Re: CRM114 script language question: divide operation in maillib

Ouch. So I need make sure my new parser statemachine can handle stuff
like this. And it passed megatest already for months...

This means it also has to know about ;@; and preferable the other
:.:...: constructs too. Ah, this info was so nicely _not_ spread
everywhere anymore. Sigh. I see the point, so I'll fix the GerH parser
and throw in the appropriate smarts.

And then a certain someone (hi, Bill ;-) ) wrote a long time ago that
he was looking for some bison fu to replace the parser. Darn, should
have asked for some serious 'flex' (lexer) fu instead: the complexity
is really in the scanner here.

I'll be back.    Like Arnold. But not as a governor, unfortunately.

Thanks for helping me out by clearing this one up,


Ger


BTW: have you tried and run, for instance, the extra statement

output /ln = :*:cache_dupe_command:/

anywhere in mailreaver.crm lately? If you do, you might find another
surprise there (at least the GerH parser does not process the escapes
in the .cf file because crm114 treats it as a *data* file (which just
looks quite like a CRM114 script) and crm114 treats all data as
binary, so without any 'C style' escapes to, er, process.

Same wickedness goes for the :mail_separator: in there. THAT one
should therefore *not* do to the logfile what one'd expect - at least
if that one thinks too much in terms of 'C' when he's doing the
expecting.

The fun only starts when you try to run this kind of stuff on a
non-UNIX sh (say, Win32, for instance) shell environment; in UNIX'
case, the sh shell instantiated by the syscall() will handle the
escapes for you, and lo, it works! ;-)







On Mon, Mar 31, 2008 at 6:19 AM, Paolo <[email protected]> wrote:
> On Mon, Mar 31, 2008 at 03:04:13AM +0100, Ger Hobbelt wrote:
>  >
>  > In maillib.crm it says @ ~ line 163:
>  >
>  >             eval /:@: :#:m_text: < (:*:decision_length: / 4) :/
>  >
>  >    Shouldn't this read:
>  >
>  >             eval /:@: :#:m_text: < (:*:decision_length: \/ 4) :/
>  >
>  >    (note the 'divide' being escaped in there)
>  >
[...]
>  yes, it's supposed to do so. IIRC it's been discussed time ago, that it'd be
>  ugly, uncomfortable, unexpected to be required to escape math symbols within
>  :@:...:, only because they happen to be our delimiters.
[...]



-- 
Met vriendelijke groeten / Best regards,

Ger Hobbelt

--------------------------------------------------
web: http://www.hobbelt.com/
 http://www.hebbut.net/
mail: [email protected]
mobile: +31-6-11 120 978
--------------------------------------------------

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace