On 12/1/06, Tomasz Zielonka wrote:

Do you accept contributions? I have some code I find very useful that would fit in the same places, like in Text.ParserCombinators.Parsec.Utils, Data.BitsUtils (btw, why not Data.Bits.Utils?), Control.Concurrent.*.

Hey, contributions. I'll throw in my haskell MIME parser if you want it. It's not the same as the one that most people use -- but I like it better. :) -- Taral "You can't prove anything." -- Gödel's Incompetence Theorem

On Fri, Dec 01, 2006 at 07:04:56AM +0100, Tomasz Zielonka wrote:

...

http://software.complete.org/missingh/wiki/TransitionPlanning

Your comments (and edits! -- must register/login first) are welcome.

As for other code (say Data.Tree.Utils), I am not sure what's best: put it in some big library like yours, or publish as separate small libraries. There is more work with the latter, but it seems more clean, and easy to review for the user.

Well, MissingH is going the direction of splitting up into smaller pieces, based on feedback received here, so I'd probably suggest that you keep that separate. Unless it's just one small/simple module. -- John

Malcolm Wallace wrote:

John Goerzen wrote:

...

Please tell me if I should just go away or go to another list here.

How about libraries@haskell.org?

Fair enough. Will do.

...

http://software.complete.org/missingh/wiki/TransitionPlanning

Your comments (and edits! -- must register/login first) are welcome.

About Text.ParserCombinators.Parsec.RFC2234 and the like. Although I can't see what this code does (no links on the wiki, and port 70 is firewalled here), I would be prepared to bet that you do not in fact define any new combinators for parsing. What you have there is a parser

Here's a snippet: This module provides parsers for the grammar defined in RFC2234, "Augmented BNF for Syntax Specifications: ABNF", http://www.faqs.org/rfcs/rfc2234.html. The terminal called char in the RFC is called character here to avoid conflicts with Parsec's char function. alpha :: CharParser st Char bit :: CharParser st Char character :: CharParser st Char cr :: CharParser st Char This one is not email-specific. The other two are. (BTW, they were written by Peter Simons)

for a specific little language. Even if it is implemented using parsec combinators, it does not belong with them.

Alternative suggestions: Language.RFC2234.Parse Language.ABNF.RFC2234.Parse Network.SMTP.RFC2821.Parse Network.Email.RFC2822.Parse

Does the above suggest a single good place for them to you? -- John

On Fri, Dec 01, 2006 at 01:30:25PM -0600, John Goerzen wrote a message of 46 lines which said:

This module provides parsers for the grammar defined in RFC2234, "Augmented BNF for Syntax Specifications: ABNF",

Do note that it is an old reference and the RFC on ABNF is 4234, issued one year ago. Also, there is another Parsec-based parser of ABNF in Haskell, Eustathius : http://www.bortzmeyer.org/eustathius-test-grammars.html