Re: [Webmacro-devel] ACK! Bad new parser bug

[Marc P. <ma...@an...>]

On Thu, 17 Jul 2003 15:50:48 +0200 (MEST), Endre Stølsvik 
<web...@st...> wrote:


> | I'd appreciate it if someone with parser knowledge can look at this as 
> soon
> | as possible. Unfortunately this is yet another thing holding up my new
> | Ignition release :(
>
> You must remember that WM is in a pre-alpha stage now, given Brian's 
> large
> overhauls - I don't think its fair to demand "immediate attention" to
> problems that arise in such situations.. Use the proper release instead,
> then.

You're entirely right, but if you look at my wording there is no "demand". 
I have done a lot of debugging and some fixing of the new WM code so far as 
an "early adopter" which has taken a lot of my time also - it's just not as 
glamourous as revamping whole parts of WM ;-)

Unfortunately I can't use the 1.1 release, because features added to WM 
since 1.1 are required by Ignition. For example, the new WM ctors that take 
a Properties class, improvements to WMServlet to make it more reusable and 
so on.

...and besides, if it wasn't for Ignition we would have found a bunch of 
bugs for which we now have fixes in most cases, and in all cases new unit 
tests for. Ignition, it is hoped, will be a companion/sister project to WM 
2.0 to increase WM's user base and as such the two do need to develop 
closely together. We can't wait for WM 2.0 to be ready and then start 
developing Ignition 1.0.

I'm just asking for some help with the parser bug because that is a very 
complex area, and if it's in the parser .jj file then there's a hell of a 
lot of a learning curve for me to fix it. I already spent days learning 
about and fixing an old  PropertyOperatorCache bug.

It should be an easy fix for someone with knowledge in the parser area, and 
if we know what files it is likely to be in we can diff them to find it.

My plea for help as soon as possible is because these things can take weeks 
for someone to fix if it's not regarded as important - like the 5 unit test 
failures in TestParseInclude that have been there for a long time.

The thing is this bug should really have been caught by unit tests because 
it is such a fundamental parser issue, but it's so hard to cover all the 
bases with tests.

I have added a specific test for this problem. I don't want to spend all 
day writing new tests... imagine all the bugs we might find ;-)

Marc
-- 
Marc Palmer
Contract Java Consultant/Developer
			
                        w a n g j a m m e r s

                 java and web software design
              experts with an ethical outlook

                   http://www.wangjammers.org