Bug 540211 – r.phyper hang

After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.

Bug 540211 - r.phyper hang


Summary:	r.phyper hang


Status:	RESOLVED FIXED

Product:	Gnumeric
Classification:	Applications
Component:	Analytics
Version:	git master
Hardware:	Other All

Importance:	Normal critical
Target Milestone:	---
Assigned To:	Morten Welinder
QA Contact:	Jody Goldberg

URL:
Whiteboard:

Depends on:
Blocks:

Reported:	2008-06-25 20:31 UTC by sum1
Modified:	2008-06-26 13:57 UTC

See Also:
GNOME target:	---
GNOME version:	---

Description sum1 2008-06-25 20:31:49 UTC

Version: r16670
OS: Ubuntu Hardy

Steps to reproduce:
- In A1, enter "=r.phyper(1.0e+65,,,)" (no quotes)


Backtrace:

Program received signal SIGINT, Interrupt.

+ Trace 201386

Thread 3061381728 (LWP 16708)

#0 pdhyper
at mathfunc.c line 3333
#1 phyper
at mathfunc.c line 3377
#2 gnumeric_r_phyper
at functions.c line 992
#3 function_call_with_exprs
at func.c line 1247
#4 gnm_expr_eval
at expr.c line 1321
#5 gnm_expr_top_eval
at expr.c line 2839
#6 gnm_cell_eval_content
at dependent.c line 1455
#7 gnm_rendered_value_new
at rendered-value.c line 210
#8 gnm_cell_render_value
at cell.c line 469
#9 row_calc_spans
at cellspan.c line 411
#10 sheet_range_bounding_box
at sheet.c line 2250
#11 scg_redraw_range
at sheet-control-gui.c line 147
#12 sc_redraw_range
at sheet-control.c line 89
#13 sheet_redraw_partial_row
at sheet.c line 1010
#14 sheet_cell_calc_span
at sheet.c line 1102
#15 sheet_cell_set_text
#16 cmd_set_text_redo
at commands.c line 851
#17 command_push_undo
at commands.c line 742
#18 cmd_set_text
#19 wbcg_edit_finish
at wbc-gtk-edit.c line 222
#20 cb_gee_key_press_event
at gnumeric-expr-entry.c line 513

Comment 1 Andreas J. Guelzow 2008-06-25 21:40:43 UTC

I am not sure whether this is really a "hang".

In R if I enter 
phyper(1.0e+65,0,0,0);
I finally give up after it used 100% of a core for a long time.

Comment 2 Morten Welinder 2008-06-26 01:44:58 UTC

It's a hang.  1e65 - 1 is still 1e65 with double.  But since there are
only 0 red balls, anything beyond 0 is going to give the result 1.

This problem has been fixed in the development version. The fix will be available in the next major software release. Thank you for your bug report.

Comment 3 Morten Welinder 2008-06-26 01:54:11 UTC

Upstreamed.

Comment 4 Andreas J. Guelzow 2008-06-26 04:16:09 UTC

anything beyond 0 is going to give the result 1 ?

x=0 is also guaranteed so that should give 1 too.

Comment 5 Morten Welinder 2008-06-26 13:57:07 UTC

The condition I put in is (x >= NR), so yes when there are no red balls in
the pot and you ask if picking 0 (or 42 or 1e65) balls will give you 0 or
less red balls, the answer is "certainly!"