Re: [PATCH 11/13] remote-curl: accept empty line as terminator
From: Jonathan Nieder <hidden>
Date: 2016-06-15 22:49:24
Sverre Rabbelier wrote:
I noticed this when my tests suddenly broke. Bisecting pointed at the 'more rigorous return value checking' patch
Shouldn't this go before "check status code of finish_command" for bisectability, then?
I'm not very sure about the error message, if anyone feels it should go (it indicates a bug in the remote helper infrastructure, not a user error) it can be left out as far as I'm concerned.
No preference here.
quoted hunk ↗ jump to hunk
--- a/remote-curl.c +++ b/remote-curl.c@@ -813,6 +813,8 @@ int main(int argc, const char **argv) do { if (strbuf_getline(&buf, stdin, '\n') == EOF) break; + if (buf.len == 0) + break;
This is just a bug, I think. Other strbuf_getline() invocations in that file all use the equivalent if (*buf->buf) break; too.
quoted hunk ↗ jump to hunk
@@ -851,6 +853,7 @@ int main(int argc, const char **argv) printf("\n"); fflush(stdout); } else { + fprintf(stderr, "Unknown command '%s'\n", buf.buf); return 1; }
Reviewed-by: Jonathan Nieder <redacted>