GNOME Bugzilla – Bug 363871
Critical when opening function guru
Last modified: 2007-11-20 07:29:22 UTC
New gnumeric A1: =isna(getenv(A2),"",getenv(A2)) finish editing cell go back to A1 Insert/Function ** (gnumeric:3889): CRITICAL **: dialog_formula_guru_load_string: assertion `okay' failed (gnumeric:3889): Gtk-CRITICAL **: gtk_tree_store_set: assertion `VALID_ITER (iter, tree_store)' failed (gnumeric:3889): Gtk-CRITICAL **: gtk_tree_store_get_value: assertion `iter->stamp == GTK_TREE_STORE (tree_model)->stamp' failed (gnumeric:3889): GLib-GObject-WARNING **: gtype.c:3337: type id `0' is invalid (gnumeric:3889): GLib-GObject-WARNING **: can't peek value table for type `<invalid>' which is not currently referenced Program received signal SIGSEGV, Segmentation fault. Backtrace at segv:
+ Trace 77448
Found an unhandled failure case in dialog_formula_guru_load_fd. This fixes the crash. We now get ** (gnumeric:13688): CRITICAL **: dialog_formula_guru_load_string: assertion `okay' failed ** (gnumeric:13688): CRITICAL **: dialog_formula_guru_load_string: assertion `okay' failed and a blank guru
The warnings are there because you are supplying two more arguments to ISNA than it wants.
Exactly. I made a mistake. (Should have been =IF(isna(getenv.... ). Entered the function guru to fix it, and it blew up. This is what the guru if *for*
This problem has been fixed in the development version (well I have the fix in my tree and will commit as soon as 1.7.90 has been released since the distribution build is in progress). The fix will be available in the next major software release after 1.7.90. Thank you for your bug report.