#19 Check return codes everywhere

[Markus Elfring]

Some checks for return codes are missing.

Examples:
Would you like to add more error handling for return values from "malloc" like in the function "rfbSendSmallRectEncodingCoRRE" and from "fclose" in the function "rfbLoadConsoleFont"?
http://libvncserver.cvs.sourceforge.net/libvncserver/libvncserver/libvncserver/corre.c?revision=1.6&view=markup
http://libvncserver.cvs.sourceforge.net/libvncserver/libvncserver/libvncserver/font.c?revision=1.6&view=markup

[Johannes Schindelin]

Logged In: YES
user_id=27066
Originator: NO

Hi,

thank you! Yes, for malloc I would like it! Although I am not so sure about fclose(). Should we really error out in that case?

Ciao,
Dscho

[Markus Elfring]

Logged In: YES
user_id=572001
Originator: YES

Yes.

Do you know more places with unchecked function calls in the source files?

[Johannes Schindelin]

Logged In: YES
user_id=27066
Originator: NO

Hi,

no, not unless I would do a code review.

Wanna help?

Ciao,
Dscho

[Markus Elfring]

Logged In: YES
user_id=572001
Originator: YES

More open issues can be seen in the function "start_client_info_sock".
http://libvncserver.cvs.sourceforge.net/libvncserver/libvncserver/x11vnc/connections.c?revision=1.30&view=markup
- strdup
- fprintf
- write

[Markus Elfring]

Logged In: YES
user_id=572001
Originator: YES

http://libvncserver.cvs.sourceforge.net/libvncserver/libvncserver/libvncserver/main.c?revision=1.44&view=markup
Are you going to add checks for returned error codes?
Examples:
- LOCK, UNLOCK
- calloc
- fflush
- vfprintf

[Johannes Schindelin]

Logged In: YES
user_id=27066
Originator: NO

I am sorry. I do not have time right now to follow up on such unspecific
issues. (If you create a patch, I will look at it, though.)

[Markus Elfring]

Logged In: YES
user_id=572001
Originator: YES

What is "unspecific" about a recommended completition of the error handling?
I will not develop a patch before the general acceptance and approach for the proposed updates is clear.

Would you like to reduce the efforts for error code checking by a class hierarchy?
http://dietmar-kuehl.de/mirror/c++-faq/exceptions.html#faq-17.1
http://cexcept.sourceforge.net/

Are you going to integrate your source files into a Subversion repository (at SourceForge)?

[Johannes Schindelin]

Logged In: YES
user_id=27066
Originator: NO

> What is "unspecific" about a recommended completition of the error
> handling?

It is not a patch. It is some handwaving "there could be somewhere some
sort of return thing".

There are clearly places, where error values are returned, and there are
places that use these error values.

> I will not develop a patch before the general acceptance and approach for
> the proposed updates is clear.

Guess what: I will not tell you beforehand if I take your patch or not,
since I know nothing about the style, quality, and completeness of your
code.

> Would you like to reduce the efforts for error code checking by a class
> hierarchy?
> http://dietmar-kuehl.de/mirror/c++-faq/exceptions.html#faq-17.1
> http://cexcept.sourceforge.net/

I will not force acceptance of these things on users of LibVNCServer.
IOW: No.

> Are you going to integrate your source files into a Subversion repository
> (at SourceForge)?

The sources are tracked with Sourceforge's CVS service, although I use Git
to manage it locally.