[omniORB] omniidl assert error about comments after upgrade from 2.8 to 3

Richard Gruet rgruet@ina.fr
Thu, 16 Mar 2000 18:10:14 +0100


--------------727CD9ACA459686B3379F50C
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi (more likely intended to Duncan or Dave),

I just upgraded today from omniORB 2.80 to omniORB 3 (compiled without
pb on NT 4.0, using the ftp snapshots omniORB3-20000316_tar.gz and
omniORBpy-20000316_tar.gz).

I had to recompile the stubs  from my application IDLs. While doing
that, I got this assertion failed:

    >omniidl -N -CmyOutDir -bpython
d:\Divan2\main\src\INA\idl\DataModel_.idl
    Assertion failed: mostRecent_, file idlast.cc, line 128

* Clues:

- The file DataModel_.idl results from a preprocessing of my main IDL
file (hence the -N option), to avoid the problems related to #includes
(we discussed this in the past), while preserving comments and
discarding line# directives. This preprocessing was performed using the
command:
            omnicpp -C -P DataModel.idl > DataModel_.idl
    (BTW, thanks for omnicpp's name change ;-)

- I use a lot of comments in DataModel_.idl. They are in 'JavaDoc'
style, as advised by the
     Orbacus HIDL tool  (http://ooc.com/ob/hidl.html) . This is a small
excerpt:

/** A representative frame extracted from  a video segment. */
interface KeyframeEventCorba : AEDIEventCorba
{
  /** The filename for the keyframe */
  attribute Frame frameref;

  /** The list of extracted faces */
  readonly attribute KeyfaceCorbaList faces;

  /** Insertion of a detected face.
      @param box The bounding box where the face has been detected.
      @param coefs The wavelet packet variances, to be used for face
recognition.
      @return The incremented number of detected faces
   */
  long addFace(in BoundingBox box,in Wavelets coefs) raises
(NotApplicable);

   //...............

  /** Histogram values for 12 regions, 16 gray scale intensities and 16
colors */
  attribute RegionHistogram rhistogram;
};

If this is necessary, I can provide you with the complete DataModel_.idl
file.

Thanks for any help,
Cheers,

Richard Gruet

PS : I got an even more serious crash (application error) with my Python
programs which seems related to reference count management or
destruction of objrefs. I'm trying to isolate the problem, more on this
later...
PPS: The echo example works fine ;-)
PPS: Despite all this, omniORB is great !




--------------727CD9ACA459686B3379F50C
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Hi (more likely intended to Duncan or Dave),
<p>I just upgraded today from omniORB 2.80 to omniORB 3 (compiled without
pb on NT 4.0, using the ftp snapshots omniORB3-20000316_tar.gz and omniORBpy-20000316_tar.gz).
<p>I had to recompile the stubs&nbsp; from my application IDLs. While doing
that, I got this assertion failed:
<p>&nbsp;&nbsp;&nbsp; >omniidl -N -CmyOutDir -bpython d:\Divan2\main\src\INA\idl\DataModel_.idl
<br>&nbsp;&nbsp;&nbsp; Assertion failed: mostRecent_, file idlast.cc, line
128
<p>* Clues:
<p>- The file DataModel_.idl results from a preprocessing of my main IDL
file (hence the -N option), to avoid the problems related to #includes
(we discussed this in the past), while preserving comments and discarding
line# directives. This preprocessing was performed using the command:
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
omnicpp -C -P DataModel.idl > DataModel_.idl
<br>&nbsp;&nbsp;&nbsp; (BTW, thanks for omnicpp's name change ;-)
<p>- I use a lot of comments in DataModel_.idl. They are in 'JavaDoc' style,
as advised by the
<br>&nbsp;&nbsp;&nbsp;&nbsp; Orbacus HIDL tool&nbsp; (<a href="http://ooc.com/ob/hidl.html">http://ooc.com/ob/hidl.html)</a>
. This is a small excerpt:
<p><font size=-1>/** A representative frame extracted from&nbsp; a video
segment. */</font>
<br><font size=-1>interface KeyframeEventCorba : AEDIEventCorba</font>
<br><font size=-1>{</font>
<br><font size=-1>&nbsp; /** The filename for the keyframe */</font>
<br><font size=-1>&nbsp; attribute Frame frameref;</font>
<p><font size=-1>&nbsp; /** The list of extracted faces */</font>
<br><font size=-1>&nbsp; readonly attribute KeyfaceCorbaList faces;</font>
<p><font size=-1>&nbsp; /** Insertion of a detected face.</font>
<br><font size=-1>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; @param box The bounding
box where the face has been detected.</font>
<br><font size=-1>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; @param coefs The wavelet
packet variances, to be used for face recognition.</font>
<br><font size=-1>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; @return The incremented
number of detected faces</font>
<br><font size=-1>&nbsp;&nbsp; */</font>
<br><font size=-1>&nbsp; long addFace(in BoundingBox box,in Wavelets coefs)
raises (NotApplicable);</font>
<p><font size=-1>&nbsp;&nbsp; //...............</font>
<p><font size=-1>&nbsp; /** Histogram values for 12 regions, 16 gray scale
intensities and 16 colors */</font>
<br><font size=-1>&nbsp; attribute RegionHistogram rhistogram;</font>
<br><font size=-1>};</font>
<p>If this is necessary, I can provide you with the complete DataModel_.idl
file.
<p>Thanks for any help,
<br>Cheers,
<p>Richard Gruet
<p>PS : I got an even more serious crash (application error) with my Python
programs which seems related to reference count management or destruction
of objrefs. I'm trying to isolate the problem, more on this later...
<br>PPS: The echo example works fine ;-)
<br>PPS: Despite all this, omniORB is great !
<br>&nbsp;
<br>&nbsp;
<br>&nbsp;</html>

--------------727CD9ACA459686B3379F50C--