You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@ant.apache.org by Ben Pracht <bp...@nc.rr.com> on 2004/02/06 20:51:27 UTC

FixCRLF not processing files specified

I'm trying to remove the EOF's and reformat source files to keep
the GCC people happy.  Unfortuanately, I can get the FixCRLF task to
work on one file, but not on my source tree.

I have code like this:
         <fixcrlf srcdir="native"
                  eof="remove">
                 <include name="ephcmwin.c"/>
         </fixcrlf>

which works, and
         <apply executable="ls"
               dir="filtered_source/debug/Linux/2.4.18-14/i386"
                failonerror="true">
             <arg line="-ld"/>
             <fileset dir="filtered_source/debug/Linux/2.4.18-14/i386">
               <include name="**/*.c"/>
               <include name="**/*.cpp"/>
               <include name="**/*.h"/>
               <include name="**/*.hpp"/>
               <include name="**/*.java"/>
               <include name="**/*.txt"/>
               <include name="**/*.fgl"/>
               <include name="**/*.shl"/>
               <include name="**/*.rul"/>
               <include name="**/*.ipr"/>
             </fileset>
         </apply>

which lists all the files I want to process, and

         <fixcrlf srcdir="filtered_source/debug/Linux/2.4.18-14/i386"
                  eof="remove">
               <include name="**/*.c"/>
               <include name="**/*.cpp"/>
               <include name="**/*.h"/>
               <include name="**/*.hpp"/>
               <include name="**/*.java"/>
               <include name="**/*.txt"/>
               <include name="**/*.fgl"/>
               <include name="**/*.shl"/>
               <include name="**/*.rul"/>
               <include name="**/*.ipr"/>
         </fixcrlf>

which was supposed to fix the CRLF/EOF issues to keep GCC/G++ from
fussing.

The apply task lists all the files I'm interested in.  The second 
fixcrlf task does nothing that I can see.  When I looked at the 
documentation and source, it looked like it checked timestamps between 
the source and destination directories.  I want to do an inplace 
transformation, so comparing timestamps doesn't make any sense.  But in 
case it does somehow make sense to someone, is there any way I can force
the fixcrlf task to just do it?

Btw.  Would anyone else be interested in this being implemented as a 
Filter also?

Thanks,
Ben Pracht
Thanks in advance,
Ben



---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@ant.apache.org
For additional commands, e-mail: user-help@ant.apache.org


Re: FixCRLF not processing files specified

Posted by "Alexey N. Solofnenko" <A....@mdl.com>.
<fixcrlf> seems broken. I had to write my own <scriptdef> task to 
replace it. Fortunately it is simple now.

- Alexey.

Ben Pracht wrote:

> I'm trying to remove the EOF's and reformat source files to keep
> the GCC people happy.  Unfortuanately, I can get the FixCRLF task to
> work on one file, but not on my source tree.
>
> ....


---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@ant.apache.org
For additional commands, e-mail: user-help@ant.apache.org


Re: FixCRLF not processing files specified

Posted by Antoine Lévy-Lambert <an...@antbuild.com>.
Ben Pracht wrote:

> I'm trying to remove the EOF's and reformat source files to keep
> the GCC people happy.  Unfortuanately, I can get the FixCRLF task to
> work on one file, but not on my source tree.
>
> I have code like this:
>         <fixcrlf srcdir="native"
>                  eof="remove">
>                 <include name="ephcmwin.c"/>
>         </fixcrlf>
>
> which works, and
>         <apply executable="ls"
>               dir="filtered_source/debug/Linux/2.4.18-14/i386"
>                failonerror="true">
>             <arg line="-ld"/>
>             <fileset dir="filtered_source/debug/Linux/2.4.18-14/i386">
>               <include name="**/*.c"/>
>               <include name="**/*.cpp"/>
>               <include name="**/*.h"/>
>               <include name="**/*.hpp"/>
>               <include name="**/*.java"/>
>               <include name="**/*.txt"/>
>               <include name="**/*.fgl"/>
>               <include name="**/*.shl"/>
>               <include name="**/*.rul"/>
>               <include name="**/*.ipr"/>
>             </fileset>
>         </apply>
>
> which lists all the files I want to process, and
>
>         <fixcrlf srcdir="filtered_source/debug/Linux/2.4.18-14/i386"
>                  eof="remove">
>               <include name="**/*.c"/>
>               <include name="**/*.cpp"/>
>               <include name="**/*.h"/>
>               <include name="**/*.hpp"/>
>               <include name="**/*.java"/>
>               <include name="**/*.txt"/>
>               <include name="**/*.fgl"/>
>               <include name="**/*.shl"/>
>               <include name="**/*.rul"/>
>               <include name="**/*.ipr"/>
>         </fixcrlf>
>
> which was supposed to fix the CRLF/EOF issues to keep GCC/G++ from
> fussing.
>
> The apply task lists all the files I'm interested in.  The second 
> fixcrlf task does nothing that I can see.  When I looked at the 
> documentation and source, it looked like it checked timestamps between 
> the source and destination directories.  I want to do an inplace 
> transformation, so comparing timestamps doesn't make any sense.  But 
> in case it does somehow make sense to someone, is there any way I can 
> force
> the fixcrlf task to just do it?
>
> Btw.  Would anyone else be interested in this being implemented as a 
> Filter also?
>
> Thanks,
> Ben Pracht
> Thanks in advance,
> Ben
>
>
>
> Hi Ben,


I think it is a good idea to have fixcrlf implemented as a filter.

If you want to suggest a fix for the bug of timestamp comparison on a 
fileset, it is a good idea.

Antoine




---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@ant.apache.org
For additional commands, e-mail: user-help@ant.apache.org