X-Git-Url: http://git.vanrenterghem.biz/git.ikiwiki.info.git/blobdiff_plain/6d3ff9f6491f1e420b1cec2c4dec3b65cec024ed..fc7119505840b4156c5a13f1dab0cafbf3af2700:/doc/bugs/img_test_failure_on_nix.mdwn?ds=inline diff --git a/doc/bugs/img_test_failure_on_nix.mdwn b/doc/bugs/img_test_failure_on_nix.mdwn index e1321198c..911cc66a1 100644 --- a/doc/bugs/img_test_failure_on_nix.mdwn +++ b/doc/bugs/img_test_failure_on_nix.mdwn @@ -1,17 +1,33 @@ Peter Simons wrote: > I'm getting the following test suite failure in Nix, probably because > some Perl dependency was updated to a newer version recently: -> -> # Failed test at t/img.t line 79. -> # got: 'no image' -> # expected: '12x12' -> -> # Failed test at t/img.t line 80. -> # got: 'no image' -> # expected: '16x2' -> # Looks like you failed 2 tests of 19. -> t/img.t ........................ -> Dubious, test returned 2 (wstat 512, 0x200) -> Failed 2/19 subtests -> + + # Failed test at t/img.t line 79. + # got: 'no image' + # expected: '12x12' + + # Failed test at t/img.t line 80. + # got: 'no image' + # expected: '16x2' + # Looks like you failed 2 tests of 19. + t/img.t ........................ + Dubious, test returned 2 (wstat 512, 0x200) + Failed 2/19 subtests + > Do you have any idea how to fix this issue? + +>> As far as I can tell, this indicates that your ImageMagick doesn't +>> support PDF (in Debian and its derivatives this requires `libmagickcore-extra`). +>> Please run tests on a system with a fully-featured ImageMagick (in Debian we put +>> `libmagickcore-extra` in the `Build-Depends-Indep` for this), or if you are testing +>> at install time on arbitrary end-user systems, please tell me what this says on a +>> system where the test fails: +>> +>> perl -MData::Dumper -MImage::Magick -e 'print Dumper([(new Image::Magick)->QueryFormat("pdf")])' +>> +>> If the result is `$VAR1 = [ undef ]`, then we can easily make the test skip that bit if PDFs don't work. +>> --[[smcv]] + +>>> That configuration test doesn't actually work because of the way +>>> ImageMagick processes PDFs via ghostscript, but loading a sample +>>> PDF and probing its width does seem to work. [[done]] --[[smcv]]