From ta-meyer at ihug.co.nz Mon Sep 1 17:55:26 2003 From: ta-meyer at ihug.co.nz (Tony Meyer) Date: Mon Sep 1 00:56:02 2003 Subject: [Email-SIG] Email package for Python < 2.2.2 Message-ID: <1ED4ECF91CDED24C8D012BCF2B034F13026F28E4@its-xchg4.massey.ac.nz> Now that the sourceforge mimelib project has (effectively) closed down, is there anywhere that people can get hold of the email package if they are using Python < 2.2.2? Thanks, Tony Meyer From sdrees at sdrees.de Mon Sep 1 08:16:57 2003 From: sdrees at sdrees.de (Stefan Drees) Date: Mon Sep 1 01:17:00 2003 Subject: [Email-SIG] Email package for Python < 2.2.2 In-Reply-To: <1ED4ECF91CDED24C8D012BCF2B034F13026F28E4@its-xchg4.massey.ac.nz> References: <1ED4ECF91CDED24C8D012BCF2B034F13026F28E4@its-xchg4.massey.ac.nz> Message-ID: <20030901051657.GA32559@knoten.biz> On Mon, Sep 01, 2003 at 04:55:26PM +1200 - a wonderful day - Tony Meyer wrote: > Now that the sourceforge mimelib project has > (effectively) closed down, is there anywhere that people > can get hold of the email package if they are > using Python < 2.2.2? ... may I suggest a place on starship.python.net? Opinions? All the best, Stefan Drees. -- Stefan Drees, sdrees@python.net, www.sdrees.biz Fingerprint = 516C C4EF 712A B26F 15C9 C7B7 5651 6964 D508 1B56 From barry at python.org Mon Sep 1 19:40:08 2003 From: barry at python.org (Barry Warsaw) Date: Mon Sep 1 14:40:09 2003 Subject: [Email-SIG] Email package for Python < 2.2.2 In-Reply-To: <1ED4ECF91CDED24C8D012BCF2B034F13026F28E4@its-xchg4.massey.ac.nz> References: <1ED4ECF91CDED24C8D012BCF2B034F13026F28E4@its-xchg4.massey.ac.nz> Message-ID: <1062441575.12564.2.camel@anthem> On Mon, 2003-09-01 at 00:55, Tony Meyer wrote: > Now that the sourceforge mimelib project has (effectively) closed down, is > there anywhere that people can get hold of the email package if they are > using Python < 2.2.2? I've tried to get away with not releasing the standalone email package any more, but I've had a number of requests for it for Python 2.1, so I think I'll have to reverse my decision here. Note that, the distutils package is available in a roundabout way: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/mailman/mailman/misc/ I distribute it with Mailman so the latest version is always checked into the Mailman cvs. Here's a thought: we could make the distutils package available on the sig pages. I'll go ahead and do that. -Barry From ta-meyer at ihug.co.nz Tue Sep 2 12:35:53 2003 From: ta-meyer at ihug.co.nz (Tony Meyer) Date: Mon Sep 1 19:36:33 2003 Subject: [Email-SIG] Email package for Python < 2.2.2 In-Reply-To: <1ED4ECF91CDED24C8D012BCF2B034F13031B6EED@its-xchg4.massey.ac.nz> Message-ID: <1ED4ECF91CDED24C8D012BCF2B034F130212ADE9@its-xchg4.massey.ac.nz> > I've tried to get away with not releasing the standalone > email package any more, but I've had a number of requests for > it for Python 2.1, so I think I'll have to reverse my > decision here. [...] > Here's a thought: we could make the distutils package available > on the sig pages. I'll go ahead and do that. Thanks Barry. The reason I asked was that (potential) spambayes users have been asking for it since the link on the spambayes website died. I'll update it to point to the email sig page. Thanks, Tony From esj at harvee.org Mon Sep 1 21:47:16 2003 From: esj at harvee.org (Eric S. Johansson) Date: Mon Sep 1 20:49:38 2003 Subject: [Email-SIG] Email package for Python < 2.2.2 In-Reply-To: <1ED4ECF91CDED24C8D012BCF2B034F130212ADE9@its-xchg4.massey.ac.nz> References: <1ED4ECF91CDED24C8D012BCF2B034F130212ADE9@its-xchg4.massey.ac.nz> Message-ID: <3F53E894.4010007@harvee.org> Tony Meyer explained: >>I've tried to get away with not releasing the standalone >>email package any more, but I've had a number of requests for >>it for Python 2.1, so I think I'll have to reverse my >>decision here. > > [...] > >>Here's a thought: we could make the distutils package