Support for index data by splitdiff command

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Support for index data by splitdiff command

SF Markus Elfring
Hello,

I would like to point out an open issue for the tool "splitdiff" from the software package "patchutils 0.3.0-1.52".

A special header "Index:" is mentioned in the article "http://en.wikipedia.org/wiki/Diff". Tools like Subversion write it into difference files. The current command does not handle this data correctly.

The following input ...

Index: abc.h
===================================================================
--- abc.h (Revision 123)
+++ abc.h (work copy)
@@ -1,1 +1,1 @@
-check
+out
Index: xyz.h
===================================================================
--- xyz.h (Revision 987)
+++ xyz.h (work copy)
@@ -1,1 +1,1 @@
-test
+this

... is converted into the following arrangement by "splitdiff -a".

First source file:
--- abc.h (Revision 123)
+++ abc.h (work copy)
@@ -1,1 +1,1 @@
-check
+out
Index: xyz.h
===================================================================

Second source file:
--- xyz.h (Revision 987)
+++ xyz.h (work copy)
@@ -1,1 +1,1 @@
-test
+this


I would appreciate if the first file will keep its own header and will not additionally get informations from the second file.

Regards,
Markus
Reply | Threaded
Open this post in threaded view
|

Re: Support for index data by splitdiff command

Sergio Monteiro Basto
Hi sorry for the SPAM , my fault , elmz is too ... , last try ( I promise :))

Hi,
Don't know if it help, I use svn and I use filterdiff.
filterdiff cleans all things that don't belongs to patch and that
includes Index ...
If you just want a clean splitdiff, you may use :

svn diff | filterdiff  > txt; splitdiff -a txt


On Wed, 2009-10-07 at 03:22 +0000, [hidden email]
wrote:

>         Hello,
>        
>         I would like to point out an open issue for the tool
>         "splitdiff" from the software package "patchutils 0.3.0-1.52".
>        
>         A special header "Index:" is mentioned in the article
>         "http://en.wikipedia.org/wiki/Diff". Tools like Subversion
>         write it into difference files. The current command does not
>         handle this data correctly.
>        
>         The following input ...
>        
>         Index: abc.h
>         ===================================================================
>         --- abc.h       (Revision 123)
>         +++ abc.h       (work copy)
>         @@ -1,1 +1,1 @@
>         -check
>         +out
>         Index: xyz.h
>         ===================================================================
>         --- xyz.h       (Revision 987)
>         +++ xyz.h       (work copy)
>         @@ -1,1 +1,1 @@
>         -test
>         +this
>        
>         ... is converted into the following arrangement by "splitdiff
>         -a".
>        
>         First source file:
>         --- abc.h       (Revision 123)
>         +++ abc.h       (work copy)
>         @@ -1,1 +1,1 @@
>         -check
>         +out
>         Index: xyz.h
>         ===================================================================
>        
>         Second source file:
>         --- xyz.h       (Revision 987)
>         +++ xyz.h       (work copy)
>         @@ -1,1 +1,1 @@
>         -test
>         +this
>        
>        
>         I would appreciate if the first file will keep its own header
>         and will not additionally get informations from the second
>         file.
>        
>         Regards,
>         Markus
--
Sérgio M. B.

smime.p7s (2K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Support for index data by splitdiff command

SF Markus Elfring
In reply to this post by SF Markus Elfring
> filterdiff cleans all things that don't belongs to patch and that
> includes Index ...

I would prefer to keep all informations intact. Is still any fine-tuning needed for the deletion of unwanted source lines?

The command "cvs diff -u" generates also this data together with the field "RCS file:" which will be lost, too. Would you like to consider such details as supported extensions for the patch data format?

Regards,
Markus
Reply | Threaded
Open this post in threaded view
|

Re: Support for index data by splitdiff command

Sergio Monteiro Basto
On Sat, 2009-10-10 at 21:10 +0200, SF Markus Elfring wrote:
> > filterdiff cleans all things that don't belongs to patch and that
> > includes Index ...
>
> I would prefer to keep all informations intact.

> Is still any fine-tuning needed for the deletion of unwanted source lines?

I don't understand the question ... , filterdiff ?!

> The command "cvs diff -u" generates also this data together with the field "RCS file:" which will be lost, too.

> Would you like to consider such details as supported extensions for the patch data format?

I can't answer on that, I am not the maintainer of sources.

Regards,
--
Sérgio M. B.

smime.p7s (2K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Support for index data by splitdiff command

Sergio Monteiro Basto
In reply to this post by SF Markus Elfring
On Sat, 2009-10-10 at 21:10 +0200, SF Markus Elfring wrote:
> > filterdiff cleans all things that don't belongs to patch and that
> > includes Index ...
>
> I would prefer to keep all informations intact. Is still any fine-tuning needed for the deletion of unwanted source lines?
>
> The command "cvs diff -u" generates also this data together with the field "RCS file:" which will be lost, too. Would you like to consider such details as supported extensions for the patch data format?
>
> Regards,
> Markus


http://cyberelk.net/tim/software/patchutils/
Fixcvsdiff is for correcting the output of ‘cvs diff’.

--
Sérgio M. B.

smime.p7s (4K) Download Attachment