It's partly interesting, but as every static analyzer it's at least half wrong, so it has to be used carefully, and it takes time. For example, the first issue is both irrelevant and mostly wrong (it probably comes from the fact Assert is defined conditionally as a non-aborting check, which makes the analyzer incorrectly think the checked value can be NULL), and the third is totally wrong, which might be interesting to report to the authors (it wrongly assumes control flow could skip the body of a while(1) loop). src/encodings.c:608: error: PREMATURE_NIL_TERMINATION_ARGUMENT is also incorrect, because of the semantic of the "incriminated" function.

Anyway, thank you for this, but it has to be sorted out, as it's not very useful straight out of the analyzer.


You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.