Issue 35191 - Not all, only ~11 rows displayed in MailMerge / Table window
Summary: Not all, only ~11 rows displayed in MailMerge / Table window
Status: CONFIRMED
Alias: None
Product: Base
Classification: Application
Component: code (show other issues)
Version: OOo 1.1.3
Hardware: PC All
: P4 Trivial with 1 vote (vote)
Target Milestone: ---
Assignee: AOO issues mailing list
QA Contact:
URL:
Keywords: oooqa
: 47670 48418 56583 60054 72405 74710 75434 97450 101509 (view as issue list)
Depends on:
Blocks:
 
Reported: 2004-10-08 12:29 UTC by tuharsky
Modified: 2013-02-07 22:40 UTC (History)
3 users (show)

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


Attachments
what I mean (3.04 KB, image/png)
2004-10-12 08:32 UTC, tuharsky
no flags Details

Note You need to log in before you can comment on or make changes to this issue.
Description tuharsky 2004-10-08 12:29:53 UTC
Get the attachment from issue 35178. Add the spreadsheet to Data Sources. Create
an empty text document and add field pointing to any column of the sheet.

Now go to MailMerge. In the selection window, there are only 11 rows of the
Spreadsheet visible. If U click at "forward to the end" button, remaining rows
appear.

If U use a sort function, again only 13 rows appear and only after Forward to
end envisibles remaining rows.
Comment 1 marc.neumann 2004-10-11 14:53:12 UTC
Hi,

yes that's right.
First you see only a couple of records f.e. 11. And then after scrolling you see
all records.
You don't want to wait until f.e. 100.000 records are displayed in bug tables.

So I close this issue as "wont fix"

Bye Marc
Comment 2 marc.neumann 2004-10-11 14:53:28 UTC
close
Comment 3 tuharsky 2004-10-12 07:25:47 UTC
Well, it is well-thought, I agree
.
However, shouldn't the user be noticed about this? Maybe some cute text on the
bottom or so could do.. Or some note that would display when mouse the first
time affects the fields selection window.
Comment 4 Frank Schönheit 2004-10-12 07:47:46 UTC
Well, there's an asterik behind the record number. The documentation states (to
my best knowledge) that this means the record count is incomplete. However, I
admit that this is pretty non-obvious.
A tooltip could do, but where? What do you mean by "fields selection window"?
Comment 5 tuharsky 2004-10-12 08:32:06 UTC
Created attachment 18295 [details]
what I mean
Comment 6 tuharsky 2004-10-12 08:55:23 UTC
When U go to MailMerge window, this part is what I mean. The tooltip or so could
cover also the scroll and other controls.

Hmm, what about other small trick. The scrollbar, that is used to scroll down,
changes its size by default to reflect length of the list. If we changed the
size to make it smaller by default, the user would naturally expect more records
and longer list. Of course, accessing the lower parts of list would change the
size to appropriate. Of course, the fader should shrink only if there are really
more than 11 records avaiable to display ;o)

Additional improvement: a "full list" button somewhere near. User usually
dosen't work with 100000's of records as You say, but some 100's usually do
(when we say "MailMerge" in Office product, we're almost speaking about
generating mail labels in some company). I agree that the 100000's of records
would take long time, but 100's shouldn't. This way ("Full list" button), user
is acknowledged that the list ISN'T FULL and that HE SHOULD TAKE SPECIAL ACTION
TO DISPLAY FULL LIST -thus it is much more OK for him.

The actual implementation of the window is not very lucky anyway. If user isn't
aware, he dosen't feel comfortably with only 11 records showed. He feels like
there are no more records, or that program is eating his data. If he does any
operation (sort or so), the records are reorganised, but again only 11 appear.

The may be best possible solution? Listing is real-time rendered anytime the
Mail-merge window is opened. Only first visible records are displayed, but mouse
cursor shows activity and the scrollbar 's size is changing contineously to
reflect real amount of records. Cursor is active (not blocked) and user can do
any action he wants. I understand that such solution may be the most difficult
to code.

What do You mean?
Comment 7 tuharsky 2004-12-08 14:14:37 UTC
Hmm, I don't think the behaviour is obvious for user. He should be noticed by
some logical, self-documenting and meaningful way. The sole asterisk dosen't do
the job.

