he's still able to trigger such errors.
see https://github.com/pure-data/pure-data/issues/477 and a proposed fix: https://github.com/pure-data/pure-data/pull/492
Le lun. 24 sept. 2018 à 11:45, Roman Haefeli reduzent@gmail.com a écrit :
Hi
0.49test4 fixed the regression for me. I don't see any 'INVALID COMMAND NAME' errors anymore. Claude told me in #dataflow, though, that he's still able to trigger such errors.
Thanks, Roman
On Sun, 2018-09-23 at 09:40 +0200, Roman Haefeli wrote:
Hey Claude
It seems you already did the hard work. Sorry for not having checked more carefully. I'm glad attention is already paid to the issue.
Roman
On Sat, 2018-09-22 at 23:29 +0100, Claude Heiland-Allen wrote:
Hi Roman,
On 22/09/18 23:19, Roman Haefeli wrote:
Hi
Sorry, need to report something probably more serious. Some of my patches trigger messages like the following:
(Tcl) INVALID COMMAND NAME: invalid command name ".x558795b0c400.c" while executing ".x558795b0c400.c itemconfigure 558795b682c0LABEL -font {{DejaVu Sans Mono} -11 bold}" ("uplevel" body line 10) invoked from within "uplevel #0 $docmds"
Side effects are that sometimes the label of [cnv] is not updated. Or the canvas stays blank after dynamically creating stuff and objects are only shown after closing the window and opening it again.
You might try my (unfortunately too slow for anything other than debugging) attempt here: https://github.com/claudeha/pure-data/tree/gui-tcl-robustness
It may reduce the impact of the error, or perhaps show further errors...
I don't have a small patch yet that triggers the problem, but will try to figure one out.
I reported something possibly-similar here: https://github.com/pure-data/pure-data/issues/488
toplevel patch contains a non-GOP abstraction which contains a GOP abstraction which sets iemgui properties on loadbang results in invalid command name tcl error
BTW.: this is with 0ee284ecfb5215
Claude_______________________________________________
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list