I can't really answer you regarding why ExecuteProcess doesn't work.
But just wanted to ask: Have you thought of trying TProcess instead? I
normally execute any external programs via TProcess with good results
- no matter the platform.
Regards,
- Graeme -
On 08/09/2011, brian wrote:
> Reinfor
In our previous episode, Graeme Geldenhuys said:
> I can't really answer you regarding why ExecuteProcess doesn't work.
> But just wanted to ask: Have you thought of trying TProcess instead? I
> normally execute any external programs via TProcess with good results
> - no matter the platform.
The i
On 09/08/2011 02:20 AM, Graeme Geldenhuys wrote:
I can't really answer you regarding why ExecuteProcess doesn't work.
But just wanted to ask: Have you thought of trying TProcess instead? I
normally execute any external programs via TProcess with good results
- no matter the platform.
Nope. I'l
In our previous episode, brian said:
> What's driving me crazy is that running the two commands via
> ExecuteProcess does the first step OK, but oggenc fails with an exit
> code of 1, operation not permitted.
>
> If I replace the ExecuteProcess with a call to fpSystem, concatenating
> the Comma
On 08/09/2011, Marco van de Voort wrote:
>
> The inner plumbing of executeprocess and tprocess should be the same.
> TProcess just has several options (shell and piping)
I thought of that, right after I sent my message. But maybe there is a
slim chance that he is simply not using ExecuteProcess()
In our previous episode, Graeme Geldenhuys said:
> > The inner plumbing of executeprocess and tprocess should be the same.
> > TProcess just has several options (shell and piping)
>
> I thought of that, right after I sent my message. But maybe there is a
> slim chance that he is simply not using E
On Wed, 7 Sep 2011, brian wrote:
Reinforcing the subject, this is using Linux - to be specific, FPC 2.4.2-0
under 64 bit Mint 9.
I'm trying to convert a large number of MP3 files to Ogg Vorbis.
This is a two step process, first I run mpg321 to switch them to a .wav file,
then oggenc to con
- "brian" schreef:
> This is reproducible on a whole batch of files, and yes, I've checked
>
> all the file permissions.
>
> It's got to be something obvious, or some quirk of Linux programming
> that I haven't met up with yet (I'm still a novice with FreePascal and
>
> Linux, though I'v
Brian:
> What's driving me crazy is that running the two
> commands via ExecuteProcess does the first step
> OK, but oggenc fails with an exit code of 1, oper-
> ation not permitted.
>
> If I replace the ExecuteProcess with a call to
> fpSystem, concatenating the CommandString and
>
In our previous episode, Anton Shepelev said:
> In the case of ExecuteProcess() parameter separation
> takes place on FPC side, while with fpSystem() the
> shell is responsible for it.
Executeprocess has two forms. One does parameter separation,
and one not, and directly passes the separated p
On 09/08/2011 01:07 PM, Marco van de Voort wrote:
In our previous episode, Anton Shepelev said:
In the case of ExecuteProcess() parameter separation
takes place on FPC side, while with fpSystem() the
shell is responsible for it.
Executeprocess has two forms. One does parameter separation,
On 09/08/11 19:40, brian wrote:
>
> What are the commands?
>
> mpg321 -q -w "tempfile.wav" "inputfile.mp3"
>
> oggenc -Q --output="outputfile.ogg" "tempfile.wav"
>
> It makes no difference whether or not I use a full pathname for mpg321
> and oggenc, and all the other files are in the current
On 09/08/11 19:52, Andrew Haines wrote:
> You can try something like
> oggenc -Q '--output="outputfile.ogg"' "tempfile.wav"
> to see if that works or if you use TProcess directly I understand that
> options have been added recently to overcome the problem quotes can have
> on the arguments.
or co
In our previous episode, brian said:
> write a batch file for the actual conversions.
>
>
> What are the commands?
>
> mpg321 -q -w "tempfile.wav" "inputfile.mp3"
>
> oggenc -Q --output="outputfile.ogg" "tempfile.wav"
> between the ExecuteProcess and fpSystem calls, I just commented out
> the
On Fri, 9 Sep 2011, Marco van de Voort wrote:
In our previous episode, brian said:
write a batch file for the actual conversions.
What are the commands?
mpg321 -q -w "tempfile.wav" "inputfile.mp3"
oggenc -Q --output="outputfile.ogg" "tempfile.wav"
between the ExecuteProcess and fpSystem c
The quotes are bash commands, they are not part of command line. I
wrote on this topic not long ago, but for TProcess:
http://wiki.lazarus.freepascal.org/Executing_External_Programs#Parameters_which_contain_spaces_.28Replacing_Shell_Quotes.29
--
Felipe Monteiro de Carvalho
__
Andrew Haines:
> I suspect that the oggenc command is failing
> because TProcess behind the scenes is splitting
> --output="outputfile.ogg" into two params.
