Re: possibility of java style // end of line comments?

Wednesday, October 15, 2008, 1:15:33 PM, Attila Szegedi wrote:

> Yeah, problem is, we add a one-line comment syntax, it'll blow some  
> poor sap's templates apart who happens to use these characters  
> literally. I.e. double-forward-slash certainly wouldn't fly for people
> who use FreeMarker to generate Java or C++ source code (well, who  
> bother to emit comments into them, that is).
>
> Something similar to the existing [#-- would maybe be acceptable, i.e.
> with one more dash [#--- but that gives you a five character prefix...
> Or maybe [#- with a single dash, assuming our tokenizer requires a  
> whitespace after the token, so it wouldn't mistake [#-- for [#-  
> followed by another dash. Actually, same goes for [#---, that will  
> only work if tokenizer wouldn't mistake it for [#-- followed by  
> another dash.

Bad idea... A "[" visually asks for for a "]" pair, an even
technically in many text editors due to the pair highlighting. (Then,
"[#--- --#]" is legal comment, so it's not even 100% BC.) Then it
complicates the rule as I said in the other mail, and for what? For
writing "[#--- blah" instead of "[#-- blah --#]"? No a that big win
than writing "// blah".

Also, there are much higher priority tasks right now... I'm saying it
because I'm already seeing as Jon, returning from the FreeCC world,
committing this with a sudden flame of anti-ersatz enthusiasm. :-)

> Attila.
>
> On Oct 15, 2008, at 12:47 PM, Daniel Dekany wrote:
>
>> Tuesday, October 14, 2008, 11:28:02 PM, Newman, John W wrote:
>>
>>> Hi,
>>>
>>> I’m working on a template now that is exceptionally complicated.
>>> So complicated, that for once I’m actually writing comments in  
>>> it.  :-o
>>>
>>> It’s less than ideal to always have to close them with --], it
>>> would be nice if FM added some kind of // like feature.  Just a
>>> random idea I figured I’d pass on to kick around for 2.4.  But I do
>>> remember years ago when I wrote a lexer & parser for a java variant,
>>> and both /* and // were surprisingly difficult to handle.
>>
>> That's not the real problem, Jonathan is a smart guy... ;-) The
>> problem with all the syntactical shortcuts (because we have seen a few
>> proposals in that topic) that they complicate the language rules, and
>> for a template language that's seldom acceptable, since it's something
>> that people should learn as quickly as possible and then also remember
>> it if they don't use the language every day. More general purpose
>> languages can accept themselves more. For this reason I'm reluctant to
>> vote for any additional syntactical tricks... Of course I know that in
>> some applications they are quite much desirable, but the interest of
>> the majority is more important. (I think the only way to address these
>> kind of issues is that the engine has a better architecture that makes
>> it easy to define and correctly treat(!) alternative template
>> languages.)
>>
>>> -John
>>
>> -- 
>> Best regards,
>> Daniel Dekany
>

-- 
Best regards,
 Daniel Dekany


-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
FreeMarker-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/freemarker-user