[Laszlo-dev] For Review: Change 20090304-hqm-l Summary: make script cache persistent by default when running lzc

[P T Withington]

Sorry to be such a pain but, should you use something like SafeArg  
that the rest of the arg processing does?  It seems your code will  
blow up if the user doesn't supply an "=".

Is there no getOpt library for Java?

Should you add to the help message that the default value for script  
cache is in lps.properties?

On 2009-03-04, at 11:40EST, Henry Minsky wrote:

> Change 20090304-hqm-l by hqm at badtzmaru.home on 2009-03-04 11:28:03 EST
>    in /Users/hqm/openlaszlo/trunk6
>    for http://svn.openlaszlo.org/openlaszlo/trunk
>
> Summary:  make script cache persistent by default when running lzc
>
> New Features:
>
> Bugs Fixed: LPP-7699
>
> Technical Reviewer: ptw
> QA Reviewer: (pending)
> Doc Reviewer: (pending)
>
> Documentation:
>
> Release Notes:
>
> Details:
>
> + script cache is persisted by default when compiling via lzc  
> (compiler/Main.java)
>
> + removed --keepscriptcache, as it is redundant with --flush-script- 
> cache
>
> + option boolean arg supported for --flush-script-cache[=true|false]
>
> + default behavior can be set in lps.properties:  
> compiler.scache.persist=true|false
>
> + the script cache is (still) unconditionally persisted when  
> compiling via the LPS server
>
> Tests:
>
> compile an an app from command line , timing the execution of the  
> compiler
>
>  time lzc demos/calendar/calendar.lzx
>
> then compile again and see if time is significantly less, due to  
> caching in script compiler
>
>
> then repeat with --flush-script-cache, and see if time goes up to  
> the first compile time
>
>  time lzc --flush-script-cache demos/calendar/calendar.lzx
>
>
> Files:
> M      WEB-INF/lps/config/lps.properties
> M      WEB-INF/lps/server/src/org/openlaszlo/compiler/Main.java
>
> Changeset: http://svn.openlaszlo.org/openlaszlo/patches/20090304-hqm-l.tar