Git的传输协议如何工作 [英] How does Git's transfer protocol work

问题描述

我在Git上工作了一年多，现在我必须向团队中的其他人解释它.这就是为什么我需要更多背景知识的原因. 去年，我花了大部分时间阅读Git书，最近我又继续阅读第10章.在

I am working with Git for more than one year and now I have to explain it to others in our group. That is why I need a bit more backround. I went thourgh most of the Git Book in the last year and recently I continued with chapter 10. In chapter 10.6 I got completely stuck:

让我们遵循simplegit库的http-fetch过程:

Let’s follow the http-fetch process for the simplegit library:

$ git clone http://server/simplegit-progit.git

此命令要做的第一件事是下拉info/refs文件. 此文件由update-server-info命令编写，因此 您需要将其启用为接收后挂钩，以进行HTTP 运输以正常工作:

The first thing this command does is pull down the info/refs file. This file is written by the update-server-info command, which is why you need to enable that as a post-receive hook in order for the HTTP transport to work properly:

=> GET info/refs
ca82a6dff817ec66f44342007202690a93763949     refs/heads/master

我有一个小型测试仓库https://github.com/to_my/repo，git clone效果很好.但是

I have a small test repo https://github.com/to_my/repo and git clone works well. But

• 文件夹info/refs在哪里?我只在clone ...
之后找到一个/.git/info/exclude
• 我应该如何使用update-server-info命令?是git clone的一部分吗?
• 尽管我了解钩子(我认为)并使用预提交钩子来自动增加软件包版本，但我对"...这就是为什么您需要将其启用为接收后钩子"感到迷惑. /li>
• 我在git bash工作中无法获得命令GET info/refs.

• Where is the folder info/refs? I only find a /.git/info/exclude afther the clone...
• How should I use the update-server-info command? Is it part of git clone somehow?
• I am competely lost with "...which is why you need to enable that as a post-receive hook" although I understand hooks (I thought) and use a pre-commit hook for automatically increasing the package version.
• I can't get the command GET info/refs in git bash work.

很抱歉，如果问题很愚蠢，但我只是不明白如何将文档中的这些内容放在一起.

Sorry if the questions are stupid, but I just don't understand how to put these pieces from the documentation together.

推荐答案

文件夹信息/引用在哪里?我只找到/.git/info/排除此克隆...

Where is the folder info/refs? I only find a /.git/info/exclude afther the clone...

没有这样的文件夹(它不是目录)，但是-.git/info/refs-将是 file 所在的位置(如果有文件的话)

There is no such folder (it's not a directory), but that—.git/info/refs—would be where the file would be, if there were a file there.

我应该如何使用update-server-info命令?它是git clone的一部分吗?

How should I use the update-server-info command? Is it part of git clone somehow?

通常，您应该不使用它:它仅用于哑"传输. 智能"(双向对话)传输不需要它.

In general, you should not use it: it's only for "dumb" transports. "Smart" (two way conversation) transports don't need it.

尽管我了解钩子(我认为)并使用预提交钩子来自动增加软件包版本，但我对"...这就是为什么需要将其启用为接收后钩子"感到迷惑. /p>

I am competely lost with "...which is why you need to enable that as a post-receive hook" although I understand hooks (I thought) and use a pre-commit hook for automatically increasing the package version.

如果出于某种原因要启用哑传输，则每次文件需要创建或更新时，都需要运行某些文件来创建或更新多个文件.每当引用更改时，都需要更新info/refs文件，因此在接收后挂钩中可以很好地运行"something". 东西"是命令git update-server-info.

If, for some reason, you want to enable dumb transports, you need to run something to create or update several files every time they need creating or updating. The info/refs file needs to be updated whenever references change, so a good place to run the "something" is in a post-receive hook. The "something" is the command git update-server-info.

请注意，如果您未在服务器上运行仅推送的裸存储库，则仅运行后接收脚本git update-server-info是不够的，因为可以通过其他方式添加提交和其他对象(手动git commit例如).在这种情况下，您可能会使用例如cron作业，以时钟驱动为基础来创建或更新哑传输信息.

Note that if you are not running a push-only bare repository on a server, having a post-receive script run git update-server-info is not sufficient, since commits and other objects can be added by other means (manual git commits for instance). In this case you might use, e.g., a cron job to create-or-update dumb-transport information on a clock-driven basis.

我无法在git bash工作中获得命令GET info/refs.

如果该文件存在，则可以通过HTTP获得该文件，例如，从浏览器或使用curl命令.

If the file existed, you would obtain it via HTTP, e.g., from a browser or with the curl command.

这篇关于Git的传输协议如何工作的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！