[Laszlo-dev] __updatefieldsize() for 'onkeypress'

[P T Withington]

Can't you just check length < Infinity?

On Mar 18, 2009, at 8:58, André Bargull <andre.bargull at udo.edu> wrote:

> Hmm, I think it wasn't defaulted to Infinity when I implemented that  
> part. So this if-condition needs to look like:
>> if (sprite.restrict || (sprite.multiline && view.maxlength > 0 &&  
>> isFinite(view.maxlength)) {
>
>
>
> On 3/18/2009 1:21 PM, P T Withington wrote:
>> Also, LzView/maxlength defaults to Infinity, so the last clause  
>> will always be true.
>> On 2009-03-18, at 05:59EDT, André Bargull wrote:
>>> If that's the case, a separate if-statement needs to be used..
>>>
>>>
>>> On 3/18/2009 3:00 AM, Max Carlson wrote:
>>>> I'm not sure about the sprite.restrict part, but my original  
>>>> intent was to have multiline inputtext fields update their scroll  
>>>> properties when typed into.
>>>> André Bargull wrote:
>>>>> In "LzInputTextSprite.prototype.__textEvent":
>>>>> ---
>>>>>  } else if (eventname == 'onkeypress') {
>>>>>      if (sprite.restrict || (sprite.multiline && view.maxlength  
>>>>> > 0)) {
>>>>>          sprite.__updatefieldsize();
>>>>> ---
>>>>>
>>>>> "__updatefieldsize()" is only called for "onkeypress" if the  
>>>>> inputtext has got a pattern (sprite.restrict) or if it's  
>>>>> multiline and maxlength was specified? I don't understand that  
>>>>> logic.
>>>>> And it doesn't really match the svn message:
>>>>>> Update the field size metrics onkeypress.
>>>>>
>>>>>
>>>>> --- 
>>>>> --- 
>>>>> ------------------------------------------------------------------
>>>>> r12785 | max | 2009-02-07 15:53:13 +0100 (Sat, 07. Feb 2009) |  
>>>>> 22 lines
>>>>> Ge?\195?\164nderte Pfade:
>>>>> M /openlaszlo/trunk/WEB-INF/lps/lfc/kernel/dhtml/ 
>>>>> LzInputTextSprite.js
>>>>> M /openlaszlo/trunk/lps/components/incubator/scrolledittext.lzx
>>>>> M /openlaszlo/trunk/lps/components/incubator/test/scrolledittext- 
>>>>> test.lzx
>>>>>
>>>>> Change 20090206-maxcarlson-Q by maxcarlson at Bank.local on  
>>>>> 2009-02-06 06:43:58 PST
>>>>>  in /Users/maxcarlson/openlaszlo/trunk-clean
>>>>>  for http://svn.openlaszlo.org/openlaszlo/trunk
>>>>>
>>>>> Summary: Fix runtime setMultiline() and scrolledittext sizing in  
>>>>> DHTML
>>>>>
>>>>> Bugs Fixed: LPP-6580  scrolledittext doesn't work in 4.1  
>>>>> (partial), LPP-7726 - DHTML: setMultiline() has no effect after  
>>>>> init time
>>>>>
>>>>> Technical Reviewer: ptw
>>>>> QA Reviewer: hminsky
>>>>>
>>>>> Documentation: LzInputTextSprite - Refactor inputtext div  
>>>>> initialization to separate method, call when creating an  
>>>>> inputtext, or when the mu
>>>>> ltiline state changes in setMultiline().  Update the field size  
>>>>> metrics onkeypress.
>>>>>
>>>>> scrolledittext - Use constraints instead of  
>>>>> applyConstraintMethod() for maintaining height/width.
>>>>>
>>>>> scrolledittext-test - Uncomment second scrolledittext, use  
>>>>> default scrollbar class to test functionality.
>>>>>
>>>>> Details: This makes scrolledittext begin working in DHTML, with  
>>>>> some issues.  Sometimes, the scroll attribute assertion is  
>>>>> triggered, and ma
>>>>> xscroll doesn't seem to be tall enough.  Also, typing carriage  
>>>>> returns into the inputtext doesn't cause it to automatically  
>>>>> scroll down when
>>>>> typing into the bottom of the field like it does with native  
>>>>> scrollbars - the scrollTop/scroll doesn't change.  I'm not sure  
>>>>> how to fix thi
>>>>> s because AFAIK there's no way to accurately figure out the  
>>>>> cursor pixel offset when typing and set it manually - maybe  
>>>>> Tucker has a clue...
>>>>>
>>>>>
>>>>> Tests: scrolledittext-test runs as before in swf8/9, and now  
>>>>> (almost) works in DHTML.
>>>>>
>>>>> --- 
>>>>> --- 
>>>>> ------------------------------------------------------------------