Skip Menu |

This queue is for tickets about the Image-Info CPAN distribution.

Report information
The Basics
Id: 130629
Status: resolved
Priority: 0/
Queue: Image-Info

People
Owner: Nobody in particular
Requestors: nick [...] ccl4.org
Cc:
AdminCc:

Bug Information
Severity: Wishlist
Broken in: (no value)
Fixed in: (no value)



Please find attached a patch to add WebP support for Image::Info We need this at work, so it's been done on work time, and I have permission to upload and submit it to you. [it's not much use to us if I don't do this :-)] It's a "clean" implementation from the specs (aargh) - I didn't copy the other code on CPAN or elsewhere. Some things I'm not quite sure about - could you check these: 1) I had to add a module file, and as all the others stated copyright, I figured this one should too, so I put the legal name of geizhals in there. 2) I used done_testing() in the test file, but I'm not sure if that creates an implied dependency on a version of Test::More that isn't the default in core as far back as the module supports. 3) I tried to ensure that it was portable (pack formats) but I'm not in a position to test it on a big endian system (and I slack, I didn't test it on a 32 bit little endian system). 4) I didn't update CHANGES, but I hope that I did correctly update all the other "meta" files. I *did* test it on x86_64 against all the images in https://chromium.googlesource.com/webm/libwebp-test-data Imager::File::WEBP and the test images I added here - for all, it gives the same results as Image::ExifTool and Google's own command line tools. The test images are copies of test.jpg and test.png, or an "animation" made from 1x1 frames. Nothing new here.
Subject: 0001-WebP-support-for-Image-Info-Lossy-Lossless-and-Exten.patch

Message body is not shown because it is too large.

On 2019-10-02 12:19:13, NWCLARK wrote: Show quoted text
> Please find attached a patch to add WebP support for Image::Info > > We need this at work, so it's been done on work time, and I have > permission to upload and submit it to you. [it's not much use to us if > I don't do this :-)] > > It's a "clean" implementation from the specs (aargh) - I didn't copy > the other code on CPAN or elsewhere. > > Some things I'm not quite sure about - could you check these: > > 1) I had to add a module file, and as all the others stated copyright, > I figured this one should too, so I put the legal name of geizhals in > there. > 2) I used done_testing() in the test file, but I'm not sure if that > creates an implied dependency on a version of Test::More that isn't > the default in core as far back as the module supports. > 3) I tried to ensure that it was portable (pack formats) but I'm not > in a position to test it on a big endian system (and I slack, I didn't > test it on a 32 bit little endian system). > 4) I didn't update CHANGES, but I hope that I did correctly update all > the other "meta" files. > > I *did* test it on x86_64 against all the images in > https://chromium.googlesource.com/webm/libwebp-test-data > Imager::File::WEBP and the test images I added here - for all, it > gives the same results as Image::ExifTool and Google's own command > line tools. > > The test images are copies of test.jpg and test.png, or an "animation" > made from 1x1 frames. Nothing new here.
Thanks. travis looks good so far: https://travis-ci.org/eserte/image-info/builds/592728327 But this is not a surprise --- no big endian or 32 bit systems involved here. At least it seems that there are a few cpantesters with 32 bit systems around... so we'll see the results after the first dev release. Regarding the other notes: I am fine with it. Just the done_testing will be converted into a plan or no_plan, no big deal. I think I can do a proper release in about two weeks. Regards, Slaven
On 2019-10-02 12:19:13, NWCLARK wrote: Show quoted text
> 3) I tried to ensure that it was portable (pack formats) but I'm not > in a position to test it on a big endian system (and I slack, I didn't > test it on a 32 bit little endian system).
32 bit seems to be OK: http://www.cpantesters.org/cpan/report/96a129a2-f11a-11e9-a546-542be505d1f1
On 2019-10-19 02:29:37, SREZIC wrote: Show quoted text
> On 2019-10-02 12:19:13, NWCLARK wrote: >
> > 3) I tried to ensure that it was portable (pack formats) but I'm not > > in a position to test it on a big endian system (and I slack, I > > didn't > > test it on a 32 bit little endian system).
> > 32 bit seems to be OK: > http://www.cpantesters.org/cpan/report/96a129a2-f11a-11e9-a546- > 542be505d1f1
1.42 is out. Regards, Slaven