peczenyj / lua-gdpr-iab-tcfv2 / 25672369211
90%

DEFAULT BRANCH: devel
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: 89.694% (-0.02%) from 89.711%
25672369211

push

github

web-flow 
test(parity): wire field-level corpus parity and upgrade claim wording (#20)

* test(parity): wire field-level corpus parity and upgrade claim wording

Earns the parity claim that Phase 2 softened: golden_spec.lua now
deep-compares parser:to_table() against Perl's tests.to_json output
for the rich tier of the Golden Corpus (the first 128 rows still
carry to_json after scripts/optimize_golden.sh's pass).

* test/reference/golden_spec.lua: new "Golden Field-Level Parity"
  describe. For each rich-tier row (up to TCF_DEEP_LIMIT, default 16),
  project the Lua parser output into Perl's snake_case + nested shape
  and run parity_helper.deep_compare. Verified against all 128 rich
  rows locally; default 16 keeps CI fast while still catching
  regressions on a meaningful sample.
* Comparison scope is intentional. Compared: 11 scalar metadata fields
  (cmp_id, cmp_version, ..., version) plus the bool maps
  (purpose.{consents,legitimate_interests}, vendor.{consents,
  legitimate_interests}, special_features_opt_in) with false entries
  stripped to match Perl's sparse JSON. Deferred (with reasons inline
  in the spec): created/last_updated (date format diverges -- Lua
  emits deciseconds-since-epoch, Perl emits ISO 8601), publisher.*
  (structure diverges -- Lua keeps publisher fields flat, Perl nests
  them), tc_string (always matches by construction).
* README.md, lua-gdpr-iab-tcfv2-0.1.0-1.rockspec, DEVELOPMENT.md:
  upgrade the parity wording from Phase 2's "validator-decision
  parity" to "field-level structural parity ... plus validator-
  decision parity across all 1,024 rows", accurately reflecting what
  test-reference now verifies.
* DEVELOPMENT.md: re-add the TCF_DEEP_LIMIT env-var row (removed in
  Phase 2 when there was no implementation backing it).
* test/fuzz/fuzz_spec.lua: seed math.random deterministically (42)
  so the "does not crash on random data" test is reproducible across
  CI runs. Override with TCF_FUZZ_SEED for ad-hoc exploration.
*... (continued)

557 of 621 relevant lines covered (89.69%)

10503.02 hits per line

Jobs
ID Job ID Ran Files Coverage
1 25672369211.1 11
89.69
 GitHub Action Run
Source Files on build 25672369211