If this cache is supposed to be shared by all instances maybe it
should be a class variable instead of hanging off the prototype? (In
fact, there won't be a prototype in JS2, so we should be on the look
out for similar cases.)
On Feb 14, 2008, at 9:45, Philip Romanik <[EMAIL PROTECTED]>
wrote:
Change 20080214-Philip-4 by [EMAIL PROTECTED] on 2008-02-14 08:54:44
EST
in /cygdrive/f/laszlo/svn/src/svn/openlaszlo/trunk
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Access text cache as LzTextSprite.prototype._sizecache
New Features:
Bugs Fixed: LPP-5390
Technical Reviewer: max
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
The code used to access the global cache through 'this'. However,
once the cache
is full and is cleared, access through 'this' it not reliable. I
changed this._
sizecache to LzTextSprite.prototype._sizecache. Although I only need
to do this
for the line that actually resets the cache, I changed all
references to make it
clear this is a global.
Tests:
See test in LPP-5390. This is slow to run since you need to wait
until the _size
cache.counter reaches 1000. During testing I set the cache size to
100 (ie. LzTe
xtSprite.prototype.__sizecacheupperbound = 100) to verify the bug
and the fix.
Files:
M WEB-INF/lps/lfc/kernel/dhtml/LzTextSprite.js
Changeset: http://svn.openlaszlo.org/openlaszlo/patches/20080214-Philip-4.tar
