Implement the lexer, AST and parser for the SQLite dialect - #3

Merged

kyleconroy merged 23 commits into

mainfrom

claude/the-loop-claude-md-4bnogc

Jul 30, 2026

Commits on Jul 30, 2026

Implement the lexer, AST and parser for the SQLite dialect
claude
committed
Record how far SQLite's parser got, so truncated parses do not lie
claude
committed
Extract the corpus from every pinned test script, and assert error offsets
claude
committed
Add the SQL renderer, AST dump, round-trip property and snapshots
claude
committed
Harden the parser: recursion limit, fuzzing, benchmarks, debug-parse
claude
committed
Vendor the reference grammar, give errors a position, refresh the docs
claude
committed
Add cmd/difftest, and fix the nine divergences it found
claude
committed
Teach difftest second-order edits and the round trip, and fix trans_opt
claude
committed
Check span invariants over the corpus, and fix named constraints
claude
committed
Extract do_eqp_test cases too
claude
committed
Stop allocating per token and per operator
claude
committed
Show enums whose zero value means something, and chain difftest edits
claude
committed
Test ast traversal and the dump renderer directly
claude
committed
Run difftest weekly
claude
committed
Distinguish NOTNULL from NOT NULL, and start on the review findings
claude
committed
Put the round trip, the split hazard and the truncation rule in one place
claude
committed
Apply the rest of the review: fewer allocations, less repetition
claude
committed
Check the transcriptions against the vendored upstream sources
claude
committed
Feed SQLite's own fuzz corpus to difftest, and stop at NUL like SQLite does
claude
committed
Bring PLAN.md's layout and hardening milestone up to date
claude
committed
Record the fuzzdata seeds in the README, and the sqllogictest decision
claude
committed
Give the weekly difftest a fresh seed and a bigger budget
claude
committed
Document the oracle's -seeds mode where the other modes are documented
claude
committed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Implement the lexer, AST and parser for the SQLite dialect - #3

Implement the lexer, AST and parser for the SQLite dialect#3
kyleconroy merged 23 commits into
mainfrom
claude/the-loop-claude-md-4bnogc

Commits on Jul 30, 2026

Implement the lexer, AST and parser for the SQLite dialect

Record how far SQLite's parser got, so truncated parses do not lie

Extract the corpus from every pinned test script, and assert error offsets

Add the SQL renderer, AST dump, round-trip property and snapshots

Harden the parser: recursion limit, fuzzing, benchmarks, debug-parse

Vendor the reference grammar, give errors a position, refresh the docs

Add cmd/difftest, and fix the nine divergences it found

Teach difftest second-order edits and the round trip, and fix trans_opt

Check span invariants over the corpus, and fix named constraints

Extract do_eqp_test cases too

Stop allocating per token and per operator

Show enums whose zero value means something, and chain difftest edits

Test ast traversal and the dump renderer directly

Run difftest weekly

Distinguish NOTNULL from NOT NULL, and start on the review findings

Put the round trip, the split hazard and the truncation rule in one place

Apply the rest of the review: fewer allocations, less repetition

Check the transcriptions against the vendored upstream sources

Feed SQLite's own fuzz corpus to difftest, and stop at NUL like SQLite does

Bring PLAN.md's layout and hardening milestone up to date

Record the fuzzdata seeds in the README, and the sqllogictest decision

Give the weekly difftest a fresh seed and a bigger budget

Document the oracle's -seeds mode where the other modes are documented