空提交上的 git format-patch 返回意外结果

Question

在 git docs 上找不到关于这个问题的任何解释：

如果我创建一个虚拟提交，带有一些虚拟差异，我会得到一个正常的补丁 当我跑步时

git format-patch -1 -o outgoing/ -p -k

但如果最后一次提交是空提交，则由

生成

git commit --allow-empty "Some commit message"

那么格式补丁的输出将是一个空补丁。如果 第一种情况会产生这样的结果：

From 08cfdb2994554d834b89309ca96d9bf513e26a90 Mon Sep 17 00:00:00 2001
From: User <mail@example.com>
Date: Fri, 8 Jan 2016 12:44:57 +0000
Subject: dummy commit


diff --git a/lol.txt b/lol.txt
new file mode 100644
index 0000000..f944b38
--- /dev/null
+++ b/lol.txt
@@ -0,0 +1 @@
+:)
--
2.5.4 (Apple Git-61)

那么第二种情况不应该生成这样的东西吗？

From 2d486f25c48780e2e132047e681929fcccb7e60c Mon Sep 17 00:00:00 2001
From: User <mail@example.com>
Date: Fri Jan 8 12:43:55 2016 +0000
Subject: Some commit message


2.5.4 (Apple Git-61)

Answer 1

2022 年更新：使用 Git 2.35（2022 年第一季度），“git am”^(man) 学习 --empty=(stop|drop|keep)" 和 --allow-empty 选项来调整对 e 的处理- 没有补丁的邮件。

见“git-am with mailbox patch fails when it contains a cover letter”。

---

2016：原始答案：

注意：如果空提交不是最后一个，它会起作用（如“Git patch of empty commits”中所述）

关于empty commit patch in this thread back in 2010的争论。

一个好消息是format-patch 已经使用--always 命令行选项从空提交中生成消息，但是因为它不能与“am”一起应用，所以它是毫无意义的。

（--always 已通过to git diff-tree）

你需要做一些测试，但默认情况下，确实不包括空提交。

Answer 2

从 git 邮件列表中提取：

杰夫金说：

I'm not sure if this is a bug or not.

In the beginning, git's revision-traversal machinery generally does not show 
commits which have no diff. Over the years, commands like "git log"
learned to set the "always_show_header" option to show even empty commits. 
But format-patch never did.

然后 Junio C Hamano 补充道：

The patch based workflow support is geared towards helping the recipient 
of the patches a lot more than the contributors, and to prevent mistakes while 
applying the patches, "am" would stop when it sees such an empty e-mail as you saw 
(in the later part of message I am not quoting). After all, a "format-patch" output 
that does not have any patch would be indistinguishable from discussion e-mail 
messages and the recipient would not want to end up with no-op commits 
that record such messages.

So I think skipping no-op commit from the output was done pretty much deliberately 
and it is definitely not a bug.  I however do not think it is incorrect 
to say that it is a lack of feature that nobody 
so far found necessary or beneficial.

I would not refuse to consider adding a new option to "format-patch" 
to emit such a no-op message, and add a "having no patch is OK, just record a 
no-op commit" option to "am", though.  But I do not see a clear benefit from 
such change--it sounds more like a set of"because we could" not 
"because we need to" changes to me.

线程在http://git.661346.n2.nabble.com/git-format-patch-on-empty-commit-td7645342.html 感谢@VonC 找到它