[antlr-interest] ANTLR v4 status / website functionality moving forward - Ruby Target ?

[Martin Van Aken]

Hello again,
Thanks for the clarification. I can assume that to have a target that
works, you need an implementation of the runtime also. I've foud some
documentation about the .stg part, but almost nothing about the runtime.
Any pointer (excepted looking at the java reference) ?

Thanks,

Martin

On 24 September 2012 19:52, Terence Parr <parrt at cs.usfca.edu> wrote:

>
> On Sep 18, 2012, at 11:33 PM, Martin Van Aken wrote:
>
> > Hello Terence,
> > Thanks for the info, this is far better than nothing, and I like the
> incremental process outlined there. There is still something I'm not sure
> to understand : the String Template .stg file is used to generate the
> parser/lexer code for the grammar. But those classes inherit from more
> general classes in Antlr (in the java version, at least). Are those also
> generated by the .stg ?
>
> The string template files just generate code specific to the grammar. Most
> of the common functionality has been factored out into a runtime library.
> So Lexer, Parser and so on are all part of the runtime library, not
> generated by antlr. This is in contrast to, for example, javacc that
> generates all of the runtime for each grammar. If you have multiple
> grammars in the same  project, you get duplicates.
>
> :)
>
> T
> >
> > In other word, what about the runtime of Antlr, is it an output of the
> .stg, or is it another part to implement to be able to have a parser
> running under a given language (let say Ruby).
> >
> > Thanks !
> >
> > Martin
> >
> > On 17 September 2012 00:44, Terence Parr <parrt at cs.usfca.edu> wrote:
> > Unfortunately, no one has stepped up to bring the Ruby target up to
> date. The other bit of bad news is that we really don't have much in the
> way of how to create a target. just this page:
> >
> >
> http://www.antlr.org/wiki/display/ANTLR3/How+to+build+an+ANTLR+code+generation+target
> >
> > Ter
> > On Sep 16, 2012, at 5:51 AM, Martin Van Aken wrote:
> >
> >> Terrence/list,
> >> I've the same question about the Ruby target. For what I can see from
> >> github : https://github.com/antlr/antlr3/tree/master/runtime/Ruby it
> has
> >> not move since two years so it is probably lagging behind. Do anyone is
> >> still maintaining it ?
> >>
> >> If not, I may be interested in trying to update it myself. Any resource
> for
> >> (would be) goal maintener that I could start with ? Anyone that would be
> >> interested to contribute (time, advice, test, anything). I may take a
> look
> >> at the python target (that seems to be keeping up well) as a reference
> >> (closer to Ruby than Java).
> >>
> >> Thanks a lot.
> >>
> >> Martin
> >>
> >> On 16 September 2012 09:44, Kieran Simpson <kierans777 at gmail.com>
> wrote:
> >>
> >>> Ter/list,
> >>>   Thanks again for all your efforts.  In terms of other language
> >>> targets is there an idea/outline of when they'll be available.  I'm
> >>> specifically thinking of the C target.  In a list thread from January
> >>> there was the indication that a C++ target was still a while away so
> any
> >>> progress updates?
> >>>
> >>> Cheers,
> >>>
> >>>
> >>> On 16/09/12 4:36 AM, Terence Parr wrote:
> >>>> Howdy folks,
> >>>>
> >>>> ANTLR v4 release is rapidly approaching. The beta of the reference
> book
> >>> will be out next week and the remaining two or three chapters should
> appear
> >>> within a month or so afterwards. Sam Harwell and I have been working
> very
> >>> hard on the tool itself and we should have 4.0 ready by the time the
> book
> >>> goes final. In the meantime, 4.0b1 will be available for use with the
> beta
> >>> book. Oh, and we need to release 3.4.1 before 4.0.
> >>>>
> >>>> I have paid for a new website design for both ANTLR and
> StringTemplate,
> >>> which looks great. We will continue to use the same wiki software for
> >>> documentation. The current website is generated by a Java server I
> built
> >>> whereas the new one is going to be static so I have less software to
> >>> maintain. In other words,  rather than using some kind of include
> mechanism
