The GitHub Actions job "Java CI with Ant" on poi.git/trunk has succeeded. Run started by GitHub user pjfanning (triggered by pjfanning).
Head commit for run: a400d35eecdbe22b3cab4b6698fa88bc9a8b53b9 / PJ Fanning <[email protected]> RLEDecompressingInputStream: recover from bad chunks more consistently (#1497) * RLEDecompressingInputStream: recover from bad chunks more consistently Follow-up to #1496. - VBAMacroReader: a module whose stream was read before its dir stream now falls back to the brute-force search when the dir offset is bad, as the other read order already did. A dir stream that fails to decompress is reported as an IOException with the record id, like other dir errors. - A chunk that fails to decode leaves the stream at its end instead of leaving bytes from the previous chunk to be read. - Input that ends inside a literal or copy token keeps the bytes already decompressed in that chunk, as input that ends at a flag byte does. - Uncompressed chunks must hold exactly 4096 bytes, per MS-OVBA. - Tests: cover reading past an empty chunk, compare the 4096-byte chunk in one assertion. Co-Authored-By: Claude Opus 5.5 <[email protected]> * TestRLEDecompressingInputStream: drop forbidden InputStream.available() call Co-Authored-By: Claude Opus 5.5 <[email protected]> * TestRLEDecompressingInputStream: check available() after a failed chunk under @SuppressForbidden Co-Authored-By: Claude Opus 5.5 <[email protected]> --------- Co-authored-by: Claude Opus 5.5 <[email protected]> Report URL: https://github.com/apache/poi/actions/runs/40756110599 With regards, GitHub Actions via GitBox --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
