Sebastian Humenda: > Hi all > > I'd like to report and discuss with you an issue that I'm having with the > reproducible builds page, linked from the QA page of a package of mine. > My package, freedict, is affected by "unstable" builds. I suppose that's > because > the build time stamp is inserted into some of the distributed files. > Therefore I have tried to take a look at the diff output, but found this view > to be very inaccessible on the web site. I am using a screen reader (BRLTTY or > gnome-orca) to read either the console or the GUI, respectively. [...] > However, the text file contains tables made of Unicode symbols and hence I > cannot navigate them properly. The screen reader is reading every character on > the line, from left to right, ignoring the columns in the plain text format. > This usually works fine for plain texts or code, but in these > plain-text-tables, > I am unable to read the information properly. It basically gets read to me > without structure. [...] > 3. The diff (or build) output contains these characters mentioned above. It > would be great, if somebody could explain the sense to me. As far as I > understand the output, it is some slightly polished diff output. If so, > would it be possible to provide a plain diff output as well?
The text (and the html) output is generated by 'diffoscope'. citing https://diffoscope.org/: > diffoscope will try to get to the bottom of what makes files or > directories different. It will recursively unpack archives of many > kinds and transform various binary formats into more human readable > form to compare them. It can compare two tarballs, ISO images, or PDF > just as easily. In the text output it uses "unified" diff output to compare the text representation of different parts of the file. Those diffs are indented to reflect the nested structure of the files in question. To make the structure visually more easily recognizable additional to the indentation there a lines drawn with those unicode chars. As a temporary work-around you can try to remove the unicode stuff and replace them with simple spaces with a command like: sed -e ':a;s/^\(\s*\)[╵│├─┄]/\1 /;t a' -i freedict_2016.10.22-1.diffoscope.txt Although I'm not sure if the result is much easier for you to read. How do you deal with other text content where the indentation matters like for example python source code?
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Reproducible-builds mailing list Reproducible-builds@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/reproducible-builds