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 343176 - [xls] INDEX problem
[xls] INDEX problem
Status: RESOLVED OBSOLETE
Product: Gnumeric
Classification: Applications
Component: Analytics
git master
Other All
: High major
: ---
Assigned To: Morten Welinder
Jody Goldberg
: 621350 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2006-05-28 11:29 UTC by Ivan Wong
Modified: 2018-05-22 13:16 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
DataValDealer.xls (33.50 KB, application/vnd.ms-excel)
2006-05-28 11:33 UTC, Ivan Wong
Details

Description Ivan Wong 2006-05-28 11:29:48 UTC
Loading the attached XLS file gives the following messages:

Reading file:///tmp/4.xls
Excel 97 +
Employees!L2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!L2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!M2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!M2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!N2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!N2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!O2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!O2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!P2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!P2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!Q2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!Q2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!R2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!R2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!AC2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!AC2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!AD2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!AD2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!AE2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!AE2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!AF2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!AF2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!AG2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!AG2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!AH2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!AH2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Employees!AI2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3A
Employees!AI2 : Hmm, ptgAttr of type 0 ??
I've seen a case where an instance of this with flag A and another with flag 3
bracket a 1x1 array formula.  please send us this file.
Flags = 0x3
Comment 1 Ivan Wong 2006-05-28 11:33:37 UTC
Created attachment 66362 [details]
DataValDealer.xls
Comment 2 Jon Kåre Hellan 2006-10-18 17:17:23 UTC
No warnings on load, but the center table is computed incorrectly. All non-empty cells get 'Date', in XL they get 'Ann', 'Ben', 'Ed', etc.
Formula example (Emplyees!L2): 
=if(row($A2:$A8)-row(A2)+1>count(T2:T8),"",
 index($A:$A,small(T2:T8,row(indirect("1:"&rows(A2:A8))))))
Comment 3 Jody Goldberg 2006-10-26 12:54:59 UTC
Another eval monstrosity hiding under the ptg=0 warning.  I suppose it makes sense.  The ptg=0 seemed to mark array functions in single cells.  Now that we have most array iteration working these pieces stand out in sharp relief.
Comment 4 Jody Goldberg 2007-07-27 12:17:45 UTC
I've boiled down the test case a bit.  The problem actually seems completely different than expected.  Gnumeric defines SMALL/LARGE as =small(n1,n2,n3....k)
whereas XL and OOo use =small(n,k) which allows for implicit iteration and intersection on 'k'.

Patched for 1.7.12 and pile of new test cases added.

Fixing INDEX will take a bit more time.
Comment 5 Andreas J. Guelzow 2009-08-01 22:40:40 UTC
According to Comment #4 small and large are fixed. 

What is wrong with INDEX?
Comment 6 Morten Welinder 2010-06-10 14:14:13 UTC
I would take a guess that the problem is that INDEX has definition

    gnumeric_index (GnmFuncEvalInfo *ei, int argc, GnmExprConstPtr const *argv)

i.e., it gets expressions, not values.  As such, it cannot work with implicit
iteration.  And INDEX' special handling of GNM_EXPR_OP_SET will probably make
it hard to fix.
Comment 7 Morten Welinder 2010-06-12 12:09:39 UTC
*** Bug 621350 has been marked as a duplicate of this bug. ***
Comment 8 GNOME Infrastructure Team 2018-05-22 13:16:06 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to GNOME's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.gnome.org/GNOME/gnumeric/issues/58.