good news

[David Relson]

At 11:41 AM 2/26/03, Nick Simicich wrote:
>At 12:22 PM 2003-02-25 -0500, David Relson wrote:
>
>>Nick,
>>
>>Your new code works!  True it causes 3 of the regression tests to 
>>fail.  However my unified lexer causes the same 3 tests to fail.  When I 
>>run the tests with the "-v" switch (so as to keep all the generated 
>>files) and then use "diff -ur njs tst", there are _no_ differences in the 
>>output files.
>
>Since I know that my code is broken because I know that it is in the wrong 
>state from time to time, as I described in my last missive, I believe we 
>(you?) need to look at how the code is state changing and why it is in 
>body state while parsing headers.
>
>>Now I need to think about testing, releasing, etc.  More later ...
>
>Do you want me to try and work on a performance run?

Nick,

FWIW, this message was timestamped 11:40 yesterday, but just arrived. ???

I'll take a look at the header/body thing (and will throw in a breakpoint 
at all the "From " checks). Can you send me a tarball of your src 
directory?  I'm busy this AM, but should have time later today.

Let's get the header/body thing corrected.  Then we can do performance 
testing.  If you want to put together some tests sooner, rather than later, 
that'd be great.

David