RE: Getting involved in Bison

[Morales Cayuela, Victor (NSB - CN/Hangzhou)]

Hello!


> I recommend that you first build by hand a push parser before trying to get 
> this done by m4.  For instance take examples/c++/calc++, and change the 
> generated parser so that it works as a push parser.  Keep this preciously 
> somewhere to make sure "make" doesn't kill it.  When it works, you can show 
> us a diff between the regular pull parser and the push parser, and when we 
> agree, we can proceed to make it work in m4.

Ok, I will first create a draft parser, review it with everyone and then make 
it work in m4. I will use Java as reference, should be closer to the final 
solution than C.
I have been fiddling this weekend with the push parsers and some examples, but 
I didn't look at pushcalc. Although I have used a lot Bison I am still 
unfamiliar with jargon and some notions. Thanks for all the suggestions and 
examples.

> Maybe you should start with that: find a means to benchmark two pull parsers: 
> one off-the-shelf, generated by today's lalr1.cc, and another one where the 
> local variables that need to be member variables in push-mode are made member 
> variables.

What is the expected difference in performance between local and member 
variables? Besides constructor creation/destruction and variable lifetime, I 
would say there should not be other issues. Anyway let's test both 😊



Related to this topic:
#################
Do you remember Akim that I told you that in my Mac tests did not pass 
completely but in yours they did? Specifically all in these categories:
        LALR(1) Calculator
        LALR(1) C++ Calculator

Since this time will I need them, I decided to checked them. Seems that in Mac 
`wc -l` indents the result with the number of lines, mismatching with the 
expected pattern:
        tests/testsuite.log:
        ./calc.at:1111: grep -v 'Return for a new token:' stderr | wc -l 
        --- -   2020-02-22 13:15:01.000000000 +0800 
        +++ 
/Users/Victor/Projects/bison/tests/testsuite.dir/at-groups/456/stdout       
2020-02-22 13:15:01.000000000 +0800 
        @@ -1,2 +1,2 @@
        -0
        +       0
Has someone previously reported this? Seems I am predestined to deal with 
indentations XD



Separate topic about C++:
#####################
I believe we could improve a bit the C++ generated code and standardize it with 
modern syntax.

For example, C++ does no longer recommend constructing objects with 
parenthesis, but with braces. I saw yesterday this statement:
        symbol_type yylookahead (yylex ());
Which should be rewritten as:
        symbol_type yylookahead {yylex ()};

There are also some keywords that help the compiler optimizing. For example, 
this structure has a default constructor:
        struct by_type  { /* Default constructor */ by_type (); ... }
We might add "= default" (or even "= delete" if it is not used and we prefer to 
avoid it being created without parameters):
        struct by_type  { /* Default constructor */ by_type () = default; ... }

We can discuss all this after the push parser.


Regards,
Victor

-----Original Message-----
From: Akim Demaille <address@hidden> 
Sent: 2020年2月23日 15:08
To: Morales Cayuela, Victor (NSB - CN/Hangzhou) <address@hidden>
Cc: Bison Bugs <address@hidden>; Bison Patches <address@hidden>
Subject: Re: Getting involved in Bison

Hi Victor!

> Le 20 févr. 2020 à 04:07, Morales Cayuela, Victor (NSB - CN/Hangzhou) 
> <address@hidden> a écrit :
> 
> Hi!
> 
>> It really depends what you'd like to do.  You reported you're fluent in C++. 
>>  Something quite interesting would be to enable push-parser in lalr1.cc.  
>> It's more ambitious than the subcomplain task, but it will probably take you 
>> less time, because it's more compact, it will impact much less code.
> 
> Ok, seems very interesting! I use C++ (currently C++17) in my daily job as a 
> developer and I always try to keep up with the latest of the standard.
> I will familiarize myself first with push/pull parsers and then I will start 
> with it 😊 I will also take a look at TODO to check if there is already any 
> specification about how it should look like.

I think all the information you need is in the manual, in the push example in C 
(see examples/c/pushcalc), and in yacc.c's implementation of push.  Actually, 
Java also has push-parser support, and is probably a better source of 
inspiration than C.

I recommend that you first build by hand a push parser before trying to get 
this done by m4.  For instance take examples/c++/calc++, and change the 
generated parser so that it works as a push parser.  Keep this preciously 
somewhere to make sure "make" doesn't kill it.  When it works, you can show us 
a diff between the regular pull parser and the push parser, and when we agree, 
we can proceed to make it work in m4.

Actually, maybe examples/c++/calc++ is too bloated: first create a very simple 
example in C++, similar to examples/c/calc.  Start from this one, it is much 
simpler, and yet sufficient.

Most of the effort, I guess, is finding what local variables must be turned 
into member variables, to survive across calls to yyparse (amusingly enough 
coroutines would make this easy, but we can't afford them yet ;-).  But 
lalr1.java should help.

Note that C++ allows to consider other options than the C parsers.  Since the 
C++ parser is an object, maybe we can have push and pull use the same object.  
I mean, *maybe*, we can *always* move most local variables into member 
variables, even for regular pull-parsing.  Of course it's technically possible, 
the "maybe" refers to the runtime cost: *if* benchmarks show that it is not a 
regression to move to a model with more member variables and less local 
variables, then the m4 part will be much easier.

Maybe you should start with that: find a means to benchmark two pull parsers: 
one off-the-shelf, generated by today's lalr1.cc, and another one where the 
local variables that need to be member variables in push-mode are made member 
variables.

How does that sound?