Re: template.loader syntax in velocity.properties : parser fix

["Geir Magnusson Jr." <ge...@optonline.net>]

I found and fixed Christoph's template bug.  I will check them in later
when the template.path issue is solved.  If anyone wants a copy of the
Parser stuff that fixes it, let me know and I will send it..

geir

"Geir Magnusson Jr." wrote:
> 
> Christoph Reck wrote:
> >
> > Hi,
> >
> > I'm trying to run velocity standalone via the ...test.misc.Test.java
> > class to be able to track down some problems and supply the fixes.
> 
> 0) What is the problem you originally were trying to solve?
> 
> 1) If this was a fresh CVS tree, there are a few problems because of the
> work that Jason did on the Runtime/Loader/Template system.  We know what
> they are and are being worked on.
> 
> For now, I would suggest hardwiring, or backup to the 2 day old nightly
> build.
> 
> That might be the best course, the 2 day old nightly build....
> 
> > Since I'm running JBuilder under Win95, it's having trouble with
> > the loading path (it doesn't find the properties nor the template
> > at '.').
> >
> > Hardwiring the path to velocity.properties brings other bugs into
> > appearance:
> > * Runtime.java is accessing "external_init" from the properties,
> >   but it's not there, so bang :(
> >   -> could you add it to the example properties files and maybe
> >      change lines 340 and 409 to do a getString on EXTERNAL_INIT
> >      with an empty "" default value?
> 
> Yes, that's fixed.  Throw "false" in there if you wish as the default.
> 
> > * The current properties file contain a single template.loader
> >   entry. Runtime.java is searching for template.loader.N entries.
> >   The TemplateFactory is trying to access some "class" parameter
> >   of it.
> >   -> What is the current syntax for the template.loader entry(s)?
> 
> Yep, that is the fix Jason has coming.  Will be explained when done, but
> it's recognized as a problem.  Affects Torque, Turbine and Texen as
> well, so know you are not alone :)
> 
> > The bug I'm trying to track is that
> > ----
> > foo
> > ## ##
> > bar
> > ----
> > emits
> > ----
> > foo
> > ----
> > swallowing the bar (seems like the second ## within a ## makes this!
> >
> > Also if the file would end after bar (without a EOL), it throws an
> > exception:
> >   Lexical error at line 4, column 4.  Encountered: <EOF> after : ""
> 
> I'll nail that one now, if you want.  Simple parser issue... my bad...
> 
> > Thanks for any help,
> > :) Christoph
> 
> --
> Geir Magnusson Jr.                               geirm@optonline.com
> Dakota tribal wisdom: "when you discover you are riding a dead horse,
> the best strategy is to dismount."

-- 
Geir Magnusson Jr.                               geirm@optonline.com
Dakota tribal wisdom: "when you discover you are riding a dead horse,
the best strategy is to dismount."