# [PATCH 19/23] http-push.c::fetch_symref(): use the new http API 5 messages from 2009-06-05 to 2009-06-06. Participants: Tay Ray Chuan, Jakub Narebski, Junio C Hamano. Thread: https://gitlist.dev/t/19695 ## Tay Ray Chuan, 2009-06-05 16:05 Subject: [PATCH 19/23] http-push.c::fetch_symref(): use the new http API Message-ID: <20090606000559.483f22da.rctay89@gmail.com> URL: https://gitlist.dev/e/20090606000559.483f22da.rctay89%40gmail.com ``` From: Mike Hommey Subject: [WIP Patch 09/12] Use the new http API in fetch_symref() Date: Sun, 18 Jan 2009 09:04:34 +0100 Signed-off-by: Mike Hommey Signed-off-by: Tay Ray Chuan --- http-push.c | 20 +++----------------- 1 files changed, 3 insertions(+), 17 deletions(-) diff --git a/http-push.c b/http-push.c index 455e0aa..1714e4e 100644 --- a/http-push.c +++ b/http-push.c @@ -2029,27 +2029,13 @@ static void fetch_symref(const char *path, char **symref, unsigned char *sha1) { char *url; struct strbuf buffer = STRBUF_INIT; - struct active_request_slot *slot; - struct slot_results results; url = xmalloc(strlen(repo->url) + strlen(path) + 1); sprintf(url, "%s%s", repo->url, path); - slot = get_active_slot(); - slot->results = &results; - curl_easy_setopt(slot->curl, CURLOPT_FILE, &buffer); - curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite_buffer); - curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, NULL); - curl_easy_setopt(slot->curl, CURLOPT_URL, url); - if (start_active_slot(slot)) { - run_active_slot(slot); - if (results.curl_result != CURLE_OK) { - die("Couldn't get %s for remote symref\n%s", - url, curl_errorstr); - } - } else { - die("Unable to start remote symref request"); - } + if (http_get_strbuf(url, &buffer, 0) != HTTP_OK) + die("Couldn't get %s for remote symref\n%s", url, + curl_errorstr); free(url); free(*symref); -- 1.6.3.1 ``` ## Jakub Narebski, 2009-06-05 20:54 Subject: Re: [PATCH 19/23] http-push.c::fetch_symref(): use the new http API Message-ID: URL: https://gitlist.dev/e/m37hzq5rod.fsf%40localhost.localdomain In-Reply-To: <20090606000559.483f22da.rctay89@gmail.com> ``` Tay Ray Chuan writes: > From: Mike Hommey > Subject: [WIP Patch 09/12] Use the new http API in fetch_symref() What is the subject of this patch: "http-push.c::fetch_symref(): use the new http API" or "Use the new http API in fetch_symref()" > Date: Sun, 18 Jan 2009 09:04:34 +0100 > > Signed-off-by: Mike Hommey > Signed-off-by: Tay Ray Chuan > --- If you wanted to note old subject, it would be better to put it in comments (here). > http-push.c | 20 +++----------------- > 1 files changed, 3 insertions(+), 17 deletions(-) > > diff --git a/http-push.c b/http-push.c > index 455e0aa..1714e4e 100644 > --- a/http-push.c > +++ b/http-push.c > @@ -2029,27 +2029,13 @@ static void fetch_symref(const char *path, char **symref, unsigned char *sha1) > { > char *url; > struct strbuf buffer = STRBUF_INIT; > - struct active_request_slot *slot; > - struct slot_results results; > > url = xmalloc(strlen(repo->url) + strlen(path) + 1); > sprintf(url, "%s%s", repo->url, path); > > - slot = get_active_slot(); > - slot->results = &results; > - curl_easy_setopt(slot->curl, CURLOPT_FILE, &buffer); > - curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite_buffer); > - curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, NULL); > - curl_easy_setopt(slot->curl, CURLOPT_URL, url); > - if (start_active_slot(slot)) { > - run_active_slot(slot); > - if (results.curl_result != CURLE_OK) { > - die("Couldn't get %s for remote symref\n%s", > - url, curl_errorstr); > - } > - } else { > - die("Unable to start remote symref request"); > - } > + if (http_get_strbuf(url, &buffer, 0) != HTTP_OK) > + die("Couldn't get %s for remote symref\n%s", url, > + curl_errorstr); > free(url); > > free(*symref); > -- > 1.6.3.1 > -- Jakub Narebski Poland ShadeHawk on #git ``` ## Tay Ray Chuan, 2009-06-06 01:02 Subject: Re: [PATCH 19/23] http-push.c::fetch_symref(): use the new http API Message-ID: URL: https://gitlist.dev/e/be6fef0d0906051802n4f6d5df3g453b292787ed62d0%40mail.gmail.com In-Reply-To: ``` Hi, On Sat, Jun 6, 2009 at 4:54 AM, Jakub Narebski wrote: >> From: Mike Hommey >> Subject: [WIP