbox uses Zebra (it cannot use ES), with MARC21 records The syspref have their default values QueryAutoTruncate : automatically QueryFuzzy : try QueryStemming : try I change them to : QueryAutoTruncate : only if * is added QueryFuzzy : dont try QueryStemming : dont try It changes nothing, the search gives no result after a filtering In the search bar I can read : "ccl=(kw:the*) AND ((holdingbranch:CPL) OR (holdingbranch:FPL))" -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #14 from Mathieu Saby <mathsabypro@gmail.com> --- But it is working without the * the query in the search bar is : ccl=(kw:the) AND ((holdingbranch:CPL) OR (holdingbranch:FPL)) -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #15 from Mathieu Saby <mathsabypro@gmail.com> --- Oh, my 2d example "book" + 2 author facets is now working. The search fails if QueryAutoTruncate is activated, but gives results otherwise. And it fails if an explicit truncature is added at the end of the searched word. So I suppose you need to fix something with the handling of truncation. -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #16 from David Cook <dcook@prosentient.com.au> --- (In reply to Arthur Suzuki from comment #12)
So yeah I think maybe it's a bug. It probably makes sense to AND different types of facets/filters, but when the type is the same it should probably be ORed.
That is what we want to propose with this bug :)
Ok but the way to do that would be in how the "limit" parameter is processed rather than changing the CCL query. -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #17 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- (In reply to David Cook from comment #16)
(In reply to Arthur Suzuki from comment #12)
So yeah I think maybe it's a bug. It probably makes sense to AND different types of facets/filters, but when the type is the same it should probably be ORed.
That is what we want to propose with this bug :)
Ok but the way to do that would be in how the "limit" parameter is processed rather than changing the CCL query.
yep, I just got the implementation working, it is indeed much simpler and the compatibility with the current behaviour of searching is better preserved (QueryFuzzy, Truncate, etc...) I will do some more checks and post a bit later. Thx for the hints! -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #18 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 202617 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=202617&action=edit Bug 41608: (QA followup) change the JS to build a request with limit params instead of ccl query Patch from commit 78945fc -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Arthur Suzuki <arthur.suzuki@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #202617|0 |1 is obsolete| | -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #19 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 202618 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=202618&action=edit Bug 41608: (QA followup) change the JS to build a request with limit params instead of ccl query Patch from commit 78945fc -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #20 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 202619 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=202619&action=edit Bug 41608: (QA followup) Change labels for facets form title and button Patch from commit 39a3d58 -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #21 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 202620 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=202620&action=edit Bug 41608: (QA followup) checkboxes are only show if SearchEngine is Elasticsearch Sponsored-by: Kohala Association -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #22 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Hello, I've made the desired changes, also the checkboxes now only appears if the SearchEngine is ElasticSearch (zebra doesn't allow the mc-* indexes). I didn't add another syspref on top of that yet, wanted to make sure the rest is fine before. @Mathieu, we don't plan on making it possible to make complex queries. I'd say if you want to make more complex things, the way to go would be to either use the advanced search or go with CCL. Best, Arthur -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Arthur Suzuki <arthur.suzuki@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|Failed QA |Needs Signoff -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #23 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 202660 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=202660&action=edit Bug 41608: (QA followup) Add the syspref "MultiChoiceFacets" to enable facets checkboxes in the search results Patch from commit 3e75e1a -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Arthur Suzuki <arthur.suzuki@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #202660|0 |1 is obsolete| | -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #24 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 202661 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=202661&action=edit Bug 41608: (QA followup) Add a new syspref MultiChoiceFacets Patch from commit 938ef83 -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Laurence Rault <laurence.rault@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|Needs Signoff |Failed QA CC| |laurence.rault@biblibre.com --- Comment #25 from Laurence Rault <laurence.rault@biblibre.com> --- On ktd Patch application failed for attachment 202661 - Bug 41608: (QA followup) Add a new syspref MultiChoiceFacets Error: Apply failed: Git command (git am -3 /tmp/8KsH0sqt4l/0005-202661.patch) failed: error: sha1 information is lacking or useless (installer/data/mysql/atomicupdate/bug41608_multichoicefacets.pl). error: could not build fake ancestor -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Laurence Rault <laurence.rault@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|Failed QA |Patch doesn't apply -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Arthur Suzuki <arthur.suzuki@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #202661|0 |1 is obsolete| | -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #26 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 202662 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=202662&action=edit Bug 41608: (QA followup) Add the syspref "MultiChoiceFacets" to enable facets checkboxes in the search results Sponsored-by: Kohala Association -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Arthur Suzuki <arthur.suzuki@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |m.de.rooy@rijksmuseum.nl Component|Searching |Z39.50 / SRU / OpenSearch | |Servers Status|Patch doesn't apply |Needs Signoff --- Comment #27 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- there was an issue with the last patch, I've re-uploaded it, it should apply now. -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Laurence Rault <laurence.rault@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|Needs Signoff |Failed QA --- Comment #28 from Laurence Rault <laurence.rault@biblibre.com> --- Patch applies First comments : 1) when the syspref MultiChoiceFacets is not active, the button Apply is still visible 2) when the syspref is active, the checkboxes are in the page, but not visible, so it is not easy for testing... input[type=checkbox], input[type=radio] { height: unset; margin: 3px 3p If I remove height: unset;, they are visible and usable, and filtering seems to be ok. But I prefer this to be solved before going on with the search tests. -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #29 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 202663 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=202663&action=edit Bug 41608: if MultiChoiceFacets is disabled, do not show 'Apply' button Patch from commit a0af240 -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #30 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- (In reply to Laurence Rault from comment #28)
Patch applies
First comments :
1) when the syspref MultiChoiceFacets is not active, the button Apply is still visible
Just solved that one in a new patch.
2) when the syspref is active, the checkboxes are in the page, but not visible, so it is not easy for testing...
input[type=checkbox], input[type=radio] { height: unset; margin: 3px 3p
If I remove height: unset;, they are visible and usable, and filtering seems to be ok.
But I prefer this to be solved before going on with the search tests.
Huum, I cannot reproduce this issue... Once you've started ktd can you try to recompile the CSS? with the command "yarn build"? You may have an older version of the CSS loaded. -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #31 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 202665 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=202665&action=edit Bug 41608: (QA followup) do not check removed facets upon page load Patch from commit 2ed711b -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Arthur Suzuki <arthur.suzuki@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|Failed QA |Needs Signoff -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Arthur Suzuki <arthur.suzuki@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Assignee|koha-bugs@lists.koha-commun |arthur.suzuki@biblibre.com |ity.org | Component|Z39.50 / SRU / OpenSearch |Searching - Elasticsearch |Servers | -- You are receiving this mail because: You are watching all bug changes. You are the assignee for the bug.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Laurence Rault <laurence.rault@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|Needs Signoff |Patch doesn't apply --- Comment #32 from Laurence Rault <laurence.rault@biblibre.com> --- Patch application failed for attachment 201434 - Bug 41608: add support for multiple facets Application tested on ktd and on sandbox -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #33 from Mathieu Saby <mathsabypro@gmail.com> --- If you can rebase it, I am available for testing it again this week. -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Arthur Suzuki <arthur.suzuki@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|Patch doesn't apply |Needs Signoff -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Arthur Suzuki <arthur.suzuki@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #201434|0 |1 is obsolete| | Attachment #202618|0 |1 is obsolete| | Attachment #202619|0 |1 is obsolete| | Attachment #202620|0 |1 is obsolete| | Attachment #202662|0 |1 is obsolete| | Attachment #202663|0 |1 is obsolete| | Attachment #202665|0 |1 is obsolete| | -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #34 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 203724 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=203724&action=edit Bug 41608: add support for multiple facets Test plan: - start KTD - on staff interface, search for "complete" - you should get around 10 results - there are checkboxes in front of the facets, check "Austen, Jane" and "Beckett, Samuel" - click on "Refine Search" at the bottom of the facets menu - You should get a refined search with "complete" from both Jane Austen and Samuel Beckett. - If you check "Centerville" the CCL query should also show "holdingbranch:CPL". - If you uncheck "Centerville" the CCL should go back to "kw:complete and (au:Austen, Jane or au:Beckett, Samuel)" - repeat for OPAC. -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #35 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 203725 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=203725&action=edit Bug 41608: (QA followup) change the JS to build a request with limit params instead of ccl query -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #36 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 203726 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=203726&action=edit Bug 41608: (QA followup) Change labels for facets form title and button -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #37 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 203727 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=203727&action=edit Bug 41608: (QA followup) checkboxes are only show if SearchEngine is Elasticsearch Sponsored-by: Kohala Association -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #38 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 203728 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=203728&action=edit Bug 41608: (QA followup) Add the syspref "MultiChoiceFacets" to enable facets checkboxes in the search results Sponsored-by: Kohala Association -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #39 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 203729 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=203729&action=edit Bug 41608: if MultiChoiceFacets is disabled, do not show 'Apply' button -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #40 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 203730 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=203730&action=edit Bug 41608: (QA followup) do not check removed facets upon page load -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #41 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- (In reply to Mathieu Saby from comment #33)
If you can rebase it, I am available for testing it again this week.
Done, can you try on a sandbox? you'll need an ElasticSearch sandbox for that to work, it doesn't work with Zebra -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #42 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 203736 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=203736&action=edit Bug 41608: (QA followup) make the multifacet feature available for Zebra as well -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #43 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- (In reply to Arthur Suzuki from comment #41)
(In reply to Mathieu Saby from comment #33)
If you can rebase it, I am available for testing it again this week.
