|Formatting of Language LRMs email@example.com (Seima Rao) (2014-06-17)|
|Re: Formatting of Language LRMs firstname.lastname@example.org (Ivan Godard) (2014-06-20)|
|Re: Formatting of Language LRMs Pidgeot18@verizon.net.invalid (=?UTF-8?B?Sm9zaHVhIENyYW5tZXIg8J+Qpw==?=) (2014-06-22)|
|RE: Formatting of Language LRMs email@example.com (Costello, Roger L.) (2014-07-03)|
|Re: Formatting of Language LRMs firstname.lastname@example.org (Ivan Godard) (2014-07-03)|
|Re: syntax checkers, was Formatting of Language LRMs email@example.com (glen herrmannsfeldt) (2014-07-04)|
|From:||glen herrmannsfeldt <firstname.lastname@example.org>|
|Date:||Fri, 4 Jul 2014 17:35:34 +0000 (UTC)|
|Organization:||Aioe.org NNTP Server|
|References:||14-06-010 14-06-016 14-06-021 14-07-004 14-07-009|
|Keywords:||parse, syntax, comment|
|Posted-Date:||04 Jul 2014 14:00:39 EDT|
Ivan Godard <email@example.com> wrote:
> If the document were a compiler written in Java (i.e. a recognizer and
> checker) then formally there is no difference. Practically, an immense
> difference. VWG is an ASL for specifying languages. All of Algol68 in
> VWG was about four pages of appendix; a compiler for Algol68 written in
> Java would be much more than that, even after you strip out optimization
> and code generation.
There are programs written for OS/360 called "syntax checkers".
They are not compilers, but only enough of one to tell whether
the input is valid syntax in the specific language.
I never used the actual program, but the PLM (Program Logic Manual)
has the (I believe) BNF form for the language, unlike the normal
As far as I know, they were not normally used in batch systems, but
were for use by TSO.
Seems to me it should be possible to write a program that will
take BNF input, and a program that is supposed to be written in
that language, and will tell you whether or not it is.
[That's basically what yacc and other parser generators do, they
compile BNF or something similar into a program that recognizes input
in the BNF. Of course, for syntax checking you also have to tokenize
the input which is another whole can of worms. -John]
Return to the
Search the comp.compilers archives again.