Re: Expose subprojects as special files to "git diff" machinery

[Andy Parkins <andyparkins@gmail.com>]

On Sunday 2007, April 15, Linus Torvalds wrote:

> 	diff --git a/sub-A b/sub-A
> 	index 2de597b..e8f11a4 160000
> 	--- a/sub-A
> 	+++ b/sub-A
> 	@@ -1 +1 @@
> 	-Subproject commit 2de597b5ad348b7db04bd10cdd38cd81cbc93ab5
> 	+Subproject commit e8f11a45c5c6b9e2fec6d136d3fb5aff75393d42

Isn't this dangerous because it looks just like a normal diff with a 
file being rewritten, when in truth it is a tree entry record being 
rewritten.  Possibly this is in the same category as the enhancements 
needed for rename support

cf.

diff --git a/file1 b/file2
similarity index 100%
rename from file1
rename to file2

There's no need to have a line marker, as there is only one "line" in 
the subproject "file", and no need for the index line because that 
information is contained in the change itself.  So something like:

diff --git a/sub-A b/sub-B
subproject from commit 2de597b5ad348b7db04bd10cdd38cd81cbc93ab5
subproject to commit e8f11a45c5c6b9e2fec6d136d3fb5aff75393d42

Regardless of the format, I'm sure this should be treated as part of 
git's extended diff syntax, so that if a diff like this ever got into 
the wild it wouldn't be a disaster if someone tried to apply it.



Andy
-- 
Dr Andy Parkins, M Eng (hons), MIET
andyparkins@gmail.com