Cache per-decision look-1 to skip non-viable speculative paths #5
There are no files selected for viewing
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
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -19,7 +19,7 @@ jobs: | |
|
|
||
| steps: | ||
| - name: Checkout | ||
| uses: actions/checkout@v6 | ||
| with: | ||
| persist-credentials: false | ||
|
|
||
|
|
||
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
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
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
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,86 @@ | ||
| # Development notes | ||
|
|
||
| ## Kotlin parser parity perf benchmark | ||
|
|
||
| Reproduces the timings tracked in GitHub issue #4 against the Kotlin grammar from `antlr/grammars-v4`. | ||
|
|
||
| ### One-time setup (fresh checkout) | ||
|
|
||
| ```bash | ||
| # 1. ANTLR jar (any path; pin v4.13.2) | ||
| mkdir -p /tmp/antlr-cleanroom/tools | ||
| curl -fLo /tmp/antlr-cleanroom/tools/antlr-4.13.2-complete.jar \ | ||
| https://www.antlr.org/download/antlr-4.13.2-complete.jar | ||
|
|
||
| # 2. grammars-v4 checkout (sparse, just the kotlin grammar) | ||
| mkdir -p /tmp/antlr-cleanroom/grammars-v4 | ||
| git -C /tmp/antlr-cleanroom/grammars-v4 init -q | ||
| git -C /tmp/antlr-cleanroom/grammars-v4 remote add origin https://github.com/antlr/grammars-v4.git | ||
| git -C /tmp/antlr-cleanroom/grammars-v4 sparse-checkout init --cone | ||
| git -C /tmp/antlr-cleanroom/grammars-v4 sparse-checkout set kotlin/kotlin | ||
| git -C /tmp/antlr-cleanroom/grammars-v4 fetch --depth 1 origin 284602b3f23ca54dc30778204ab7ae9e969145e9 | ||
| git -C /tmp/antlr-cleanroom/grammars-v4 checkout FETCH_HEAD | ||
| ``` | ||
|
|
||
| ### Run the parity smoke + dumper build | ||
|
|
||
| ```bash | ||
| tests/kotlin-parity/run.sh \ | ||
| --antlr-jar /tmp/antlr-cleanroom/tools/antlr-4.13.2-complete.jar \ | ||
| --grammars-v4 /tmp/antlr-cleanroom/grammars-v4 | ||
| ``` | ||
|
|
||
| That regenerates the Rust parser from the Kotlin grammar `.interp`, builds `tests/kotlin-parity/dumper`, and asserts the parse trees match `antlr4-python3-runtime` byte-for-byte. | ||
|
|
||
| ### Measure parse-only timings | ||
|
|
||
| The dumper has a built-in parse-only stopwatch so process startup (~10 ms) is excluded: | ||
|
|
||
| ```bash | ||
| DUMPER=tests/kotlin-parity/dumper/target/release/kotlin-parity-dumper | ||
| for snippet in tests/kotlin-parity/snippets/*.kt; do | ||
| echo "=== $(basename "$snippet") ===" | ||
| "$DUMPER" --input "$snippet" --output /tmp/dump.txt --iters 5 --time | ||
| done | ||
| ``` | ||
|
|
||
| `--iters N` repeats parse N times within one process; `--time` prints `min`/`avg` to stderr. CI baseline numbers from issue #4: Python `antlr4-python3-runtime` parse-only ≈ 0.9–14.9 ms (`01-nested-types.kt`) and 226–317 ms (`02-dataframe.kt`). | ||
|
|
||
| ## ANTLR runtime testsuite (357 cases) | ||
|
|
||
| Validates the Rust runtime against ANTLR's upstream conformance descriptors. | ||
|
|
||
| ### One-time setup | ||
|
|
||
| ```bash | ||
| git clone --depth 1 https://github.com/antlr/antlr4 /tmp/antlr-cleanroom/antlr4-upstream | ||
| ``` | ||
|
|
||
| The harness reads `antlr4-upstream/runtime-testsuite` and the same ANTLR jar fetched above. | ||
|
|
||
| ### Run the full sweep | ||
|
|
||
| ```bash | ||
| cargo run --release --quiet --bin antlr4-runtime-testsuite | ||
| ``` | ||
|
|
||
| Defaults to `ANTLR4_JAR=/tmp/antlr-cleanroom/tools/antlr-4.13.2-complete.jar` and `ANTLR4_RUNTIME_TESTSUITE=/tmp/antlr-cleanroom/antlr4-upstream/runtime-testsuite`. Override with `--antlr-jar`/`--descriptors` or env vars. Expected outcome: `summary: 357 passed, 0 failed, 0 skipped, 357 run`. Wall-clock ≈ 10–15 minutes on Apple Silicon, ≈ 30 minutes on the GitHub Linux runner. | ||
|
|
||
| ### Run a subset while iterating | ||
|
|
||
| ```bash | ||
| # One descriptor: | ||
| cargo run --release --quiet --bin antlr4-runtime-testsuite -- --case LexerExec/KeywordID | ||
|
|
||
| # One group (e.g. while debugging left-recursion): | ||
| cargo run --release --quiet --bin antlr4-runtime-testsuite -- --group LeftRecursion --limit 20 | ||
|
|
||
| # Keep the per-case temp crates for inspection: | ||
| cargo run --release --quiet --bin antlr4-runtime-testsuite -- --case ParserErrors/SingleSetInsertion --keep | ||
| ``` | ||
|
|
||
| Per-case scratch crates land under `target/antlr-runtime-testsuite/<case>/`. Stale dirs from a killed run can fail a re-run with `Os { code: 66, ... DirectoryNotEmpty }` — `rm -rf target/antlr-runtime-testsuite/*` to recover. | ||
|
|
||
| ## CI parity | ||
|
|
||
| CI runs `cargo clippy --locked --all-targets --all-features -- -D warnings`, so reproduce locally with the same flags before pushing — `clippy::excessive-nesting`, `clippy::disallowed_types`, and similar nursery/pedantic lints all promote to errors there. |
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
Oops, something went wrong.
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.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This change replaces a previously SHA-pinned action with a mutable tag in the publish pipeline, which runs with crates.io publishing privileges. In
.github/workflows/publish.yml, using@v1.0.4(and similarly@v6elsewhere in this commit) allows upstream retags or compromised releases to change executed code without any change in this repository, undermining reproducibility and increasing supply-chain risk for releases. Keep theseuses:entries pinned to full commit SHAs (with version comments) as before.Useful? React with 👍 / 👎.