Hmm, I guess either Tucker or Henry is going to remind you not to use multi-line comments to comment out code. Apart from that: 1) If delayed setters aren't handled anymore in __LZresolveReferences(), there is no need to keep them in code base at all. So I'd recommend to remove them completely from LzNode, LzView and LzCanvas. 2) Testing for -1 instead of "is Function" means going back to the roots ;-) (just compare it with [1]). If my memory doesn't fool me, it was Tucker who once changed the test to "is Function"...
[1] http://svn.openlaszlo.org/openlaszlo/tags/3.1.1/WEB-INF/lps/lfc/core/LzNode.as On 1/13/2010 8:21 PM, Max Carlson wrote: > Change 20100113-maxcarlson-h by maxcarl...@bank on 2010-01-13 11:05:24 PST > in /Users/maxcarlson/openlaszlo/trunk-clean > for http://svn.openlaszlo.org/openlaszlo/trunk > > Summary: Optimize LzNode > > Bugs Fixed: LPP-8716 - Improved LzNode startup time > > Technical Reviewer: ptw > QA Reviewer: hminsky > > Details: Check for presence of __LZresolveDict before installing computed > initializations (could save a function call). Explicitly test for read-only > setters in __LZapplyArgs() (testing '!= -1' should be faster than 'is > Function'). In __LZresolveReferences(), install layouts explicitly and > comment out generalized loop, formerly used for backward compatibility > (should be much faster). > > Tests: All apps run as before, including smoketest. > > Files: > M WEB-INF/lps/lfc/core/LzNode.lzs > > Changeset: > http://svn.openlaszlo.org/openlaszlo/patches/20100113-maxcarlson-h.tar > > _______________________________________________ Laszlo-reviews mailing list [email protected] http://www.openlaszlo.org/mailman/listinfo/laszlo-reviews
