Loading ...
Sorry, an error occurred while loading the content.

RE: [antlr-interest] Re: RFC: ASTFactory change

Expand Messages
  • John Green
    I think I see the light. Yes, click through on the rules would be very cool, and there is probably lots more.-----Original Message----- From: Ric Klaren
    Message 1 of 21 , Aug 9, 2001
    • 0 Attachment
      I think I see the light. Yes, click through on the rules would be very cool,
      and there is probably lots more.


      > -----Original Message-----
      > From: Ric Klaren [mailto:klaren@...]
      > Sent: Thursday, August 09, 2001 10:08 PM
      > To: antlr-interest@yahoogroups.com
      > Subject: Re: [antlr-interest] Re: RFC: ASTFactory change
      >
      >
      > On Thu, Aug 09, 2001 at 09:59:45PM +1000, John Green wrote:
      > > Sorry for being a little slow - Ric, can you describe what you
      > have in mind
      > > by a documentation template?
      > >
      > > I have in mind that the Antlr grammar for the tree parser would be, in
      > > itself, an ideal format for the documentation. It sounds like
      > you have an
      > > idea for something better - yes?
      >
      > In the development snapshots I added an option to generate a docbook file
      > from a grammar. If you process the docbook you get or html files you can
      > click through all rules.. or pdf files that work the same. For generated
      > trees you might want to add some extra fluff to it I can imagine. Or maybe
      > add some doxygen like things to the comment parsing..
      >
      > Guess I'll second you on the wanting to get my hands on the
      > critter =) Then
      > I can have a play around with it...
      >
      > Ric
      > --
      > -----+++++*****************************************************+++
      > ++++++-------
      > ---- Ric Klaren ----- klaren@... ----- +31 53 4893722 ----
      > -----+++++*****************************************************+++
      > ++++++-------
      > Time what is time - I wish I knew how to tell You why - It hurts
      > to know -
      > Aren't we machines - Time what is time - Unlock the door
      > - And see the truth - Then time is time again
      > From: 'Time what is Time' by Blind Guardian
      >
      >
      >
      >
      > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
      >
      >


      _________________________________________________________
      Do You Yahoo!?
      Get your free @... address at http://mail.yahoo.com
    • Terence Parr
      ... Yep, Loring and I will be writing a paper on the converter. He s wading through the bowels of ANTLR s code generation at the moment ;) I have warned him
      Message 2 of 21 , Aug 9, 2001
      • 0 Attachment
        Thursday, August 09, 2001, Ric Klaren hath spoken:
        > On Thu, Aug 09, 2001 at 09:59:45PM +1000, John Green wrote:
        >> Sorry for being a little slow - Ric, can you describe what you have in mind
        >> by a documentation template?
        >>
        >> I have in mind that the Antlr grammar for the tree parser would be, in
        >> itself, an ideal format for the documentation. It sounds like you have an
        >> idea for something better - yes?

        > In the development snapshots I added an option to generate a docbook file
        > from a grammar. If you process the docbook you get or html files you can
        > click through all rules.. or pdf files that work the same. For generated
        > trees you might want to add some extra fluff to it I can imagine. Or maybe
        > add some doxygen like things to the comment parsing..

        > Guess I'll second you on the wanting to get my hands on the critter =) Then
        > I can have a play around with it...

        Yep, Loring and I will be writing a paper on the converter. He's
        wading through the bowels of ANTLR's code generation at the moment ;)
        I have warned him of the interest generated for it ;)

        Boy I would love to rewrite ANTLR's code generation engine to use
        templates / trees, but what a mess ;) I think Loring is having a
        heartattack in there ;) Bring your wading boots!

