dschmidt commented on PR #2936: URL: https://github.com/apache/tika/pull/2936#issuecomment-4985130422
Good catch on the entry size, that one was real: without the guard the crafted stsd escapes parse() as a NegativeArraySizeException (the new regression test reproduces exactly that when the guard is removed). Entry sizes beyond the remaining payload are now rejected as malformed. I also took the test gap point: the zero fixture now places a delayed foreign-key track before the undelayed still-image-time track, so an empty edit duration leaking between tracks would surface as a non-zero value in the existing assertion. Blank lines are gone too. The moov-level hdlr point I left alone: it is latent as you say, real-world container-level metadata declares 'mdta' there rather than 'meta', so guarding against it felt speculative. -- 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]
