Bug in InvokePostscriptDelegate function
Posted: 2008-03-28T15:17:44-07:00
Bug in InvokePostscriptDelegate in coders\ps.c and coders\pdf.c - ImageMagick 6.4.0-0
I reviewed earlier versions of ImageMagick and in Ghostscript the files API.htm and ierrors.h.
--------------------------------------------------------------------------------
Source ImageMagick 6.3.1-2
--------------------------------------------------------------------------------
Source ImageMagick 6.4.0-0
--------------------------------------------------------------------------------
API.htm
--------------------------------------------------------------------------------
Source Suggested ImageMagick 6.4.0-0
--------------------------------------------------------------------------------
My conclusion was that:
(status == 0) - No errors
(status == -101) - "quit" has been executed (e_Quit). No error
Other values - Are errors or not apply
It removed the line return(SystemCommand(verbose,command) != 0 ? MagickTrue : MagickFalse); because at this point if a mistake is due to the contents of the file and run this line is useless because the error occurs again.
The bug existed in version 6.3.1-2, but was masked by the fact of not being tested the return of InvokePostscriptDelegate with the changes in version 6.4.0-0 this bug is evidence to run twice Ghostscript.
I reviewed earlier versions of ImageMagick and in Ghostscript the files API.htm and ierrors.h.
--------------------------------------------------------------------------------
Source ImageMagick 6.3.1-2
Code: Select all
static MagickBooleanType InvokePostscriptDelegate(const MagickBooleanType verbose,const char *command)
{
...
if ((status == 0) || (status <= -100))
{
(void) LogMagickEvent(CoderEvent,GetMagickModule(),
"Ghostscript returns status %d, exit code %d",status,code);
return(MagickFalse);
}
return(MagickTrue);
#else
return(SystemCommand(verbose,command) != 0 ? MagickTrue : MagickFalse);
#endif
}Source ImageMagick 6.4.0-0
Code: Select all
static MagickBooleanType InvokePostscriptDelegate(const MagickBooleanType verbose,const char *command)
{
...
if ((status == 0) || (status <= -100))
{
(void) LogMagickEvent(CoderEvent,GetMagickModule(),
"Ghostscript returns status %d, exit code %d",status,code);
return(SystemCommand(verbose,command) != 0 ? MagickTrue : MagickFalse);
}
return(MagickTrue);
#else
return(SystemCommand(verbose,command) != 0 ? MagickTrue : MagickFalse);
#endif
}API.htm
Code: Select all
0 No errors
e_Quit (-101) "quit" has been executed. This is not an error. gsapi_exit() must be called next.
e_interrupt (-6) The polling callback function returned a negative value, requesting Ghostscript to abort.
e_NeedInput (-106) More input is needed by gsapi_run_string_continue(). This is not an error.
e_Info (-110) "gs -h" has been executed. This is not an error. gsapi_exit() must be called next.
< 0 Error
<= -100 Fatal error. gsapi_exit() must be called next.Source Suggested ImageMagick 6.4.0-0
Code: Select all
static MagickBooleanType InvokePostscriptDelegate(const MagickBooleanType verbose,const char *command)
{
...
if ((status == 0) || (status = -101))
return(MagickFalse);
(void) LogMagickEvent(CoderEvent,GetMagickModule(),
"Ghostscript returns status %d, exit code %d",status,code);
return(MagickTrue);
#else
return(SystemCommand(verbose,command) != 0 ? MagickTrue : MagickFalse);
#endif
}My conclusion was that:
(status == 0) - No errors
(status == -101) - "quit" has been executed (e_Quit). No error
Other values - Are errors or not apply
It removed the line return(SystemCommand(verbose,command) != 0 ? MagickTrue : MagickFalse); because at this point if a mistake is due to the contents of the file and run this line is useless because the error occurs again.
The bug existed in version 6.3.1-2, but was masked by the fact of not being tested the return of InvokePostscriptDelegate with the changes in version 6.4.0-0 this bug is evidence to run twice Ghostscript.