[Cocci] -o option fails when all files are skipped

[christoph@burschka.de (Christoph Burschka)]

On 02/08/2014 06:30 PM, Julia Lawall wrote:
> On Sat, 8 Feb 2014, Christoph Burschka wrote:
> 
>> (Posted earlier at https://github.com/coccinelle/coccinelle/issues/16 )
>>
>> -----
>> Using the latest source (1.0.0-rc19), I'm trying to run this command:
>>
>> spatch --no-includes --sp-file repr.cocci in.c -o out.c
>>
>> This is the output I'm getting:
>>
>> init_defs_builtins:
>> /home/christoph/Software/install/share/coccinelle/standard.h
>> warning: line 9: should self be a metavariable?
>> warning: line 22: should self be a metavariable?
>> (ONCE) Expected tokens self PyObject PyString_FromString PyString_FromFormat
>> Skipping:in.c
>> Fatal error: exception Failure("-o can not be applied because there are
>> multiple modified files")
>> -----
>>
>> Apparently this is because the only input file is skipped.
>>
>> I was assuming that spatch would simply copy the unchanged input file in
>> that case, probably because I misinterpreted the following code block in
>> main.ml:
>>
>>     (match outfiles with
>>     | [infile, Some outfile] when infile =$= x && null xs ->
>>         Common.command2 ("cp " ^outfile^ " " ^ !output_file)
>>     | [infile, None] when infile =$= x && null xs ->
>>         Common.command2 ("cp " ^infile^ " " ^ !output_file)
>>     | _ ->
>>         failwith
>>
>> (Instead, outfiles is an empty list. I guess it filters out the
>> irrelevant files before reaching this point? Sorry, I've never looked at
>> this source code before today.)
>>
>> I'm trying to use this tool: https://fedorahosted.org/2to3c/
>>
>> It applies a collection of semantic patches to a Python2 C extension to
>> upgrade it to Python3... but apparently it fails when any of those
>> patches are irrelevant. Is there a better way to do this?
> 
> Personally, I never use -o.  Normally, Coccinelle prints a patch to the 
> standard output, and then you can apply the patch to your code.
> 
> I can make -o simply copy the file when there is nothing to do, if that is 
> the desired behavior.
> 
> julia
> 

Well, it's probably best if I do it differently.

The project I forked was originally trying to wrap spatch in a string
function, so it read the source file, wrote it into a temporary file,
then ran spatch on it with -o, then read the output file, *then*
generated a diff against the input file... that seems needlessly
complicated.
Instead, I've changed it to just run spatch and print the diff output
that it generates directly.

--
Christoph

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 246 bytes
Desc: OpenPGP digital signature
URL: <https://systeme.lip6.fr/pipermail/cocci/attachments/20140208/fcf19205/attachment.asc>