Re: contents of a .git file
- From
brian m. carlson <sandals@crustytoothpaste.net>
- Date
- May 18, 2020, 01:01 UTC
- Message-ID
- <20200518010131.GM6362@camp.crustytoothpaste.net>
- In-Reply-To
- <CAP16ngpd_GbfUkSuBks5jk2vskks5_SRmVHOmfaEUTLoWiD8hQ@mail.gmail.com>
On 2020-05-17 at 21:58:41, bruce wrote:
Show 12 quoted lines
> Hi. > > Hope this is a place to ask this. > > There are lots of foo.git files. > > I'm trying to figure out what's un a git file. As an example, > gets the drupal repos. > However, I can't seem to find the drupal.git file. Even a net search > doesnt seem to turn up a file that I can view!! > > Any pointers/thoughts would be helpful.
The ending ".git" is just a conventional approach to indicate that a URL or directory is a bare repository. When using HTTP, the actual files are served from other locations, depending on whether the site is using the smart or dumb HTTP protocol. Most major hosting platforms prefer the smart protocol because it can send a lot less data.
So there really isn't a single file; instead, there's either some static files and some manifests (for the dumb protocol) or conversations with a CGI client (for the smart protocol). You can read more about it in Documentation/technical/http-protocol.txt, and if you're interested, you can run your operations with GIT_CURL_VERBOSE=1 to see the endpoints that are contacted. (Be aware that that output may contain credentials.)
-- brian m. carlson: Houston, Texas, US OpenPGP: https://keybase.io/bk2204