[NTG-context] limit of index entries?

Hans Hagen j.hagen at xs4all.nl
Sun Aug 2 10:18:47 CEST 2020

On 8/1/2020 6:08 PM, Henning Hraban Ramm wrote:
>> Am 01.08.2020 um 14:10 schrieb Hans Hagen <j.hagen at xs4all.nl>:
>> On 8/1/2020 1:54 PM, Henning Hraban Ramm wrote:
>>> Hi,
>>> my aforementioned person index has more than 600 entries. If I get over ~500, ConTeXt (MkIV, several versions) breaks with
>>> ! error (push_node): stack overflow
>>> mtx-context     | fatal error: return code: 256
>>> Is there a hard limit? Can I change it?
>> it might relate to protrusion (which needs to enter nested boxes to determine an edge) but how that then can relate to the number of entries puzzles me, unless we pop too less ... in which case i need to trace (with a mwe)
> I can’t reproduce it with a simple example like
> \setupalign[hanging]
> \starttext
> \placeregister[index][criterium=all]
> \dorecurse{10000}{\par\strut\index{Test\recurselevel}}
> \stoptext
> \recurselevel is empty in that context, \expanded didn’t help, but I guess we need different entries.
> Find attached the actual data, reduced to the minimum setup – the problem is indeed \setupalign[hanging].
> I guess I can omit "hanging" for the index, but of course if would be nice if it would work.
it has to do with how you do that index:

- you do a lot of \index but you do that in vmode
- do, they get collected for the next paragraph
- that is done in a robust way which involves packing
- eventually you flush with a \strut
- so we get deeply nested boxes
- and that means trouble for protrusion pre-analysis

now, I can try to avoid it but we then can have side effects but we'll see

can you try to avoid collecting:


make sure no spaces get introduced


                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl

More information about the ntg-context mailing list