peczenyj / GDPR-IAB-TCFv2 / 25463002434
96%
main: 96%

LAST BUILD BRANCH: devel
DEFAULT BRANCH: main
Ran
Jobs 1
Files 11
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 95.613%. Remained the same
25463002434

push

github

web-flow 
chore: gzip the golden test corpus to shrink on-disk footprint (#53)

t/corpus/golden.jsonl was 32 MB of one-JSON-object-per-line text,
which made `git clone` / dev-checkout footprint ugly (the compressed
.gz is 1.8 MB — a 16x ratio on disk).

Effects:

  - On-disk corpus: 32 MB -> 1.8 MB
  - CPAN dist tarball: ~2.0 MB -> ~1.6 MB (~21% smaller; the dist
    is itself gzipped, so the corpus-internal compression mostly
    overlaps with what `make dist` was already doing — modest but
    real savings on top)
  - Repo clone size: marginal additional gain on top of git's own
    pack compression

Implementation:

  - t/corpus/golden.jsonl -> t/corpus/golden.jsonl.gz
  - t/07-golden.t reads via IO::Uncompress::Gunzip (core in 5.10+,
    declared in TEST_REQUIRES so 5.8 still works after a CPAN
    install of IO-Compress)
  - t/generate_golden.pl writes via IO::Compress::Gzip Level=>9
  - MANIFEST swaps the entry; Makefile.PL declares both modules
    in TEST_REQUIRES (no MIN_PERL_VERSION change — stays at 5.008
    so the recently-validated Perl 5.8.9 CPAN Tester path still
    runs the full suite)

Trade-off worth knowing:

  - Diffs after `REGEN_CORPUS=1 prove -lv t/07-golden.t` no longer
    render in GitHub's PR view (it shows "Binary file differs").
    A comment in t/07-golden.t points reviewers at the local
    inspection recipe:
      gunzip -c t/corpus/golden.jsonl.gz \
        | diff - <(git show HEAD:t/corpus/golden.jsonl.gz | gunzip -c)

741 of 775 relevant lines covered (95.61%)

11624.71 hits per line

Jobs
ID Job ID Ran Files Coverage
1 25463002434.1 11
95.61
 GitHub Action Run
Source Files on build 25463002434