        You know it's funny: After building jGuru (i.e., actual commercial
        software) I've learned a lot about just how much extra work has to go
        into something to make it a real product instead of a "cool thing I
        whipped up". I think that previously you could characterize my work as
        having the right project strategy, the right algorithms, many of the
        right ideas, but the code has always been done in panic mode and
        chasing a moving target. Mea culpa, I'm beating my breast. Oh well,
        the book is probably the most important thing for me to work on.
        Perhaps ANTLR 3.0 will happen after that and from a spec approved by
        the community since I think by now we have a pretty damn good idea of
        what's useful and what is not. Sounds like a good
        foodfight/feeding-frenzy for a the next "Dr. T's Traveling Parsing
        Revival and Beer Tasting Festival." I'm pretty sure I'm coming out of
        the woods this Fall; I'm getting that urge to have another conference.
        Anybody object to Hawaii in February? ;)

        Adios,
        Ter
        --
        Chief Scientist & Co-founder, http://www.jguru.com
        Creator, ANTLR Parser Generator: http://www.antlr.org
      • Ric Klaren
        Hi, ... Hmmm while he s walking there he might add some comments where some were uhm.. forgotten ;) ;) (at least if he figured stuff out there =) ) ... Hmmm
        Message 3 of 21 , Aug 9, 2001
        • 0 Attachment
          Hi,

          On Thu, Aug 09, 2001 at 08:45:31AM -0700, Terence Parr wrote:
          > > Guess I'll second you on the wanting to get my hands on the critter =) Then
          > > I can have a play around with it...
          >
          > Yep, Loring and I will be writing a paper on the converter. He's
          > wading through the bowels of ANTLR's code generation at the moment ;)
          > I have warned him of the interest generated for it ;)

          Hmmm while he's walking there he might add some comments where some were
          uhm.. forgotten ;) ;) (at least if he figured stuff out there =) )

          > Boy I would love to rewrite ANTLR's code generation engine to use
          > templates / trees, but what a mess ;) I think Loring is having a
          > heartattack in there ;) Bring your wading boots!

          Hmmm just survived Wacken Open Air in Germany.. nothing much scares me
          although I have been to worse festivals (in the mud/rain sense of the word
          =) )

          > foodfight/feeding-frenzy for a the next "Dr. T's Traveling Parsing
          > Revival and Beer Tasting Festival."

          I'm still the proud owner of a few La Trappe Quadrupple belgium beers that
          are now way past their expiry date (in the case of belgium beers this is a
          good thing (tm) the ones that survive only get better =) ) So drop by and
          You can have a taste... Then see about revival and survival ;)

          > I'm pretty sure I'm coming out of the woods this Fall; I'm getting that
          > urge to have another conference. Anybody object to Hawaii in February? ;)

          Hmmm I should get a better paying job eh?

          Fun aside, doing a complete rebuild with Gnu make from the toplevel of my
          devel version now works... Now glue in some configure stuff, and some stuff
          to build different versions of the library (for different compiler
          versions), and I'm up for a new devel release. (Big maybe for tomorrow else
          next week)

          Ric

          PS How could a language designer ever get the idea to make whitespace part
          of the syntax (indentation in stead of {..} or begin .. end) *shiver* (just
          looking at cyberchair conference paper submission/review system, which is
          written in Python, which reminded me of good old Occam (another *shiver*))
          --
          -----+++++*****************************************************+++++++++-------
          ---- Ric Klaren ----- klaren@... ----- +31 53 4893722 ----
          -----+++++*****************************************************+++++++++-------
          Time what is time - I wish I knew how to tell You why - It hurts to know -
          Aren't we machines - Time what is time - Unlock the door
          - And see the truth - Then time is time again
          From: 'Time what is Time' by Blind Guardian
        • John D. Mitchell
          ... [...] ... IMNSHO, because neither of them were actually designed by language designers . Of course, most computer languages are built/designed by
          Message 4 of 21 , Aug 9, 2001
          • 0 Attachment
            >>>>> "Ric" == Ric Klaren <klaren@...> writes:
            [...]
            > PS How could a language designer ever get the idea to make whitespace
            > part of the syntax (indentation in stead of {..} or begin .. end)
            > *shiver* (just looking at cyberchair conference paper submission/review
            > system, which is written in Python, which reminded me of good old Occam
            > (another *shiver*)) --

            IMNSHO, because neither of them were actually designed by "language
            designers".

            Of course, most computer languages are built/designed by non-designers. :-)

            Take care,
            John
          • Sinan
            ... Or an interpreter...... Sinan
            Message 5 of 21 , Aug 9, 2001
            • 0 Attachment
              Ric Klaren wrote:
              >
              > Hi,
              >
              > On Thu, Aug 09, 2001 at 09:28:14PM +1000, John Green wrote:
              > > > his automatic tree-parser construction code (i.e., given a text-parser
              > > > grammar with tree construction actions, automatically generate the
              > > > tree parser for it).
              > >
              > > Is there any way that I could get my hands on that critter?
              > >
              > > My current task is to provide documentation for the tree that my parser
              > > generates. I thought that the best format for documenting the tree would be
              > > a working Antlr tree-parser. I wasn't going to worry about the "working"
              > > part right now. It's just the documentation part that's important.
              >
              > Hmmm if one would automatically generate the treeparser it might be one
              > easy step away to generate a documentation template from it?
              >
              > Just a thought,
              >
              > Ric

              Or an interpreter......

              Sinan
            • James Mansion
              ... Hey! I wrote an Occam 1 compiler for the IBM 3081D. It even featured NOSFERATU - the New Occam Scheduler, Fortunately All Runs Terminate Usefully. Well,
              Message 6 of 21 , Aug 14, 2001
              • 0 Attachment
                > written in Python, which reminded me of good old Occam (another *shiver*))

                Hey! I wrote an Occam 1 compiler for the IBM 3081D.

                It even featured NOSFERATU - the New Occam Scheduler, Fortunately All Runs
                Terminate Usefully.

                Well, it was better than doing the coursework ... ;-)

                James
              • jsrs701@yahoo.com
                (was: Re: Loring s auto tree grammar / doc generator [+Ter s Mea Culpa]) Well, maybe not ethics per se... ... whitespace part ... *shiver* (just ... which is
                Message 7 of 21 , Aug 15, 2001
                • 0 Attachment
                  (was: Re: Loring's auto tree grammar / doc generator [+Ter's Mea
                  Culpa])

                  Well, maybe not "ethics" per se...

                  --- In antlr-interest@y..., Ric Klaren <klaren@c...> wrote:
                  > PS How could a language designer ever get the idea to make
                  whitespace part
                  > of the syntax (indentation in stead of {..} or begin .. end)
                  *shiver* (just
                  > looking at cyberchair conference paper submission/review system,
                  which is
                  > written in Python, which reminded me of good old Occam (another
                  *shiver*))

                  I agree--mostly.

                  I honestly think that our bias against whitespace-sensitive languages
                  is because the ones that existed Way Back When were not very good
                  languages in the first place. The ones that were good were not
                  senstive to whitespace (odd commenting syntax aside), and so we
                  became accustomed to them.

                  However, don't we all have our own ideas of how code should *look* on
                  paper? Our own set of indentation heuristics? For example, I think
                  the public and protected keywords in C++ class definitions should be
                  indented one level in from the class keyword. Other people think
                  they should be at the same level. Whatever, we all have our ideas.

                  Now <flamebait> aside from the fact that Guido is an insufferable
                  jerk </flamebait> I think his decision to make Python whitespace-
                  senstive was a good one, for the purposes of the language.

                  Python is a groovy little language for teaching newbies structured
                  programming. Why? Because most people don't instantly grok the
                  whitespace-insensitive grammars. Tell a beginning programming class
                  that the open curly brace can be on the same line as the if
                  statement, or on the next line, or wherever, and half of the students
                  will give you that priceless blank stare that all teachers have come
                  to love. However, tell them that the next line has to be indented,
                  and they get that instantly. For most people, purely *visual* rules
                  make a lot of sense. They'll hopefully build on that later to the
                  point where they understand and appreciate the malleability of the C
                  grammar. But until then, you at least have a chance to teach
                  structured programming in a context that they'll understand and be
                  able to work within very easily.

                  <flamebait part=2>
                  Of course after that, Python's usefulness drops off considerably
                  because it has such a god-awful syntax... :-)
                  </flamebait>

                  (Wow. I'm gonna get crucified for this post, ain't I?)

                  -JSRS
                Your message has been successfully submitted and would be delivered to recipients shortly.