With LifeLines 3.0.41, compiled on Linux Mandrake 10.1,
llines crashe with this message (translation, I use
french):
Fatal Error
in file <keytonod.c> in line 1395
I tried with an other Gedcom file, and the problem is
strictly the same.
Hope this help.
Logged In: YES
user_id=1213099
Sorry, I was not logged in. My name in Source Force : hollyth.
Logged In: YES
user_id=1213099
Same problem, but at a different line, with
lifelines-3.0.42-2 compiled on Linux Mandrake 10.1:
Fatal Error
in file <keytonod.c> in line 1420.
Logged In: YES
user_id=1195173
Would you mind retrying with lifelines-3.0.43 release (which
I just made a couple minutes ago, attempting to fix this)?
Logged In: YES
user_id=1213099
I just tried with lifelines-3.0.43, and I obtained the same:
Fatal error
in file <keytonod.c> in line 1424
Logged In: YES
user_id=1195173
Do you have a reproducible test case on a database that you
can afford to send me?
or else, can you invoke lines with the -d switch and get a
coredump (-d will suppress the signal handlers) and send me
a copy of the call stack?
Or else, describe what you were doing before that happened?
Logged In: YES
user_id=1213099
I just try the original report ps-pedigree to see if the
translation cause problem. So the version furnished whith
lifelines cause the same error.
My replies to this report:
Select a person: Jean /Martin (name for
example)
I reply "i" to accept Lifelines selection
Max generation per chart: 7
Max depth of chart: 15
Name of ouput file: famille.ps
When I validate, the program run and:
ERREUR FATALE
x x
x x dans le fichier <keytonod.c> la ligne 1424
x x
x xBase ouverte: ./bases/famille
Also something I did'nt see on first time:
ASSERT chec
If you cannot reproduce this, I can send you my Gedcom file
for your test. I just need your promess to use it
exclusively for this test.
Logged In: YES
user_id=1195173
Thank you for the offer of the GEDCOM, but I just reproduced
it a moment ago so I won't need it.
Logged In: YES
user_id=1195173
I believe the root of this is the memset in
put_node_in_cache, which is clearing record info.
Logged In: YES
user_id=1195173
Delivery to the following recipients failed.
hollyth@users.sourceforge.net
I was emailing you to ask you if I can send you a new
tarball to test -- can you send me email (just send to my
sourceforge email address, I think mine works) and let me
know if I can email you a new test distribution tarball.
Logged In: YES
user_id=1195173
I hope that this is fixed in the (new) 3.0.44 beta release?
Logged In: YES
user_id=1213099
I just tried it this morning: It works!
I made test with my french version of ps-pedigree and
book-latex.
Thanks for your work, very efficient.
Logged In: YES
user_id=1195173
Great!
Closing as Fixed.