Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## v0.3 #110 +/- ##
=======================================
Coverage ? 67.58%
=======================================
Files ? 18
Lines ? 2147
Branches ? 0
=======================================
Hits ? 1451
Misses ? 696
Partials ? 0 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Parsing an international number under a mismatched reference country strips the wrong national prefix, so parse(Some(GB), "+390635511397") yields +39635511397 and is_valid reports false: the metadata came from the reference country rather than from the country code the number carries. The let chain is rewritten as a match guard, since v0.3 is edition 2021. (cherry picked from commit 56bd871)
db42be2 to
8d1d5cb
Compare
|
The failing
Fix is either raising |
|
@23Skidoo I'm not precious about being the 'authour' of commits etc buddy do whatever you need -- my only goal is better library... for whatever that means in this weird machine overlord era... |
I say move to a newer |
c006f1a widened the quick-xml cap to <=0.41, and quick-xml 0.41 requires rustc 1.79, so the 1.76.0 job fails at dependency build on every pull request against v0.3. 1.79.0 is the lowest rustc that satisfies the cap.
|
Not hard for me to give credit where it's due. Added a commit bumping |
2 participants
Backport of 56bd871 from
main. The let chain is rewritten as a match guard, sincev0.3is edition 2021.On
v0.3, parsing an international number under a mismatched reference country strips the wrong national prefix:parse(Some(country::GB), "+390635511397")returns+39635511397andis_validreportsfalse, because the metadata comes from the reference country rather than from the country code the number carries.mainfixed this in 56bd871; this brings the same fix to the 0.3 line.Verified on this branch: the full suite passes (98 tests), and the mismatched-hint cases now behave as they do on
main.A second commit raises
v0.3's MSRV to 1.79.0, which is the option @jer-gremlin picked in the thread below. c006f1a widened the quick-xml cap to<=0.41and quick-xml 0.41 requires rustc 1.79, so the1.76.0job fails at dependency build on every pull request againstv0.3. 1.79.0 is the lowest rustc that satisfies the cap, and it is declared inCargo.toml, the CI matrix and the README.Would you consider a
0.3.11with this in it? We're pinned to a fork solely for this fix. Moving tomainisn't an option for us yet because of the memory regression in #107, and you clearly maintain the 0.3 line — c006f1a backported the quick-xml cap here and shipped as0.3.10— so a 0.3.x release would let us drop the fork entirely.Prepared with LLM assistance (Claude Code); I see no policy either way in this repo. The code is @jer-gremlin's, kept as commit author — the only new lines are the match-guard rewrite. Tell me if you'd rather have a commit trailer recording that, or not take LLM-assisted PRs at all.