available >>on the sig pages. I'll go ahead and do that. > > > Thanks Barry. The reason I asked was that (potential) spambayes users have > been asking for it since the link on the spambayes website died. I'll > update it to point to the email sig page. another thanks here as well. The camram project counts on it as well as we frequently trip over message parsing bugs. ---eric From barry at python.org Tue Sep 2 04:26:00 2003 From: barry at python.org (Barry Warsaw) Date: Mon Sep 1 23:26:01 2003 Subject: [Email-SIG] Email package for Python < 2.2.2 In-Reply-To: <1062441575.12564.2.camel@anthem> References: <1ED4ECF91CDED24C8D012BCF2B034F13026F28E4@its-xchg4.massey.ac.nz> <1062441575.12564.2.camel@anthem> Message-ID: <1062473136.20227.27.camel@geddy> On Mon, 2003-09-01 at 14:39, Barry Warsaw wrote: > Here's a thought: we could make the distutils package available on the > sig pages. I'll go ahead and do that. Done. -Barry From bac at OCF.Berkeley.EDU Wed Sep 3 00:25:13 2003 From: bac at OCF.Berkeley.EDU (Brett C.) Date: Mon Sep 8 10:28:05 2003 Subject: [Email-SIG] Thread support and XML output request Message-ID: <3F558949.30005@ocf.berkeley.edu> Late last week I wrote a script to calculate what people and threads have caused me the most amount of work on the python-dev Summaries. I initially thought of using the email package, but I had the problem of needing to know what emails contributed to a thread. I did not see that in the email package and so I just parsed the Mailman archive page in thread view to garner that info. Now I would like to not have to do HTML parsing to figure this kind of thing out. Obviously Mailman has an algorithm to figure out what emails belong in a thread. Any chance of getting a similar algorithm added to the email package? And on top of this I think it would be nice to have a standardized XML format for emails. Yes, you can just read the mbox file and pass it to the email package, but why go through all of that parsing of data and such when you could do it once and then dump the parsed data to XML? And if threading was supported it would serve as a great way of storing that info instead of having to recalculate all of that for every read of the mbox. Mailman could, for instance, just provide an XML output of its parsed email info so that poor folks like me don't have to parse the HTML pages to get info. Now whether any of this is practical is beyond me; I am just a simple email user who probably has non-typical needs. But it sure would be nice to have. =) -Brett From anthony at interlink.com.au Wed Sep 10 00:55:33 2003 From: anthony at interlink.com.au (Anthony Baxter) Date: Tue Sep 9 09:56:02 2003 Subject: [Email-SIG] My plans for the parser Message-ID: <200309091355.h89DtXjp023121@localhost.localdomain> I'd just thought I'd drop a brief note with my basic plans for the email parser. I plan on a parser mode (or maybe we'll end up with multiple parsers) that is an absolute "best effort" parser - that is, it will always produce a message structure regardless of how mangled the message is. At the moment I'm leaning towards an attribute that includes warnings and the like that we encountered in parsing the message. For most (all) of the requirements that I have for this, it's completely useless to me to simply have the parser say "well, something was wrong with the message, so I can't show you anything about it". Show me what you can, let me know what you can't, and I'll deal with the results. Other things I'd like to add - I'd like to make sure we can handle unicode strings as inputs for everything appropriate (I understand that we don't currently do this). Anthony From matt at mondoinfo.com Tue Sep 9 16:07:58 2003 From: matt at mondoinfo.com (Matthew Dixon Cowles) Date: Tue Sep 9 16:08:26 2003 Subject: [Email-SIG] My plans for the parser In-Reply-To: <200309091355.h89DtXjp023121@localhost.localdomain> References: <200309091355.h89DtXjp023121@localhost.localdomain> Message-ID: <1063135885.33.541@sake.mondoinfo.com> Dear Anthony, > I'd just thought I'd drop a brief note with my basic plans for the > email parser. I plan on a parser mode (or maybe we'll end up with > multiple parsers) that is an absolute "best effort" parser - that > is, it will always produce a message structure regardless of how > mangled the message is. I think that's grand. You probably already know it but I've found that, very occasionally, Python's base-64 decoder will raise an error where a laxer one will produce results. I have one of those and you're welcome to it if you want, but you can probably do better off the top of your head. Regards, Matt From anthony at interlink.com.au Wed Sep 10 16:13:14 2003 From: anthony at interlink.com.au (Anthony Baxter) Date: Wed Sep 10 01:13:41 2003 Subject: [Email-SIG] My plans for the parser In-Reply-To: <1063135885.33.541@sake.mondoinfo.com> Message-ID: <200309100513.h8A5DFRh030650@localhost.localdomain> >>> Matthew Dixon Cowles wrote > I think that's grand. You probably already know it but I've found > that, very occasionally, Python's base-64 decoder will raise an error > where a laxer one will produce results. Hm. Have you got an example of one of these? I think I recall hearing about the problem... > I have one of those and > you're welcome to it if you want, but you can probably do better off > the top of your head. Is it in C or Python? Base64 decoding probably does need to be in C for speed. Anthony -- Anthony Baxter It's never too late to have a happy childhood. From matt at mondoinfo.com Wed Sep 10 11:55:44 2003 From: matt at mondoinfo.com (Matthew Dixon Cowles) Date: Wed Sep 10 12:23:09 2003 Subject: [Email-SIG] My plans for the parser In-Reply-To: <200309100513.h8A5DFRh030650@localhost.localdomain> References: <1063135885.33.541@sake.mondoinfo.com> <200309100513.h8A5DFRh030650@localhost.localdomain> Message-ID: <1063208896.51.1656@sake.mondoinfo.com> Dear Anthony, >> You probably already know it but I've found that, very >> occasionally, Python's base-64 decoder will raise an error where a >> laxer one will produce results. > Hm. Have you got an example of one of these? I think I recall > hearing about the problem... The only examples that I can find at the moment are single-part messages that are base-64 encoded but had ASCII footers stuck naively on the bottom. >> I have one of those and you're welcome to it if you want, but you >> can probably do better off the top of your head. > Is it in C or Python? Base64 decoding probably does need to be in C > for speed. It's in Python and therefore really a lot slower than the base64 module. It's fast enough for my purposes since it gets called only if the base64 module raises an error. Regards, Matt From matt at mondoinfo.com Sat Sep 13 16:10:45 2003 From: matt at mondoinfo.com (Matthew Dixon Cowles) Date: Sat Sep 13 16:11:06 2003 Subject: [Email-SIG] Message threading Message-ID: <1063482061.73.1100@sake.mondoinfo.com> A while ago, Brett mentioned that he'd like to see support for message threading in the email module. I've been meaning to mention since then that I agree. I think it's a natural for email.Utils. I have an implementation of Jamie Zawinski's message threading algorithm that I've been using for more than a year and I'd be glad to donate it. Jamie may not be Barry's favorite person, but I think that the algorithm is pretty good. It's described at: http://www.jwz.org/doc/threading.html Regards, Matt From barry at python.org Sat Sep 13 18:33:01 2003 From: barry at python.org (Barry Warsaw) Date: Sat Sep 13 18:33:08 2003 Subject: [Email-SIG] Message threading In-Reply-To: <1063482061.73.1100@sake.mondoinfo.com> References: <1063482061.73.1100@sake.mondoinfo.com> Message-ID: <1063492381.19907.174.camel@anthem> On Sat, 2003-09-13 at 16:10, Matthew Dixon Cowles wrote: > Jamie may not be Barry's favorite person Because