help-3dldf
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[help-3dldf] Fwd: 3DLDF errors


From: Laurence Finston
Subject: [help-3dldf] Fwd: 3DLDF errors
Date: Sat, 12 Feb 2005 22:42:58 +0100
User-agent: IMHO/0.98.3+G (Webmail for Roxen)

------ Forwarded message -------

From: "Laurence Finston" <address@hidden>
To: Martijn van Manen  
Date: Sat, 12 Feb 2005 22:41:38 +0100

Another way of debugging is to use the `--bison-trace' option to `3dldf'.  If
you use this, it sets the variable `yydebug' to 1 in `main()', and a flood of
information will be output to `stderr' (or maybe `stdout'), including the
rules according to which the parser states are reduced.  This will tell you
which rule is the culprit.  

Segmentation faults are usually caused by dereferencing an uninitialized
pointer, or one that doesn't point to the correct type of object.  I don't
know without looking where this could be happening in the case you found.  I
suspect it has something to do with the rules for 
the `point' expression hierarchy.  As you may know, data types in MF/MP and
3DLDF are  organized in a hierarchy of primaries, secondaries, tertiaries, and
expressions.  Another factor is that I don't have control of the way Bison
creates states from the rules that I write, and so don't have control of how
states are reduced.  It may be easy to fix this problem, or it may be
impossible.  I won't know until I try, and I won't get to it this week.  So in
the meantime, please try to work around it.

Laurence



reply via email to

[Prev in Thread] Current Thread [Next in Thread]