Bug 7576 - (int-154879) Capitalization on smileys lost
(int-154879)
: Capitalization on smileys lost
Status: RESOLVED WONTFIX
Product: Chat & Call & SMS
Messaging
: 5.0/(3.2010.02-8)
: N900 Maemo
: Unspecified normal (vote)
: ---
Assigned To: rtcomm@maemo.org
: im-chat-bugs
:
:
:
:
  Show dependency tree
 
Reported: 2010-01-02 12:11 UTC by Nick Leppänen Larsson
Modified: 2010-03-05 17:23 UTC (History)
3 users (show)

See Also:


Attachments


Note

You need to log in before you can comment on or make changes to this bug.


Description Nick Leppänen Larsson (reporter) 2010-01-02 12:11:44 UTC
SOFTWARE VERSION:
1.2009.42-11

EXACT STEPS LEADING TO PROBLEM: 
Conversations -> New SMS -> Enter ":D" as message, send to a contact using a
phone not showing smiley as pictures.

EXPECTED OUTCOME:
Message arrives as ":D"


ACTUAL OUTCOME:
Message arrives as ":d"


REPRODUCIBILITY:
always

EXTRA SOFTWARE INSTALLED:

OTHER COMMENTS:


User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 6.1; en-GB; rv:1.9.1.6)
Gecko/20091201 Firefox/3.5.6
Comment 1 Lucas Maneos 2010-01-02 13:36:12 UTC
Confirmed, also in Jabber conversations.
Comment 2 Nick Leppänen Larsson (reporter) 2010-01-05 21:47:54 UTC
I managed it to get sent as ":D" by sending this:

*A* :D *a*

A and a can be exchanged with any non-empty string, I *think*

Sending "** :D **" gets converted to "** :d **"

Tested with SMS
Comment 3 Mikhail Zabaluev nokia 2010-01-28 16:26:01 UTC
We cannot reproduce it as described in recent builds. There is a different
problem with case there, so I'll keep this open for now.
Comment 4 Andre Klapper maemo.org 2010-02-08 18:20:09 UTC
"This will be fixed in rtcom-messaging-ui 1.2.20" - setting Target Milestone.
Comment 5 Andre Klapper maemo.org 2010-02-22 19:11:05 UTC
Forwarding internal comment:

"This is not a quick fix, since the icons replace parts of the text.
Complications come when you start deleting text (or text ranges) containing
icons in the editor, it should update the original preserved text accordingly,
which isn't trivial. -> WONTFIX"
Comment 6 Lucas Maneos 2010-02-24 11:34:46 UTC
(In reply to comment #5)
> "This is not a quick fix, since the icons replace parts of the text.
> Complications come when you start deleting text (or text ranges) containing
> icons in the editor, it should update the original preserved text accordingly,
> which isn't trivial. -> WONTFIX"

Is there any context you can share on this, as it doesn't sound directly
related?

Moreover, in 3.2010-02-8 (rtcom-messaging-ui 1.1.3-1+0m5, and yes I know this
wasn't updated in PR1.1.1) it seems fixed or at least not reproducible: typing
and sending ":D" to the same recipient as in comment 1 now arrives as ":D" at
the other end.
Comment 7 Andre Klapper maemo.org 2010-03-04 17:49:42 UTC
(In reply to comment #6)
> Is there any context you can share on this, as it doesn't sound directly
> related?

Sure:

"It's because we're using an icon cache to get the icons for the text editor.
So
that if someone puts in 100 x ":)", the icon isn't loaded to memory 100 times,
but only once.
Side effect of this is that if the same icon is there multiple times, and the
replaced characters are put in it every time some characters are resolved to
that smiley, the last resolved characters are read for all the icons when
parsing the editor field.
So that if you write ":d :D :-d", you get ":-d :-d :-d".
I think this is something we can live with.
An alternative would increase memory usage."
Comment 8 Lucas Maneos 2010-03-05 16:18:18 UTC
Thanks, that makes a lot of sense now.

So the bug is actually fixed except for the corner sub-case where two or more
instances of the same smiley appear in the same message, AND in different case. 

I think FIXED with a note to that effect would be more accurate than WONTFIX,
but that's just me :-)
Comment 9 Mikhail Zabaluev nokia 2010-03-05 17:23:09 UTC
(In reply to comment #8)
> So the bug is actually fixed except for the corner sub-case where two or more
> instances of the same smiley appear in the same message, AND in different case.
> 
> I think FIXED with a note to that effect would be more accurate than WONTFIX,
> but that's just me :-)

I agree. But I'm a bit embarrassed about the implementation that confuses
cached images with the text they are replacing...