comments: cut the development log from four more normalize layers - #685
Merged
Conversation
Comments only — all four files verified code-identical to main by
transpiling both revisions with removeComments and comparing the emit.
serbian 316 → 190 zulu 314 → 171
persian 293 → 157 english 293 → 220
What went, per CONTRIBUTING: corpus tabulations and defect inventories,
decode transcripts and the arguments about how a word was sourced,
chronology ("used to", "the first version", "an earlier draft"), and the
inline trap-number glosses.
What stayed, marked with ⚠ where it bites: the ordering couplings between
rules, the guards and what each one rejects, deliberate omissions so they
are not "fixed", and the per-language invariants — Serbian's digraphia and
its lowercase-licensor guard, Zulu's rule that no rule may invent a concord
for a noun it did not emit, Persian's bare-Perso-Arabic emission constraint
and the RTL word-across-numeral move, English's lack of the shared tier.
English cuts least: much of its bulk is the unit/abbreviation tables and
the NOT_VERSION guard, which are specification rather than narrative.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012Lc3WnUgogC7okV7n53vjr
Self-review of the four rewrites. All three are cases where a paraphrase
invented or misattributed a reason; none touches code.
serbian the ± ordering note asserted a coupling to the degree rules'
spacing that does not exist — the real reason ± runs first is
that it is a single character the `+` arms cannot see
english said the ± arm requires the digit immediately; it does not,
it has `\s?`. The no-space requirement belongs to the sign
arm above it, which is what keeps `(1418 – 1450)` a range
persian listed the ordinal among the rules that move a word across
the numeral; it replaces digits with words and moves nothing
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012Lc3WnUgogC7okV7n53vjr
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fifth batch of the comment sweep. Comments only; all four files are code-identical to
main, verified by transpiling both revisions withremoveCommentsand comparing the emitted code.serbian/normalize.tszulu/normalize.tspersian/normalize.tsenglish/normalize.tsWhat went
Per
CONTRIBUTING.md: corpus tabulations and defect inventories, decode transcripts and the arguments about how a word was sourced, chronology ("used to", "the first version", "an earlier draft of this comment"), and the inline trap-number glosses left over from the issue-reference strip.What stayed
Marked with ⚠ where it bites — the ordering couplings between rules, the guards and what each rejects, deliberate omissions so they are not "fixed", and the per-language invariants:
NOT_VERSION,magnitudesandbareExponentall need local equivalents.English cuts least, and that is the honest result: much of its bulk is the unit and abbreviation tables plus the
NOT_VERSIONguard, which are specification rather than narrative.Checks
npm run typecheckcleanmain; no new debt added🤖 Generated with Claude Code
https://claude.ai/code/session_012Lc3WnUgogC7okV7n53vjr