smacc in moose 6.0

List overview All Threads
Download

newer

older

moose meeting at esug

humane assessment & moose on...

Tudor Girba

12 Aug 2016 12 Aug '16

9:41 p.m.

Hi, John Brant created a repo on SmalltalkHub with the latest version of SmaCC. And now, SmaCC is part of Moose 6.0. Cheers, Doru -- www.tudorgirba.com www.feenk.com "Value is always contextual."

Show replies by date

serge.stinckwich＠gmail.com

12 Aug 12 Aug

10:53 p.m.

Great news ! There was already a Smacc repo on github managed by Thierry: https://github.com/ThierryGoubier/SmaCC Some cooperation will happens here in order to avoid duplicate work ? Regards Envoyé de mon iPhone

...

Le 12 août 2016 à 21:41, Tudor Girba <tudor(a)tudorgirba.com> a écrit : Hi, John Brant created a repo on SmalltalkHub with the latest version of SmaCC. And now, SmaCC is part of Moose 6.0. Cheers, Doru -- www.tudorgirba.com www.feenk.com "Value is always contextual." _______________________________________________ Moose-dev mailing list Moose-dev(a)list.inf.unibe.ch https://www.list.inf.unibe.ch/listinfo/moose-dev

Tudor Girba

10:57 p.m.

I know about the repo of Thierry and I hope that we will see more collaboration around SmaCC. Some of the latest changes of John were not available publicly until now, and now he was kind enough to make them available. I hope this will be considered a good thing, and not an issue of competition. Cheers, Doru

...

On Aug 12, 2016, at 10:53 PM, serge.stinckwich(a)gmail.com wrote: Great news ! There was already a Smacc repo on github managed by Thierry: https://github.com/ThierryGoubier/SmaCC Some cooperation will happens here in order to avoid duplicate work ? Regards Envoyé de mon iPhone

_______________________________________________ Moose-dev mailing list Moose-dev(a)list.inf.unibe.ch https://www.list.inf.unibe.ch/listinfo/moose-dev

-- www.tudorgirba.com www.feenk.com "Not knowing how to do something is not an argument for how it cannot be done."

Serge Stinckwich

13 Aug 13 Aug

2:32 p.m.

On Fri, Aug 12, 2016 at 10:57 PM, Tudor Girba <tudor(a)tudorgirba.com> wrote:

...

I know about the repo of Thierry and I hope that we will see more collaboration around SmaCC.

I hope so also.

...

Some of the latest changes of John were not available publicly until now, and now he was kind enough to make them available. I hope this will be considered a good thing, and not an issue of competition.

Thierry add some modifications like the parser generator browser. Maybe this is nice to share the same issue tracker : https://github.com/ThierryGoubier/SmaCC/issues -- Serge Stinckwich UCBN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/

Alexandre Bergel

14 Aug 14 Aug

10:04 p.m.

Thanks Doru. Just wondering, I always had in mind that PetitParser is a kind of ultimate framework for parsing. Why would one want to use SmaCC if PetitParser is so robust and popular? Cheers, Alexandre

...

On Aug 12, 2016, at 3:41 PM, Tudor Girba <tudor(a)tudorgirba.com> wrote: Hi, John Brant created a repo on SmalltalkHub with the latest version of SmaCC. And now, SmaCC is part of Moose 6.0. Cheers, Doru -- www.tudorgirba.com www.feenk.com "Value is always contextual." _______________________________________________ Moose-dev mailing list Moose-dev(a)list.inf.unibe.ch https://www.list.inf.unibe.ch/listinfo/moose-dev

-- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

Tudor Girba

10:31 p.m.

Hi, PetitParser is a dynamic parser. SmaCC is a static parser. Both have their usages, and the static parsers are significantly faster. The speed problem of PetitParser will likely be significantly improved once we get the work of Jan on compiling parsers. One thing that SmaCC comes with is an extensive library of parsers and ASTs (e.g., JavaScript, Java, Swift) and this is very valuable in the context of Moose. Cheers, Doru

...

On Aug 14, 2016, at 10:04 PM, Alexandre Bergel <alexandre.bergel(a)me.com> wrote: Thanks Doru. Just wondering, I always had in mind that PetitParser is a kind of ultimate framework for parsing. Why would one want to use SmaCC if PetitParser is so robust and popular? Cheers, Alexandre

-- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. _______________________________________________ Moose-dev mailing list Moose-dev(a)list.inf.unibe.ch https://www.list.inf.unibe.ch/listinfo/moose-dev

-- www.tudorgirba.com www.feenk.com "Problem solving efficiency grows with the abstractness level of problem understanding."

stepharo

19 Aug 19 Aug

12:02 p.m.

+1 When you have a antrl grammar, smacc is really nice. Stef Le 14/8/16 à 22:31, Tudor Girba a écrit :

...

-- www.tudorgirba.com www.feenk.com "Problem solving efficiency grows with the abstractness level of problem understanding." _______________________________________________ Moose-dev mailing list Moose-dev(a)list.inf.unibe.ch https://www.list.inf.unibe.ch/listinfo/moose-dev

Johan Fabry

2:50 p.m.

I did not read this before, but now I’m interested. Dory, when you say that 'the static parsers are significantly faster’ can you give some numbers on that? And you know what the numbers are compared to the compiled petitparsers of Jan and Jan ? (Their compiler is really cool BTW :-) I use it for LRP ) -- Does this mail seem too brief? Sorry for that, I don’t mean to be rude! Please see http://emailcharter.org . Johan Fabry - http://pleiad.cl/~jfabry PLEIAD and RyCh labs - Computer Science Department (DCC) - University of Chile

...

On Aug 19, 2016, at 07:02, stepharo <stepharo(a)free.fr> wrote: Le 14/8/16 à 22:31, Tudor Girba a écrit : > Hi, > > PetitParser is a dynamic parser. SmaCC is a static parser. Both have their usages, and the static parsers are significantly faster. The speed problem of PetitParser will likely be significantly improved once we get the work of Jan on compiling parsers.

Jan Kurš

20 Aug 20 Aug

3:36 p.m.

On the Smalltalk grammar, SmaCC is two times faster than PetitParser and Compiled PetitParser is four times faster. RBParser is like four point one times faster. The SmaCC parsers are table driven, unless you rewrite SmaCC to produce top-down parsers, it will not reach the performance of RBParser or the parser compiled by PetitParser Compiler. Cheers Jan On Fri, Aug 19, 2016, 14:50 Johan Fabry <jfabry(a)dcc.uchile.cl> wrote:

...

On Aug 19, 2016, at 07:02, stepharo <stepharo(a)free.fr> wrote: Le 14/8/16 à 22:31, Tudor Girba a écrit : > Hi, > > PetitParser is a dynamic parser. SmaCC is a static parser. Both have

their usages, and the static parsers are significantly faster. The speed problem of PetitParser will likely be significantly improved once we get the work of Jan on compiling parsers. _______________________________________________ Moose-dev mailing list Moose-dev(a)list.inf.unibe.ch https://www.list.inf.unibe.ch/listinfo/moose-dev

Thierry Goubier

4:06 p.m.

Hi Jan, Le 20/08/2016 à 15:36, Jan Kurš a écrit :

...

On the Smalltalk grammar, SmaCC is two times faster than PetitParser and Compiled PetitParser is four times faster. RBParser is like four point one times faster.

Impressive results. Are you also changing the O() characteristics of the PetitParser-based parsers (i.e. that tendency to have performance/memory behavior based on the input length and not on the tree depth) ?

...

The SmaCC parsers are table driven, unless you rewrite SmaCC to produce top-down parsers, it will not reach the performance of RBParser or the parser compiled by PetitParser Compiler.

No. What you have to do with SmaCC is generate(*) a recursive ascent parser[1] instead of a table-driven parser. Regards, Thierry [1] https://en.wikipedia.org/wiki/Recursive_ascent_parser (*) This is an exercice left to the reader :)

...

Cheers Jan On Fri, Aug 19, 2016, 14:50 Johan Fabry <jfabry(a)dcc.uchile.cl <mailto:jfabry@dcc.uchile.cl>> wrote: I did not read this before, but now I’m interested. Dory, when you say that 'the static parsers are significantly faster’ can you give some numbers on that? And you know what the numbers are compared to the compiled petitparsers of Jan and Jan ? (Their compiler is really cool BTW :-) I use it for LRP ) -- Does this mail seem too brief? Sorry for that, I don’t mean to be rude! Please see http://emailcharter.org . Johan Fabry - http://pleiad.cl/~jfabry PLEIAD and RyCh labs - Computer Science Department (DCC) - University of Chile

On Aug 19, 2016, at 07:02, stepharo <stepharo(a)free.fr

<mailto:stepharo@free.fr>> wrote:

Le 14/8/16 à 22:31, Tudor Girba a écrit : > Hi, > > PetitParser is a dynamic parser. SmaCC is a static parser. Both

have their usages, and the static parsers are significantly faster. The speed problem of PetitParser will likely be significantly improved once we get the work of Jan on compiling parsers. _______________________________________________ Moose-dev mailing list Moose-dev(a)list.inf.unibe.ch <mailto:Moose-dev@list.inf.unibe.ch> https://www.list.inf.unibe.ch/listinfo/moose-dev _______________________________________________ Moose-dev mailing list Moose-dev(a)list.inf.unibe.ch https://www.list.inf.unibe.ch/listinfo/moose-dev

Johan Fabry

5:12 p.m.

...

On Aug 20, 2016, at 10:36, Jan Kurš <kurs.jan(a)gmail.com> wrote: On the Smalltalk grammar, SmaCC is two times faster than PetitParser and Compiled PetitParser is four times faster. RBParser is like four point one times faster. The SmaCC parsers are table driven, unless you rewrite SmaCC to produce top-down parsers, it will not reach the performance of RBParser or the parser compiled by PetitParser Compiler. Cheers Jan

Jan Kurš

21 Aug 21 Aug

12:13 p.m.

I added some details to the webpage for those, who are interested: http://scg.unibe.ch/research/petitcompiler @Thierry: Thanks! I am not changing any complexity characteristics. The speedup is gained by analyzing a grammar and optimizing some parts. E.g. replacing combinators with a while loop, or using the first-set to choose the correct branch without backtracking (if possible). On Sat, Aug 20, 2016 at 5:12 PM Johan Fabry <jfabry(a)dcc.uchile.cl> wrote:

...

Thanks for the numbers, Jan. And they are excellent numbers as well! :-) Congrats on the compiler, it is simply great ! -- Does this mail seem too brief? Sorry for that, I don’t mean to be rude! Please see http://emailcharter.org . Johan Fabry - http://pleiad.cl/~jfabry PLEIAD and RyCh labs - Computer Science Department (DCC) - University of Chile On Aug 20, 2016, at 10:36, Jan Kurš <kurs.jan(a)gmail.com> wrote: On the Smalltalk grammar, SmaCC is two times faster than PetitParser and Compiled PetitParser is four times faster. RBParser is like four point one times faster. The SmaCC parsers are table driven, unless you rewrite SmaCC to produce top-down parsers, it will not reach the performance of RBParser or the parser compiled by PetitParser Compiler. Cheers Jan _______________________________________________ Moose-dev mailing list Moose-dev(a)list.inf.unibe.ch https://www.list.inf.unibe.ch/listinfo/moose-dev

Thierry Goubier

12:45 p.m.

Hi Jan, Le 21/08/2016 à 12:13, Jan Kurš a écrit :

...

I added some details to the webpage for those, who are interested: http://scg.unibe.ch/research/petitcompiler

Thanks.

...

@Thierry: Thanks! I am not changing any complexity characteristics. The speedup is gained by analyzing a grammar and optimizing some parts. E.g. replacing combinators with a while loop, or using the first-set to choose the correct branch without backtracking (if possible).

This one (avoiding backtracking) does change the complexity in a good way (reduces it). A nice optimisation. Thierry

...

On Sat, Aug 20, 2016 at 5:12 PM Johan Fabry <jfabry(a)dcc.uchile.cl <mailto:jfabry@dcc.uchile.cl>> wrote: Thanks for the numbers, Jan. And they are excellent numbers as well! :-) Congrats on the compiler, it is simply great ! -- Does this mail seem too brief? Sorry for that, I don’t mean to be rude! Please see http://emailcharter.org . Johan Fabry - http://pleiad.cl/~jfabry PLEIAD and RyCh labs - Computer Science Department (DCC) - University of Chile

On Aug 20, 2016, at 10:36, Jan Kurš <kurs.jan(a)gmail.com <mailto:kurs.jan@gmail.com>> wrote: On the Smalltalk grammar, SmaCC is two times faster than PetitParser and Compiled PetitParser is four times faster. RBParser is like four point one times faster. The SmaCC parsers are table driven, unless you rewrite SmaCC to produce top-down parsers, it will not reach the performance of RBParser or the parser compiled by PetitParser Compiler. Cheers Jan

_______________________________________________ Moose-dev mailing list Moose-dev(a)list.inf.unibe.ch <mailto:Moose-dev@list.inf.unibe.ch> https://www.list.inf.unibe.ch/listinfo/moose-dev _______________________________________________ Moose-dev mailing list Moose-dev(a)list.inf.unibe.ch https://www.list.inf.unibe.ch/listinfo/moose-dev

2808

days inactive

2817

days old

moose-dev@list.inf.unibe.ch

Manage subscription

12 comments

8 participants

tags (0)

participants (8)

Alexandre Bergel
Jan Kurš
Johan Fabry
Serge Stinckwich
serge.stinckwich＠gmail.com
stepharo
Thierry Goubier
Tudor Girba