Patch 09/12] Use the new http API in fetch_symref() > > What is the subject of this patch: >  "http-push.c::fetch_symref(): use the new http API" > or >  "Use the new http API in fetch_symref()" > [...] > If you wanted to note old subject, it would be better to put it in > comments (here). Since the title of the patch differed from Mike's original, I was afraid the attribution to the author's original patch was lost on the reader of the commit message, thus I added the original subject, in addition to the author and date, into the commit message. I didn't add the subject line in Mike's other patches, like patch #11, because the title of the original patch and the current one are the same, so it's easy for readers to know what patch is being attributed to. -- Cheers, Ray Chuan ``` ## Junio C Hamano, 2009-06-06 05:32 Subject: Re: [PATCH 19/23] http-push.c::fetch_symref(): use the new http API Message-ID: <7vvdnadj0z.fsf@alter.siamese.dyndns.org> URL: https://gitlist.dev/e/7vvdnadj0z.fsf%40alter.siamese.dyndns.org In-Reply-To: ``` Tay Ray Chuan writes: > On Sat, Jun 6, 2009 at 4:54 AM, Jakub Narebski wrote: >>> From: Mike Hommey >>> Subject: [WIP Patch 09/12] Use the new http API in fetch_symref() >> >> What is the subject of this patch: >>  "http-push.c::fetch_symref(): use the new http API" >> or >>  "Use the new http API in fetch_symref()" >> [...] >> If you wanted to note old subject, it would be better to put it in >> comments (here). > > Since the title of the patch differed from Mike's original, I was > afraid the attribution to the author's original patch was lost on the > reader of the commit message, thus I added the original subject, in > addition to the author and date, into the commit message. Please don't. If you chose to use a better (in your opinion) subject from the original one, we'd want to have that improved one in our commit. The same thing if you chose to rewrite the proposed commit log message, or the contents of the patch, to add your own improvement. That is what we review here, and if people think your submitted version is good, that is what we want to record as a commit. So, in general we would want to see something like this: (1) From: Tay Ray Chuan Subject: [PATCH 19/23] http-push.c:fetch_symref(): use the new HTTP API (2) From: Mike Hommey (3) Now the contents of a file can be slurped into a strbuf with a single call to http_get_strbuf(), use it to simplify fetch_symref(). (4) Signed-off-by: Mike Hommey Signed-off-by: Tay Ray Chuan (5) --- (6) This is based on Mike's earlier patch: Subject: [WIP Patch 09/12] Use the new http API in fetch_symref() Date: Sun, 18 Jan 2009 09:04:34 +0100 I retitlted Mike's patch in order to ... Also changed frotz and nitfol so that they distim doshes better... (7) http-push.c | 20 +++----------------- 1 files changed, 3 insertions(+), 17 deletions(-) (8) diff --git a/http-push.c b/http-push.c ... That is, (1) RFC2822 headers your MUA gives us; (2) In-body (fake)headers for From: if the author is different from what (1) records. If you are forwarding somebody else's patch verbatim without modification, you may want to add Date: to keep the original datestamp, but if you are sending a version with your own improvements, we would rather use the Date: recorded in (1) in the final commit. (3) Proposed commit log message; (4) Chain of Signed-off-by lines; (5) Three-dash separator; (6) Commentary to explain the history of the patch, and/or background information if necessary, (7) diffstat; (8) the patch itself. Thanks. ``` ## Tay Ray Chuan, 2009-06-06 07:20 Subject: Re: [PATCH 19/23] http-push.c::fetch_symref(): use the new http API Message-ID: URL: https://gitlist.dev/e/be6fef0d0906060020x1a3a03a6ya4e58de6c83854d0%40mail.gmail.com In-Reply-To: <7vvdnadj0z.fsf@alter.siamese.dyndns.org> ``` Hi, On Sat, Jun 6, 2009 at 1:32 PM, Junio C Hamano wrote: > Please don't. Ok. >    (5) Three-dash separator; > >    (6) Commentary to explain the history of the patch, and/or background >        information if necessary, Is there anyway git can store and display these "commentaries"? I find myself having to repeatedly write this. -- Cheers, Ray Chuan ```