@tatsuhiro-t, thanks for your PR! By analyzing the history of the files in this pull request, we identified @Andersbakken, @jay and @bagder to be potential reviewers.

Good point.. Yes, we need to parse it. Does curl have convenient function to parse this comma separated list header fields?

Does curl have convenient function to parse this comma separated list header fields?

hm offhand I can't think of one. There's always strtok_r. Also I don't think we have to check for the semi-colon in trailers if I'm reading this right:

       TE        = "TE" ":" #( t-codings )
       t-codings = "trailers" | ( transfer-extension [ accept-params ] )
       transfer-coding         = "chunked" | transfer-extension
       transfer-extension      = token *( ";" parameter )

Thank you. I ended up not to use strtok_r since it requires NULL-terminated string, and it replaces delimiter with NUL byte. Instead, I added hand crafted "trailers" detection code.
Conflict has been resolved as well.

I ended up not to use strtok_r since it requires NULL-terminated string, and it replaces delimiter with NUL byte. Instead, I added hand crafted "trailers" detection code.

Oops I forgot about that. Your changes just landed. I made a few corrections:

• Typo. Searching for tab had used 't' instead of '\t'
• Unnecessary goto. We discourage goto when not necessary so I redid the logic.
• enum names too generic. (For example IGNORE is defined to 0 somewhere in my includes so I prefixed them with HEADERINST_)
• Unreachable code. Some compilers may warn that since the for(;;) is never exited the return after that block will never be hit, so I removed it.
• Minor style changes

Thanks!

The commit looks good. Thank you for merging this.