Issue 9300 - wrong data displayed in bug form
Summary: wrong data displayed in bug form
Status: CLOSED FIXED
Alias: None
Product: Base
Classification: Application
Component: code (show other issues)
Version: 643
Hardware: PC Windows 2000
: P1 (highest) Trivial (vote)
Target Milestone: OOo 1.1 Beta
Assignee: Frank Schönheit
QA Contact: issues@dba
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2002-11-17 12:38 UTC by Frank Schönheit
Modified: 2006-05-31 14:29 UTC (History)
1 user (show)

See Also:
Issue Type: DEFECT
Latest Confirmation in: ---
Developer Difficulty: ---


Attachments
zipped dBase file to reproduce the bug case (1.60 KB, application/zip)
2002-11-17 12:39 UTC, Frank Schönheit
no flags Details
sample form to reproduce the bug case (8.57 KB, application/octet-stream)
2002-11-17 12:41 UTC, Frank Schönheit
no flags Details

Note You need to log in before you can comment on or make changes to this issue.
Description Frank Schönheit 2002-11-17 12:38:09 UTC
When traveling through a form bound to a dBase data source, sometimes wrong data
is displayed:
Image (I'l attach documents to this bug after submitting) a form with at least a
text control which is bound to some text field of the underlying data source
table/query.
If the contains a non-empty string in one record, and no content in the next
record, then when traveling from this first to the next record the control is
not cleared, but instead the content of the first record is displayed. This
means that at this moment, the form displays content of two different records at
the same time.

Note that this does not have to do with the various bugs about the max. input
text length, where a auto-pilot-generated form does display wrong data when
encountering a too-long text. In this new bug case here there is no limit on the
input text length of the affected controls.
Comment 1 Frank Schönheit 2002-11-17 12:39:01 UTC
Created attachment 3625 [details]
zipped dBase file to reproduce the bug case
Comment 2 Frank Schönheit 2002-11-17 12:41:37 UTC
Created attachment 3626 [details]
sample form to reproduce the bug case
Comment 3 Frank Schönheit 2002-11-17 12:44:43 UTC
the attached files sample.zip and form.sxw can be used to reproduce
the bug case. The former contains a sample.dbf, which should be
extracted and accessed as a new dBase data source named "sample".
Then open the text document - it contains a form working with table
"sample" of data source "sample".
* Travel to the 8th record. Note that the field EMAIL displays a
non-empty string.
* Using the form navigation toolbar, travel to the next record.
=> The EMAIL field still contains the email from the 8th record,
though EMAIL is empty in the 9th record (as you can see when
displaying the sample table in the data source browser)
Comment 4 Frank Schönheit 2002-11-17 12:46:06 UTC
accepting issue. Sounds pretty much like a candidate for OOo 1.0.3
(1.0.2 is out of reach, because too near to be released, AFAIK).
This bug renders form functionality pretty useless (because too
unreliable).
Comment 5 Frank Schönheit 2002-11-29 14:23:50 UTC
changing target milestone to OOo 1.1 Beta
Comment 6 alex.thurgood 2002-12-21 16:05:23 UTC
This is similar to another bug I've discovered in the grid/table view
when carrying out a search using the binocular/dialog on data in
1.0.1. The record found is correct, but the corresponding cell shown
actually corresponds to data higher up in the table. Switching cells
manually doesn't alter this behaviour.

System Linux Mandrake 9, Radeo ATI Rage Mobility Pro graphics card (in
 case it's graphics related, but it seems to be the redrawing of the
grid widget/object that is at fault.
Comment 7 Frank Schönheit 2003-01-06 09:32:44 UTC
Alex, this bug here was not reproducible in the latest builds - I am
actually not sure if this held for 643, too. Can you please try your
bug in an 643 build, and if it happens there, too, submit a new issue?
This one here is likely to be set to WORKSFORME or FIXED. Thanks.
Comment 8 Frank Schönheit 2003-01-09 16:09:00 UTC
cannot reproduce in latest inhouse build (SRX644c). Because we changed
a lot in this area in the meantime, I consider this FIXED, though it's
not really satisfying that I don't know what went wrong there ....
Comment 9 Frank Schönheit 2003-05-20 11:40:15 UTC
closing
Comment 10 hans_werner67 2004-02-02 12:30:23 UTC
change subcomponent to 'none'