Relative URLs trigger warnings. Test for them, instead of letting them loose on the screen.
Created attachment 38758 [details] [review] Bug 14117: Silence warnings t/SearchPazPar2.t The use of relative URLS (null actually), triggers warnings. Catch them, instead of letting them loose. TEST PLAN --------- 1) $ prove t/Search_PazPar2.t -- noise for each test currently. 2) apply patch 3) $ prove t/Search_PazPar2.t -- No noise, and extra tests added. -v shows caught warning tests. 4) koha qa test tools.
Created attachment 39016 [details] [review] Bug 14117: Silence warnings t/SearchPazPar2.t The use of relative URLS (null actually), triggers warnings. Catch them, instead of letting them loose. TEST PLAN --------- 1) $ prove t/Search_PazPar2.t -- noise for each test currently. 2) apply patch 3) $ prove t/Search_PazPar2.t -- No noise, and extra tests added. -v shows caught warning tests. 4) koha qa test tools. Signed-off-by: Indranil Das Gupta <indradg@gmail.com>
Does the warning is expected? The tests you have added don't make any sense to me.
Actually the whole file don't make any sense...
(In reply to Jonathan Druart from comment #4) > Actually the whole file don't make any sense... Prior to the patch?
yes
(In reply to Jonathan Druart from comment #6) > yes Because of yaz client? Or is there something else I'm missing?
(In reply to M. Tompsett from comment #7) > (In reply to Jonathan Druart from comment #6) > > yes > > Because of yaz client? Or is there something else I'm missing? Sorry I don't get it. I have just said that the file does not make sense, the tests are not really useful.
Created attachment 39693 [details] [review] [PASSED QA] Bug 14117: Silence warnings t/SearchPazPar2.t The use of relative URLS (null actually), triggers warnings. Catch them, instead of letting them loose. TEST PLAN --------- 1) $ prove t/Search_PazPar2.t -- noise for each test currently. 2) apply patch 3) $ prove t/Search_PazPar2.t -- No noise, and extra tests added. -v shows caught warning tests. 4) koha qa test tools. Signed-off-by: Indranil Das Gupta <indradg@gmail.com> Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>
(In reply to Jonathan Druart from comment #8) > [SNIP] the tests are not really useful. I have to agree, but this at least makes them quiet. Though, that does mean the issue of "Will tests ever be written that are meaningful for Search_ParPaz2?" is still worth addressing, but perhaps as a separate bug, since this is not about the quality of tests, but the noisiness of them.
Patch pushed to master. Thanks Mark!