akshatshenoi-db commented on PR #56784: URL: https://github.com/apache/spark/pull/56784#issuecomment-4805000162
**Self-review (AI-assisted) — 0 blocking, 0 non-blocking, 0 nits.** Reviewed the engine hoist + `.zip` extension. After the refactor `TarArchiveReader` and `ZipArchiveReader` differ in exactly one method (`openArchiveStream`), with the shared `readEntries` engine — including `ignoredPathSegmentRegex` filtering, close-shielding, eager-first-entry error surfacing, and task-completion cleanup — inherited from the base. A state-space pass over the wrapped `ZipArchiveInputStream` found every case handled or documented: the one unstreamable-zip case (a STORED entry sized only by a trailing data descriptor) surfaces as an exception, not silent corruption. The new `ZipArchiveReader` correctly applies no Hadoop codec layer, since zip entries are individually deflated. Test coverage mirrors the merged tar suites: `ArchiveReaderSuite` `.zip` cases (dispatch, empty/single/multiple entries, dir/dotfile/marker skipping, lazy advance, non-closing entry stream) plus per-format `*ZipArchiveReadSuite` on the shared scaffolding, with the text tests extracted into a container-agnostic `TextArchiveReadBase`. Compile/test validation rests on Apache CI (no warm local OSS build). <!-- ai-code-review --> -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