Done, can you try on a sandbox? you'll need an ElasticSearch sandbox for that to work, it doesn't work with Zebra
Now works on both (since last patch uploaded) -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Mathieu Saby <mathsabypro@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|Needs Signoff |Failed QA --- Comment #44 from Mathieu Saby <mathsabypro@gmail.com> --- I tested with KTD, with ES and Zebra, and noted those issues : 1/ The search behavior is broken with Zebra if you combine "multi-facets" with "classic" facets - search "the" - check both boxes "Fiction" and "Non-Fiction" in Collections facets - Apply => 23 results - click on "Centerville" Library => 0 results It seems the "mc-ccode" is wrongly changed into a simple "ccode". It should not, and this is not the case with Elasticsearch 2/ You should rephrase the syspref (remove the "when using Elasticsearch"), as it applies also if Zebra is used 3/ The Apply button is not very visible, especially on small screens, or if you have a lot of facets. I have seen that in the Bokeh discovery tool this button is show twice, at top and at bottom of the facets. What do you think of this solution? -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Arthur Suzuki <arthur.suzuki@biblibre.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Patch complexity|--- |Medium patch Status|Failed QA |Needs Signoff -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #45 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 204072 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=204072&action=edit Bug 41608: (QA followup) properly store multichoices limits in the cgi query -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #46 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 204073 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=204073&action=edit Bug 41608: (QA followup) add the facets submit button at the top of the form -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #47 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 204074 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=204074&action=edit Bug 41608: (QA followup) fix the JS to properly escape doublequote in authorized value -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #48 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- Created attachment 204075 --> https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=204075&action=edit Bug 41608: (QA followup) change MultiChoiceFacet syspref description (not limited to ES) -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #49 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- (In reply to Mathieu Saby from comment #44)
I tested with KTD, with ES and Zebra, and noted those issues :
Hello Mathieu, I've uploaded a bunch (4) new patches to address all you comments. for all of them, including the last one, you're perfectly right. please re-apply and test again! Best, Arthur -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #50 from Mathieu Saby <mathsabypro@gmail.com> --- I believe the "MultiChoiceFacets" should be in "Results display" section (like other syspref related to facets) and not in "Search form". Do you agree? -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #51 from Mathieu Saby <mathsabypro@gmail.com> --- Sorry but with Zebra the checkboxes are not visible. In Chrome devtools I can see this CSS: "height: unset" -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 Mathieu Saby <mathsabypro@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|Needs Signoff |Failed QA -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #52 from Mathieu Saby <mathsabypro@gmail.com> --- I tried to relaunch my ktd and it seems to solve the CSS issue. But from my perspective, there's something inconsistent about the way to deselect a facet. Could you give me your opinion? In the old system, if a value is selected, it appears followed by an [x], with a tooltip that says "Remove ..." In the new system > if you select multiple values using the checkbox, there is no "[x]" > if you select a single value using the checkbox, there is no "[x]" > if you click on a value without using the checkbox (which, in terms of filtering, is functionally the same as checking the checkbox), it works like in the old system: the value is followed by an "[x]" followed by "Remove" - Then, if you refine the search by clicking on another value in a different facet, that new value will also be followed by "[x]" - Then, if you refine the search by selecting multiple values for another facet, all the "[x]" symbols disappear - if you play around with all of this by adding different layers of facets, you might end up with a pretty weird display. In short, this is inconsistent, and I think that if the preference is enabled, there should be no "[x]" or "Remove" labels at all. These labels make no or little sense in a UI where values are preceded by checkboxes, which allow the user to know which values are actually being used for filtering (unlike the legacy system). -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #53 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- (In reply to Mathieu Saby from comment #50)
I believe the "MultiChoiceFacets" should be in "Results display" section (like other syspref related to facets) and not in "Search form". Do you agree?
Ok, I'll change that -- You are receiving this mail because: You are watching all bug changes.
https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41608 --- Comment #54 from Arthur Suzuki <arthur.suzuki@biblibre.com> --- (In reply to Mathieu Saby from comment #52) > I tried to relaunch my ktd and it seems to solve the CSS issue. > > But from my perspective, there's something inconsistent about the way to > deselect a facet. Could you give me your opinion? > > In the old system, if a value is selected, it appears followed by an [x], > with a tooltip that says "Remove ..." > > In the new system > > if you select multiple values using the checkbox, there is no "[x]" > > if you select a single value using the checkbox, there is no "[x]" > > if you click on a value without using the checkbox (which, in terms of filtering, is functionally the same as checking the checkbox), it works like in the old system: the value is followed by an "[x]" followed by "Remove" > - Then, if you refine the search by clicking on another value in a different > facet, that new value will also be followed by "[x]" > - Then, if you refine the search by selecting multiple values for another > facet, all the "[x]" symbols disappear > - if you play around with all of this by adding different layers of facets, > you might end up with a pretty weird display. > > In short, this is inconsistent, and I think that if the preference is > enabled, there should be no "[x]" or "Remove" labels at all. These labels > make no or little sense in a UI where values are preceded by checkboxes, > which allow the user to know which values are actually being used for > filtering (unlike the legacy system). Actually it is complementary. - select several checkboxes, it will make an "OR" between same types/indexes. - click on several facets, it will make an "AND". - unchecking a checkbox remove the value from the OR. - click on the [x] removes the facet from the "AND". it is not just about UI. -- You are receiving this mail because: You are watching all bug changes.
participants (1)
-
bugzilla-daemon＠bugs.koha-community.org