Please consider.

Some ways to do are commented in my previous post althought I don't present them
as the best possible.
The size of scrollbar might be good and simple way to go, and pop-up tooltip "11
of 111000 records displayed" that would fade out after first click on whatever
button might be also sympathic.
Comment 8 Frank Schönheit 2004-12-08 14:32:37 UTC
The solution to display the "11000" (or whatever) isn't possible since you
cannot know this number before visiting every record - that's an inherent
problem of databases. We could guestimate this number, at least.

In SO 5.2, we were constantly counting the records in the background, thus the
user saw a permanent increase in the record number. I agree that this was
probably the best possible solution. Unfortunately, we had to kick it out for
technical reasons, and did not yet find time to re-introduce.
Comment 9 tuharsky 2004-12-08 14:46:54 UTC
Well, It may be the best solution, however user should be able to disable it in
Options to avoid CPU overuse in multiuser terminal systems.

So the LIST ALL button seems self-documenting, transparent and relatively easy
to accomplish, isn't it?

Moreover, if there was time to implement background listing, the button isn't in
contradiction with that; user could block the auto-background-listing and still
use the button if he wishes. If auto-background-listing was enabled, pressing
the button could display some "In progress" kinda note.
Comment 10 flibby05 2005-04-01 19:05:59 UTC
defect -> enhancement
Comment 11 marc.neumann 2005-04-28 13:25:36 UTC
Hi,

I reassign this issue to User Experience for evaluating.

Bye Marc
Comment 12 marc.neumann 2005-04-28 13:34:38 UTC
*** Issue 47670 has been marked as a duplicate of this issue. ***
Comment 13 marc.neumann 2005-05-03 08:58:42 UTC
changing summary
Comment 14 marc.neumann 2005-05-03 08:59:17 UTC
*** Issue 48418 has been marked as a duplicate of this issue. ***
Comment 15 marc.neumann 2005-10-25 08:42:21 UTC
*** Issue 56583 has been marked as a duplicate of this issue. ***
Comment 16 marc.neumann 2006-01-05 09:12:23 UTC
*** Issue 60054 has been marked as a duplicate of this issue. ***
Comment 17 marc.neumann 2007-02-21 14:17:39 UTC
*** Issue 74710 has been marked as a duplicate of this issue. ***
Comment 18 marc.neumann 2007-03-16 10:22:00 UTC
*** Issue 75434 has been marked as a duplicate of this issue. ***
Comment 19 Rainer Bielefeld 2007-03-24 09:39:52 UTC
*** Issue 72405 has been marked as a duplicate of this issue. ***
Comment 20 marc.neumann 2007-10-22 13:34:56 UTC
*** Issue 82797 has been marked as a duplicate of this issue. ***
Comment 21 drewjensen.inbox 2008-12-21 18:28:11 UTC
*** Issue 97450 has been marked as a duplicate of this issue. ***
Comment 22 bero 2008-12-21 19:41:34 UTC
There is another problem with this: If a user clicks the top left corner, he
expects to select the entire table (e.g. to generate a serial letter to everyone
in the database) - while it selects only the users who are already loaded.

I agree that it makes sense to give the user interactivity before downloading
the entire db, but maybe there should be a button in the toolbar saying "load
all table entries" or something?
Comment 23 Frank Schönheit 2009-01-05 09:23:09 UTC
> There is another problem with this: If a user clicks the top left corner, he
> expects to select the entire table (e.g. to generate a serial letter to
> everyone in the database) - while it selects only the users who are already
> loaded.

This is - well, should - not be true. If you click the upper left corner, then
in fact every subsequent action should apply to the whole table, no the
visible-so-far rows. At least this is how it was implemented long ago.
Everything else is a (separate) bug, I'd say.
Comment 24 Rainer Bielefeld 2009-05-06 08:48:25 UTC
*** Issue 101510 has been marked as a duplicate of this issue. ***
Comment 25 Rainer Bielefeld 2009-05-06 09:58:13 UTC
*** Issue 101509 has been marked as a duplicate of this issue. ***