Yes, indeed. It is also mentioned on the WIKI:
http://wiki.freepascal.org/Executing_External_Programs
(see section
"Para
In our previous episode, Anton Shepelev said:
>TProcess:
>If the command to be executed or any of the
>arguments contains whitespace (space, tab
>character, linefeed character) it should be
>enclosed in single or double quotes.
It should now be possible
On 09/09/2011 04:02 AM, Felipe Monteiro de Carvalho wrote:
The quotes are bash commands, they are not part of command line. I
wrote on this topic not long ago, but for TProcess:
http://wiki.lazarus.freepascal.org/Executing_External_Programs#Parameters_which_contain_spaces_.28Replacing_Shell_Quot
brian wrote:
At least IMHO, having to quote the entire parameter string, not just the
value, is anything but "common whitespace and quote rules".
I am pretty sure that I have had to deal with this in the past on unix,
writing shell or Perl scripts i.e. not in the context of FPC. It might,
in
In our previous episode, michael.vancann...@wisa.be said:
>
> You must remove the quotes around 'outputfile.org':
>
>
> Status:=ExecuteProcess('/path/to/oggenc',['-Q','--output=outputfile.org','tempfile.wav']);
While better, this is afaik not required per se, the executing binary can
strip th
Marco van de Voort:
> It should now be possible to pass the parameters
> to tprocess separated. I don't remember if that
> change made it to 2.4.4 though, otherwise it will
> be in 2.6.0.
Thanks for letting me know.
Is the parameter separation algorithm intentionally
made to split par
On Fri, September 9, 2011 13:40, Marco van de Voort wrote:
> In our previous episode, michael.vancann...@wisa.be said:
>>
>> You must remove the quotes around 'outputfile.org':
>>
>>
>> Status:=ExecuteProcess('/path/to/oggenc',['-Q','--output=outputfile.org','tempfile.wav']);
>
> While better, t
On Fri, Sep 9, 2011 at 1:40 PM, Marco van de Voort wrote:
> While better, this is afaik not required per se, the executing binary can
> strip them also while doing cmdline parsing. It doesn't matter for
> executeprocess.
I wrote that section in the wiki when I had a real issue where the app
did
On 09 Sep 2011, at 13:40, Marco van de Voort wrote:
In our previous episode, michael.vancann...@wisa.be said:
You must remove the quotes around 'outputfile.org':
Status:=ExecuteProcess('/path/to/oggenc',['-Q','--
output=outputfile.org','tempfile.wav']);
While better, this is afaik not re
On Fri, Sep 9, 2011 at 1:46 PM, Anton Shepelev wrote:
> Is the parameter separation algorithm intentionally
> made to split partially quoted arguments like
> '--file="my file"'? If not, then, maybe the
> StringToPPChar() function should be fixed?
It does not split partially quote
On Fri, 9 Sep 2011, Marco van de Voort wrote:
In our previous episode, michael.vancann...@wisa.be said:
You must remove the quotes around 'outputfile.org':
Status:=ExecuteProcess('/path/to/oggenc',['-Q','--output=outputfile.org','tempfile.wav']);
While better, this is afaik not require
In our previous episode, Jonas Maebe said:
> >> output=outputfile.org','tempfile.wav']);
> >
> > While better, this is afaik not required per se, the executing
> > binary can
> > strip them also while doing cmdline parsing.
>
> I don't know any programs other than shells that would do so. FPC
Felipe Monteiro de Carvalho:
> It does not split partially quoted arguments AFAIK
> Not sure about Free Pascal Trunk however.
>
> If FPC Trunk also does not support this, then I'm
> sure a patch to improve StringToPPChar would be
> welcome.
Version 2.4.4 does split, say
--name="Anton Shep
Felipe Monteiro de Carvalho:
> If FPC Trunk also does not support this, then I'm
> sure a patch to improve StringToPPChar would be
> welcome.
I have created issue # 0020279.
Anton
___
fpc-pascal maillist - fpc-pascal@lists.freepascal.org
http://l
2011/9/17 Anton Shepelev :
> Felipe Monteiro de Carvalho:
>
>> If FPC Trunk also does not support this, then I'm
>> sure a patch to improve StringToPPChar would be
>> welcome.
>
> I have created issue # 0020279.
If this must be like shell quoting then you also need to allow quoting
with backsl
In our previous episode, cobines said:
> > I have created issue # 0020279.
>
> If this must be like shell quoting
These routines were never meant to be shell quoting (which is silly since
it is a Runtime lib call, not a shell). I also don't see the point why it
should be expanded to do shell quo
Marco van de Voort:
> So basically this bugreport is about adding *nix
> shell quoting to a routine that was meant to mimic
> WINDOWS API (iow not even Windows shell) be-
> haviour.
I only wanted to fix the parsing of partially quoted
parameters. Of course, it is best to deprecate this
f
33 matches
Mail list logo
