Hi, I haven't yet been able to reproduce this in a minimal example, but I'm trying to place some images within a mixedcolumns environment, and I get an error message I have never seen before: pages > flushing realpage 5, userpage 5 Node 1150 has an attribute list that is free already Node 1406 has an attribute list that is free already Node 1373 has an attribute list that is free already Node 1228 has an attribute list that is free already Node 1195 has an attribute list that is free already Node 1114 has an attribute list that is free already Node 1127 has an attribute list that is free already Node 1074 has an attribute list that is free already Node 2011 has an attribute list that is free already Node 2268 has an attribute list that is free already Node 2116 has an attribute list that is free already Node 2030 has an attribute list that is free already Node 1981 has an attribute list that is free already Node 1953 has an attribute list that is free already Node 3055 has an attribute list that is free already Node 2889 has an attribute list that is free already luatex: ../../../source/texk/web2c/luatexdir/tex/texnodes.w:1985: delete_attribute_ref: Assertion `varmem[(b)].hh.u.B0==attribute_list_node' failed. mtx-context | fatal error: no return code, message: luatex: execution interrupted% This is with context 2014.03.28 15:04 and luatex 0.78.2 on linux 64-bit. The error does not appear with a beta version from january 2014. Any idea where this might come from? Thomas
On 3/29/2014 4:02 PM, Thomas A. Schmitz wrote:
Hi,
I haven't yet been able to reproduce this in a minimal example, but I'm trying to place some images within a mixedcolumns environment, and I get an error message I have never seen before:
pages > flushing realpage 5, userpage 5 Node 1150 has an attribute list that is free already Node 1406 has an attribute list that is free already Node 1373 has an attribute list that is free already Node 1228 has an attribute list that is free already Node 1195 has an attribute list that is free already Node 1114 has an attribute list that is free already Node 1127 has an attribute list that is free already Node 1074 has an attribute list that is free already Node 2011 has an attribute list that is free already Node 2268 has an attribute list that is free already Node 2116 has an attribute list that is free already Node 2030 has an attribute list that is free already Node 1981 has an attribute list that is free already Node 1953 has an attribute list that is free already Node 3055 has an attribute list that is free already Node 2889 has an attribute list that is free already luatex: ../../../source/texk/web2c/luatexdir/tex/texnodes.w:1985: delete_attribute_ref: Assertion `varmem[(b)].hh.u.B0==attribute_list_node' failed.
mtx-context | fatal error: no return code, message: luatex: execution interrupted%
This is with context 2014.03.28 15:04 and luatex 0.78.2 on linux 64-bit. The error does not appear with a beta version from january 2014. Any idea where this might come from?
Ok, we need indeed some example of this so that we can test. It can for instance relate to some prev pointer still not being ok (which can result in such errors). This means that when context looks back at some node that happens to be a head node, we can get weird effects. In due time this will be sorted out but the more tests we have the sooner. Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
On 03/29/2014 11:26 PM, Hans Hagen wrote:
Ok, we need indeed some example of this so that we can test. It can for instance relate to some prev pointer still not being ok (which can result in such errors). This means that when context looks back at some node that happens to be a head node, we can get weird effects. In due time this will be sorted out but the more tests we have the sooner.
Sadly, it is a Heisenbug - I have been unable so far to isolate it. Compilation showed this bug both on linux and OS X - commented parts of the file in question - compilation worked - uncommented line by line - compilation keeps working without error. I'm no longer sure if it is indeed a problem with the mixedcolumns environment or the new bibliography code or a combination of both... Thomas
On 4/1/2014 11:18 AM, Thomas A. Schmitz wrote:
On 03/29/2014 11:26 PM, Hans Hagen wrote:
Ok, we need indeed some example of this so that we can test. It can for instance relate to some prev pointer still not being ok (which can result in such errors). This means that when context looks back at some node that happens to be a head node, we can get weird effects. In due time this will be sorted out but the more tests we have the sooner.
Sadly, it is a Heisenbug - I have been unable so far to isolate it. Compilation showed this bug both on linux and OS X - commented parts of the file in question - compilation worked - uncommented line by line - compilation keeps working without error. I'm no longer sure if it is indeed a problem with the mixedcolumns environment or the new bibliography code or a combination of both...
no, it's one of those ransom issues where a pointer is not set/reset which needs to b esorted out in luatex itself if you put the test file that fails in the crited repos luigi and i can run it and rather easily (less than a day work -) determine where it is Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
On 04/01/2014 11:29 AM, Hans Hagen wrote:
no, it's one of those ransom issues where a pointer is not set/reset which needs to b esorted out in luatex itself
if you put the test file that fails in the crited repos luigi and i can run it and rather easily (less than a day work -) determine where it is
This must be the first time that I'm unhappy because context doesn't fail. I try everything I did the other day to make the document break, but it still compiles like there had never been any problem...
participants (2)
-
Hans Hagen -
Thomas A. Schmitz