he's got an out-of-date anti-Mailman rant on his site? I'll give him the benefit of the doubt that he just hasn't gotten around to taking it down now that it's mostly obsolete . Quick story: Back in '94 before I joined CNRI, I went out to SF to interview with Lucid. I used their C++ development environment and of course was a huge Lemacs/XEmacs fan. They were looking for someone to take over its maintenance, as Jamie was looking to do something else. Had a very good interview and a nice weekend with the wife in SF, but I never got an offer. I guess it was a good thing though because in 3 months they were out of business. :) Jamie was the one who warned me that might happen, so I'll excuse both his anti-Mailman slant, and his becoming a bar owner (the dark side to us musicians :). > , but I think > that the algorithm is pretty good. It is. I'm +1 on adding support for threading to email 3.0. > It's described at: > > http://www.jwz.org/doc/threading.html -Barry From matt at mondoinfo.com Sat Sep 13 22:23:59 2003 From: matt at mondoinfo.com (Matthew Dixon Cowles) Date: Sat Sep 13 22:24:13 2003 Subject: [Email-SIG] Message threading In-Reply-To: <1063492381.19907.174.camel@anthem> References: <1063482061.73.1100@sake.mondoinfo.com> <1063492381.19907.174.camel@anthem> Message-ID: <1063504123.31.1169@sake.mondoinfo.com> [me] > Jamie may not be Barry's favorite person [Barry] > Because he's got an out-of-date anti-Mailman rant on his site? That was kind of what I was thinking. I agree with Jaimie a lot of the time but I think that he can be a trifle, um, immoderate. > I'll excuse both his anti-Mailman slant, and his becoming a bar > owner (the dark side to us musicians :). Becoming a bar owner is good. Without bar owners, I'd have to do all my drinking at home. > I'm +1 on adding support for threading to email 3.0. Would you like me to extract my implementation of his algorithm and clean it up so that you can look it over and decide whether or not you want to use it as a basis for what goes into the module? I imagine that we'd want a function that took a list of Message objects and returned them after threading them. (It doesn't seem to me that there's any value to merely munging the original list; it's the Message objects that will be big here.) I expect that we'd want a couple of arguments: A function to sort siblings (default to soring by the date in the date header). An optional callback function to report progress since sorting a large mailbox can take a while. Perhaps an option to return either a flattened list or a list-of-lists tree. Regards, Matt From matt at mondoinfo.com Sat Sep 20 17:19:03 2003 From: matt at mondoinfo.com (Matthew Dixon Cowles) Date: Sat Sep 20 18:17:42 2003 Subject: [Email-SIG] Message threading code Message-ID: <1064091806.7.6056@sake.mondoinfo.com> I've uploaded my implementation of Jamie Zawinski's message threading algorithm to: http://www.mondoinfo.com/messagethread.py It took some significant surgery to extract it from its previous home, but I think that it's still correct. If anyone's interested in it, you're welcome to it as it stands. If Barry thinks it's a good start for something in the email module, I'll be happy to clean it up according to the python style guide and assign copyright to the PSF. Regards, Matt From barry at python.org Mon Sep 22 18:14:14 2003 From: barry at python.org (Barry Warsaw) Date: Mon Sep 22 18:14:24 2003 Subject: [Email-SIG] [Fwd: email module - deprecated functions (fwd)] Message-ID: <1064268853.1986.77.camel@anthem> Forwarding here, mostly so it gets a Subject header matching one of my virtual folders. ;) -Barry -------------- next part -------------- An embedded message was scrubbed... From: fdrake01@comcast.net Subject: email module - deprecated functions (fwd) Date: Mon, 22 Sep 2003 22:11:02 +0000 Size: 2106 Url: http://mail.python.org/pipermail/email-sig/attachments/20030922/fa964377/attachment.mht