> >>> to get the general look and feel on each page, the new websites will be
> >>> simply static files on the disk.  The current antlr.org content will
> >>> become antlr3.org, leaving the current domain pointing at v4 content.
> >>>>
> >>>> We currently have functionality on the websites to accept new grammars
> >>> and filesharing and articles and so on. Because this is so infrequent,
> I
> >>> think it's reasonable to simply have an HTML form that has an email
> action
> >>> instead of an HTTP POST. When I get those requests, I can simply add
> them
> >>> to the file on the server. (will that use the user's local mail client
> or
> >>> will it force people to set up mail in their actual browsers before it
> will
> >>> email me? does anybody know?)
> >>>>
> >>>> On to the grammar repository. Because it's likely we'll want to make
> >>> fixes / updates to existing grammars, I don't think a simple form /
> email
> >>> mechanism is the best solution. Right now, I have to go in and
> overwrite /
> >>> update a number of files for a grammar update. Naturally, this screams
> for
> >>> a revision control solution. I was thinking that we might as well just
> use
> >>> github for this so that anybody can add or modify the publicly
> available
> >>> grammars.
> >>>>
> >>>> There are a number of issues with using github for this. First, I
> would
> >>> not want to create a new repository for each grammar so we would have
> one
> >>> repository holding all grammars. This is pretty coarse granularity.
>  On the
> >>> other hand, if you just want one grammar, you can download individually
> >>> from github. The second issue is that we would really have to have a
> single
> >>> license for all grammars in the repository. I would hate for a GPL
> grammar
> >>> to get its stank on the other grammars. It would confuse people to have
> >>> multiple licenses within a single repository. Thirdly, not everyone is
> >>> comfortable with assembly language…er…I mean git. In that case, people
> >>> could simply mail me a grammar for inclusion. It would only take me a
> >>> second to add it. The fourth problem. We need a clean URI for grammars
> and
> >>> I propose:
> >>>>
> >>>> http://www.antlr.org/grammars/<name>
> >>>>
> >>>> for the root directory of that project. For example,
> >>>>
> >>>> http://www.antlr.org/grammars/java
> >>>>
> >>>> would point out a directory that contains Java.g4 and may be a test
> >>> program or something.
> >>>>
> >>>> I could easily add a redirect in the tomcat configuration files,
> >>> assuming I can stomach all of that filthy XML, but that does not scale
> very
> >>> well when people add grammars. Instead, perhaps the best solution is
> to set
> >>> up a cronjob that pulls from the grammar repository and leaves the
> grammars
> >>> on antlr.org's disk so that /grammars URI points at that directory.
> That
> >>> way, the URIs would always be up-to-date with the repository and
> without me
> >>> having to do any work. Heh, that just might work.
> >>> http://www.antlr.org/grammars by itself could redirect to the github
> >>> project.
> >>>>
> >>>> Anyway, If you have any thoughts on this stuff, please reply.
> >>>>
> >>>> Terence
> >>>
> >>> List: http://www.antlr.org/mailman/listinfo/antlr-interest
> >>> Unsubscribe:
> >>> http://www.antlr.org/mailman/options/antlr-interest/your-email-address
> >>>
> >>
> >>
> >>
> >> --
> >> Mobile : +32 486 899 652
> >> Company website : www.8thcolor.com
> >> Contact me on LinkedIn : http://www.linkedin.com/in/martinvanaken
> >> Follow me on Twitter : @martinvanaken <http://twitter.com/martinvanaken
> >
> >>
> >>
> >> List: http://www.antlr.org/mailman/listinfo/antlr-interest
> >> Unsubscribe:
> http://www.antlr.org/mailman/options/antlr-interest/your-email-address
> >
> >
> >
> >
> > --
> > Mobile : +32 486 899 652
> > Company website : www.8thcolor.com
> > Contact me on LinkedIn : http://www.linkedin.com/in/martinvanaken
> > Follow me on Twitter : @martinvanaken
> >
> >
>
>


-- 
Mobile : +32 486 899 652
Company website : www.8thcolor.com
Contact me on LinkedIn : http://www.linkedin.com/in/martinvanaken
Follow me on Twitter : @martinvanaken <http://twitter.com/martinvanaken>