On Tue, Feb 11, 2020 at 08:17:30AM -0800, Junio C Hamano wrote: > Jeff King <peff@xxxxxxxx> writes: > > > Something like the (largely untested) patch below. That would also make > > it easy to support arbitrary amounts of whitespace after the header, > > which I think are allowed by the standard (whereas now we'd parse > > "Content-type: text/plain" as " text/plain", which is silly). > > > > Worth doing? > > The result does look cleaner. I do not think we've seen much > polishing in this area of the code for quite a long time, which > might indicate that what we have may be good enough, but at the same > time it would mean it is quiescent time for the code and it is safe > to clean it up. Yeah, it feels a little like code churn. It's not like we ever add new headers to match. But I doodled the series below while I was in a meeting, and it does fix one tiny user-visible bug (that I don't think anybody has ever noticed or complained about), so maybe it's worth doing. ;) This is on top of rs/strbuf-insertstr, since it relies on 517b60564e (mailinfo: don't insert header prefix for handle_content_type(), 2020-02-10). [1/4]: mailinfo: treat header values as C strings [2/4]: mailinfo: simplify parsing of header values [3/4]: mailinfo: be more liberal with header whitespace [4/4]: mailinfo: factor out some repeated header handling mailinfo.c | 62 +++++++++++++++++++++++++++------------------ t/t5100-mailinfo.sh | 15 +++++++++++ 2 files changed, 52 insertions(+), 25 deletions(-) -Peff
