From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,XPRIO autolearn=ham
	autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,d875f8fc1f73cf09,start
X-Google-Attributes: gid103376,public
X-Google-ArrivalTime: 2001-04-21 12:22:29 PST
Path: 
 newsfeed.google.com!newsfeed.stanford.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.icl.net!dispose.news.demon.net!demon!btnet-peer0!btnet-feed5!btnet!mendelevium.btinternet.com!not-for-mail
From: "singlespeeder" <singlespeeder@32sixteen.com>
Newsgroups: comp.lang.ada
Subject: Compiling AFLEX 1.4a with Dec Ada 83 3.5-20 on Open VMS 7.1
Date: Sat, 21 Apr 2001 20:19:12 +0100
Organization: BT Internet
Message-ID: <9bsm5c$khe$1@uranium.btinternet.com>
NNTP-Posting-Host: host213-122-132-4.btinternet.com
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 5.50.4522.1200
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4522.1200
Xref: newsfeed.google.com comp.lang.ada:6830
Date: 2001-04-21T20:19:12+01:00
List-Id: <comp.lang.ada>

I've got hold of the source for AFLEX 1.4a and want to compile it to run on
Open VMS 7.1

The compiler I have access to is DEC Ada 83 3.5-20.

As far as I can tell there are two things in the code that require
modification.

1) AFLEX 1.4a assumes that the Ada 95 package Ada.Command_Line is available.
2) AFLEX 1.4a assumes that the Ada 95 implementation of Text_IO is
available, and uses the file Standard_Error. It also generates a debug
scanner that uses Standard_Error.

I've addressed the first issue by writing a package VAX_Command_Line that
wraps starlet, and I just used that in place of Ada.Command_Line - I only
had to change two packages, both of which were documented in the manual.

I'm having more trouble with the second issue. In the lex specification for
the scanner I currently get around the scanner assuimg the presence of
Standard_Error by the following kludge in my lex specification. (Note that
AFLEX 1.4a only has two user defined sections, not 4 as in some manuals)

-- definitions
%%
-- rules
%%
-- user defined section copied before yylex
package foo is
    standard_error : text_io.file_type;
    function yylex return token;
end foo;
package body foo is
##
-- user defined section copied after yylex
begin
    text_io.create (file => standard_error,
        mode => out_file,
        name => lexer_error.txt);
end foo;

So my package foo declares it's own standard_error, which it creates when
the package is elaborated. It always does this even when I instruct AFLEX
not to generate a debug scanner - in which case I get an empty
lexer_error.txt file.

Is there a better way of writing my lex specification so that I don't have
to have this kludge, or am I going to have to poke around in the source of
Aflex? I don't want to do this because I don't want to then have to generate
a set of tests for  AFLEX to make sure I haven't screwed it up.

Nick