A novice question

54 views
Skip to first unread message

撒闿要

unread,
Dec 25, 2018, 4:19:45 AM12/25/18
to nanopass-framework
In chez scheme source code , I notice that the very first nanopass language is  Lsrc in https://github.com/cisco/ChezScheme/blob/master/s/base-lang.ss,
and then there must be some parser which is used to convert raw scheme code to Lsrc , right? 
So my question is that where is the code , and how it works!  sorry for bearing with me!
Although it's a very simple question , but it takes me two day to look for it, but can not figure it out :(

Matt Jadud

unread,
Dec 25, 2018, 8:21:33 AM12/25/18
to 撒闿要, nanopass-framework
Hi Jellyzone,

Someone will probably know better. However, looking at the test driver:


The input pulls in Scheme. Then, the pass list is walked by the "run" procedure. 


LP is the first pass. So, wherever LP is defined...


and it seems to be there.

Anyone who is more familiar with the codebase may be able to give you a better answer. That's my guess from a quick read.

I have to wander off for a bit... good luck!

Cheers,
M




--
You received this message because you are subscribed to the Google Groups "nanopass-framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nanopass-framew...@googlegroups.com.
To post to this group, send email to nanopass-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/nanopass-framework/bffa951a-6064-46fe-88a9-e98183e9b8e9%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Andy Keep

unread,
Dec 25, 2018, 11:16:27 PM12/25/18
to Matt Jadud, 撒闿要, nanopass-framework
Chez Scheme reads source using the scheme reader (in read.ss) which has a few entry points: read, get-datum, get-datum/annotation, etc.  In the case of the compiler, it uses a reader that produces source annotations along with the scheme data so that we can track source positions for error reporting, profiling, and debugging.  This source intermediate representation is then handed to the expander (in syntax.ss) to turn into Lsrc records.

If you look in syntax.ss, you'll see a section that starts "(with-output-language (Lsrc Expr)" and ends a few hundred lines later.  The functions within this section are called by the expander to construct Lsrc records to had to the rest of the compiler.

-andy:)

撒闿要

unread,
Dec 26, 2018, 12:32:31 AM12/26/18
to nanopass-framework
hi, Matt Jadud, thank you for the replay, I`ll look in to it~ 

撒闿要

unread,
Dec 26, 2018, 12:38:03 AM12/26/18
to nanopass-framework
hi, Andy,  aha, this is what I want,  In racket , there is reader and expander,   and the chez scheme also have something like this!
In racket ,The reader must provide a func­tion called read-syntax.  
                The expander must provide a macro called #%module-begin. 
I'll look into it， thank you very much , save my day!!

On Wednesday, December 26, 2018 at 12:16:27 PM UTC+8, Andy Keep wrote:
Chez Scheme reads source using the scheme reader (in read.ss) which has a few entry points: read, get-datum, get-datum/annotation, etc.  In the case of the compiler, it uses a reader that produces source annotations along with the scheme data so that we can track source positions for error reporting, profiling, and debugging.  This source intermediate representation is then handed to the expander (in syntax.ss) to turn into Lsrc records.

If you look in syntax.ss, you'll see a section that starts "(with-output-language (Lsrc Expr)" and ends a few hundred lines later.  The functions within this section are called by the expander to construct Lsrc records to had to the rest of the compiler.

-andy:)



On December 25, 2018 at 8:21:34 AM, Matt Jadud (ma...@jadud.com) wrote:

Hi Jellyzone,

Someone will probably know better. However, looking at the test driver:


The input pulls in Scheme. Then, the pass list is walked by the "run" procedure. 


LP is the first pass. So, wherever LP is defined...


and it seems to be there.

Anyone who is more familiar with the codebase may be able to give you a better answer. That's my guess from a quick read.

I have to wander off for a bit... good luck!

Cheers,
M




On Tue, Dec 25, 2018 at 4:39 AM 撒闿要 <jell...@gmail.com> wrote:
In chez scheme source code , I notice that the very first nanopass language is  Lsrc in https://github.com/cisco/ChezScheme/blob/master/s/base-lang.ss,
and then there must be some parser which is used to convert raw scheme code to Lsrc , right? 
So my question is that where is the code , and how it works!  sorry for bearing with me!
Although it's a very simple question , but it takes me two day to look for it, but can not figure it out :(

--
You received this message because you are subscribed to the Google Groups "nanopass-framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nanopass-framework+unsub...@googlegroups.com.
--
You received this message because you are subscribed to the Google Groups "nanopass-framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nanopass-framework+unsub...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages