Re: [E-devel] [Patch] evas_ojbect_textblock, ellipsis handling

[Kim S. <kim...@gm...>]

Please leave your attention from the previous patch.
I have newly revised the patch on both backport and trunk.

2012/5/4 Kim Shinwoo <kim...@gm...>:
> Dear Tom,
>
> I have attached a patch to make proper edje which uses the textblock
> with ellipsis!
> Please review the patch and keep it on upstream.
>
> Sincerely,
> Shinwoo Kim.
>
> 2012/5/4 Kim Shinwoo <kim...@gm...>:
>> Dear Tom, Good afternoon again! ;p
>> I'm back with gooood news!! The "fixed:1 x" resolved this issue
>> finally. Thank you for your help!!
>>
>> 2012/5/4 Kim Shinwoo <kim...@gm...>:
>>> Dear Tom, Good afternoon! :p
>>> Thanks for your response. Tomorrow.. actually today.. I will check it with
>>> revised edje which has "fixed:1 x";.
>>>
>>>> maybe a calculation
>>> hmm.. it would be able to be caused by calculation, then we cannot say..
>>> "Now, issue is resolved."?
>>> Doesn't the calculation problem occur, if I use the "fixed:1 x;"? then we
>>> can say.. "Now, issue is resolved."!
>>> Anyhow, test is the best way to check..
>>>
>>>
>>>> I don't think it can easily be fixed, especially since it's currently
>>>> just a not supported behaviour. How would you expect it to behave?
>>> Does it mean "fixed: 0 x;" - textblock with ellipsis is used for sizing - ?
>>> yeap, we cannot expect the behavior.. ah.. delayed calculation.. dead lock..
>>> :-/
>>>
>>> I will be back. -.,-)=b
>>> Shinwoo Kim.
>>>
>>> 2012년 5월 3일 목요일에 Tom Hacohen<tom...@sa...>님이 작성:
>>>
>>>> Dear Shinwoo,
>>>>
>>>> I've checked your issue and I indeed managed to reproduce it using your
>>>> patch (actually, only half of your patch was needed, the text was
>>>> enough, no need for the style).
>>>>
>>>> It's not really a textblock issue. It's just a matter of a bad edje, or
>>>> maybe a calculation, can't tell. The problem is that the size of the
>>>> textblock is very "unstable" because of the ellipsis. In such cases, the
>>>> textblock should not be used for sizing, so it must be defined as fixed:
>>>> 1 X; (height doesn't really matter, the width is the issue) as I
>>>> suggested before.
>>>>
>>>> I don't think it can easily be fixed, especially since it's currently
>>>> just a not supported behaviour. How would you expect it to behave?
>>>>
>>>> --
>>>> Tom.
>>>>
>>>> On 24/04/12 05:29, cnook wrote:
>>>>> Dear Mr. Tom,
>>>>>
>>>>> Finally!! I have attacked patch for your example. To make the issue,
>>>>> please
>>>>> follow as below.
>>>>>
>>>>>    elementary_test>  popup>  popup-center-title + text +1 button
>>>>>
>>>>> Then the popup needs some time to display itself with following log.
>>>>>
>>>>>    ERR<32658>:evas_main evas_object_smart.c:599
>>>>> evas_object_smart_need_recalculate_set() Object 0xb76d5df8 is not stable
>>>>> during recalc loop
>>>>>    ERR<32658>:evas_main evas_object_smart.c:599
>>>>> evas_object_smart_need_recalculate_set() Object 0xb76d5df8 is not stable
>>>>> during recalc loop
>>>>>    ...
>>>>>
>>>>> Please check this on the textblock side. Thanks a lot.
>>>>>
>>>>> Sincerely,
>>>>> Shinwoo Kim.
>>>>>
>>>>> 2011/11/21 Tom Hacohen<tom...@pa...>
>>>>>
>>>>>> On 21/11/11 16:04, cnook wrote:
>>>>>>
>>>>>>> Dear Mr. Tom,
>>>>>>>
>>>>>>> The issue comes again with different example.
>>>>>>> So I have attached previous patch for resolving. (Actually I have made
>>>>>>> from the latest version)
>>>>>>>
>>>>>>
>>>>>>
>>>>>> Dear Shinwoo,
>>>>>>
>>>>>> As I said in a previous email to the list, a compilable example or an
>>>>>> edje
>>>>>> I can run with edje_player would be very appreciated. I currently can't
>>>>>> reproduce the issue and thus I can't even check you solution works (I
>>>>>> trust
>>>>>> you that it works, but still...), nor can I verify the problem.
>>>>>>
>>>>>> As for your patch:
>>>>>>
>>>>>> c->w will be smaller than c->wmax in A LOT of cases, that's actually
>>>>>> intentional, setting c->wmax to c->w makes no sense in textblock, as
>>>>>> it'll
>>>>>> break a lot of code.
>>>>>>
>>>>>> c->wmax is used to determine the actual *used* width, and c->w is the
>>>>>> width of the textblock object.
>>>>>>
>>>>>> So for example, if we have a textblock with a size of 300x300 and
>>>>>> assuming the string "a" * 100 (100 'a's in a row) have a width of 1000
>>>>>> and
>>>>>> that the textblock contains that string:
>>>>>> c->w = 300;
>>>>>> c->wmax = 1000;
>>>>>>
>>>>>> Making edje know that 300 is just not enough to show this string. That's
>>>>>> how the min size calculation actually works.
>>>>>>
>>>>>> You can try adding "fixed: 1 1;" to the textblock part if you wish to
>>>>>> implement your behaviour, that's exactly what it does.
>>>>>>
>>>>>>
>>>>>> With that being said, there might be a bug there, as your condition
>>>>>> should
>>>>>> always be false in that place (or at least in the cases that matter),
>>>>>> and
>>>>>> for that, I need the example I asked for. :)
>>>>>>
>>>>>> --
>>>>>> Tom.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> ------------------------------------------------------------------------------
>>>>>> Live Security Virtual Conference
>>>>>> Exclusive live event will cover all the ways today's security and
>>>>>> threat landscape has changed and how IT managers can respond.
>>>>>> Discussions
>>>>>> will include endpoint security, mobile security and the latest in
>>>>>> malware
>>>>>> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> enlightenment-devel mailing list
>>>>>> enl...@li...
>>>>>> https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------------------------
>>>> Live Security Virtual Conference
>>>> Exclusive live event will cover all the ways today's security and
>>>> threat landscape has changed and how IT managers can respond. Discussions
>>>> will include endpoint security, mobile security and the latest in malware
>>>> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>>>> _______________________________________________
>>>> enlightenment-devel mailing list
>>>> enl...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
>>>>