January 2004 - Hugs-Users

version naming convention
by Jens Petersen 25 Feb '04

25 Feb '04

I would like to make the suggestion that future releases of Hugs are made with a version naming convention that is more sort friendly. The current "MonYear" versioning scheme is problematic for packagers. ie the current release would be better called say "200311" (using Debian's hugs98 package versioning scheme) than "Nov2003", since this sorts much more easily. Jens

3 5

Re: New moderation policy and
by Sven Panne 25 Jan '04

25 Jan '04

Byron Hale wrote: > [...] I notice that the Hugs-users/Hugs-bugs lists do not have any identifying > header. I don't think there is a real need for prefixing the subject, the List-Post: header field works extremely well for mailman-based lists, even in the case of CC:-ing. > [...] I want to see my lists do something like make it difficult to harvest > email addresses from lists. That could be by mangling the return > address so that human intervention was needed. IMHO these are helpless measures doomed to fail. The only real place for stopping spam is at the receiving mail server (i.e. the one for haskell.org in our case). It is the only place where one still has a slight idea where the mail is really from and can apply some countermeasures, e.g. using RBLs, verifying the hostname via a reverse lookup etc. Doing this for Postfix (which seems to be used at haskell.org) is only a matter of a few lines in the configuration file. I'm not sure if haskell.org actually does something like this, but it probably should. And einfo.com, should do it, too, of course. :-) In conjuction with a well-trained adaptive junk mail filter like the one in Netscape, spam is not really a problem, at least for me... Cheers, S.

1 0

New moderation policy
by John Peterson 16 Jan '04

16 Jan '04

I'm about to take off for the holidays and I'm going to change the spam policy for the hugs mailing lists. Previously, non-listmember mail was held for approval. As part of the approval process I would whitelist email addresses so that you only had messages held the first time you post. This only affects people with multiple email addresses who are posting from an address different from thier subscription address. Since I'll be gone for a couple of weeks, I've set the list to reject instead of approve non-member email. This means that you're going to be on your own to modify your subscription address if there are problems with posting. I'll still take private email requests to whitelist addresses but I won't be attending to email much for the next few weeks. Over the last few months I've observed fewer and fewer problems with legitimate posts being held and more and more spam being held for approval so I'm hoping that this new policy will make my life easier without being a burden to too many posters. Keeping your email spam free, John

2 1

Re: make check fails, no tests directory
by nv3006802＠sneakemail.com 13 Jan '04

13 Jan '04

I haven't heard anything back, but tar ztf hugs98-Nov2003.tar.gz | grep tests reports nothing. I don't want to mess with CVS, can someone fix the tar file? Thanks. On 9 Jan 2004 06:49:17 -0000, nv3006802(a)sneakemail.com said: > Hello, > > I'm building Hugs 98 Nov 2003 on IRIX64. When I run 'make check' it > fails because there is no '../tests' directory. In fact, there is no > 'tests' directory at all. Is it simply missing from the distribution? > > Please respond by email as well. > > Thanks,

1 0

Re: Compiling HXML toolbox under Hugs/Windows
by Ross Paterson 12 Jan '04

12 Jan '04

On Mon, Jan 12, 2004 at 11:59:58AM +0000, Graham Klyne wrote: > (1) hdom/xmltreefilter.hs, incorrect section syntax: > line 555, changed to: > del1Attr an = processAttrl ((none `when` isAttr an) $$) > line 613, changed to: > = processAttrl ((modifyValue `when` isAttr an) $$) This is a known bug in Hugs (User's Guide 4.1.2). > (3) hparser/xmlinput.hs, line 38: spurious ',' Accepting this was a bug in old versions of GHC that has since been fixed. > (4) Missing module 'Socket'. > [[ > Reading file "..\hparser\XmlInput.hs": > Parsing > ERROR "..\hparser\XmlInput.hs" - Can't find imported module "Socket" > ]] > Is this a GHC/Hugs library difference? Should this be Network.Socket? I > tried using that and it seemed to be accepted. Network.Socket is the modern name. > (6) I think a probem with the build instructions in README: > [[ > Just add the modules from the directories "hdom", "hparser", "hvalidator", > "hxpath", "http" and > "parsec" to the path of your compiler or interpreter, the Makefile contains > an example. > It is planned to provide a GHC package of the Haskell XML Toolbox in the > near future. > An example ghci project file ".ghci" can be found in the examples directory. > ]] > does not mention directory popen. GHC has POpen in hslibs. > (7) in POpen.hs: > Module Posix should be Text.Regex.Posix? No, System.Posix, but the Hugs version is incomplete (lacks forkProcess).

1 0

Re: Problems building experimental Unicode version of Hugs on Windows
by Ross Paterson 10 Jan '04

10 Jan '04

On Fri, Jan 09, 2004 at 05:52:47PM +0000, Graham Klyne wrote: > This message notes some edits I'm making to try and build the Dec2003 Hugs > source kit with Unicode under MS-Windows (because I want to try and use > HXML toolbox under Hugs). It also describes a problem I'm having with the > experimental Unicode support under Windows. Thanks for trying this out, and for all the details. I believe the bugs with STDCALL_SUPPORTED and option corruption are fixed in the CVS version. You might like to try a new snapshot from http://cvs.haskell.org/Hugs/downloads/snapshots/hugs98-20040109.tar.gz (This version also has a dynamically growing heap, I believe.)

2 1