Bug#936146: archivemail - Python 3 porting
On Tue, Aug 04, 2020 at 11:12:51AM +0100, Jonathan Dowland wrote: > On Mon, Jul 27, 2020 at 12:37:36AM -0400, Sandro Tosi wrote: > > do you have any plan on completing this port? I'm not a user of > > archivemail but it looks like it should be removed, not salvaged: > > > > * no new upstream releases since 2011 (!) > > * last upload to debian in 2014 > > * retired from fedora: https://bugzilla.redhat.com/show_bug.cgi?id=1777616 > > > > maybe it's time to let it go? > > > > If i dont hear otherwise in a week, i'll file for its removal > > I think I'm happy for it to go. I'm certainly not going to be able to > work on porting it, and I'm most likely going to move to something else > for the purposes I use it. Since there were no further objections, I've just filed a removal bug. Cheers, Moritz
Bug#936146: archivemail - Python 3 porting
Hi! I used this frequently enough to be hesitant to let apt dist-upgrade remove it on my system currently. But I understand it may not make sense to port it as its hard and upstream is not active anymore. Is there an alternative in Debian repository? Best, -- Martin
Bug#936146: archivemail - Python 3 porting
On Mon, Jul 27, 2020 at 12:37:36AM -0400, Sandro Tosi wrote: do you have any plan on completing this port? I'm not a user of archivemail but it looks like it should be removed, not salvaged: * no new upstream releases since 2011 (!) * last upload to debian in 2014 * retired from fedora: https://bugzilla.redhat.com/show_bug.cgi?id=1777616 maybe it's time to let it go? If i dont hear otherwise in a week, i'll file for its removal I think I'm happy for it to go. I'm certainly not going to be able to work on porting it, and I'm most likely going to move to something else for the purposes I use it. Thanks, -- 👱🏻 Jonathan Dowland ✎j...@debian.org 🔗 https://jmtd.net
Bug#936146: archivemail - Python 3 porting
On Mon, 27 Jul 2020, Sandro Tosi wrote: On Thu, 28 May 2020 13:56:59 -0400 (EDT) Scott Talbert wrote: On Tue, 26 May 2020, Jonathan Dowland wrote: archivemail seems to be a good candidate to RM due to dead upstream. However, it still has a relatively high popcon, so people seem to be using it. I'm willing to take a stab at porting to Python 3 if anyone is available to test it? The port effort doesn't look that bad at first glance, but I don't use this package. I'm happy to test anything you produce, but I'd warn you that I think it's quite a significant piece of work. From what I remember when I last looked at hacking a feature into it (#736327), archivemail uses the older of two different APIs provided by the python "mailbox" library, and only the newer one was carried forward to Python 3. So moving away from that older API is a big part of the work. You were right. This was harder than I expected. :) Mainly it is exactly as you described - the rfc822.Message class (which doesn't exist in Python 3) does not map exactly to the email.message.Message class. I'm stuck at the moment with figuring out how to calculate the message size. In rfc822.Message, you could access the file handle directly and get the size that way. do you have any plan on completing this port? I'm not a user of archivemail but it looks like it should be removed, not salvaged: * no new upstream releases since 2011 (!) * last upload to debian in 2014 * retired from fedora: https://bugzilla.redhat.com/show_bug.cgi?id=1777616 maybe it's time to let it go? If i dont hear otherwise in a week, i'll file for its removal No, I kind of gave up. It did turn out to be harder than I expected, and some of the APIs in the older email APIs don't exist in the new ones, so it wasn't clear what to do there. I would say it can go, but I don't use it so my opinion doesn't matter much. Scott
Bug#936146: archivemail - Python 3 porting
Hey Scott and Jonathan, On Thu, 28 May 2020 13:56:59 -0400 (EDT) Scott Talbert wrote: > On Tue, 26 May 2020, Jonathan Dowland wrote: > > >> archivemail seems to be a good candidate to RM due to dead upstream. > >> However, it still has a relatively high popcon, so people seem to be using > >> it. > >> > >> I'm willing to take a stab at porting to Python 3 if anyone is available to > >> test it? The port effort doesn't look that bad at first glance, but I > >> don't use this package. > > > > I'm happy to test anything you produce, but I'd warn you that I think > > it's quite a significant piece of work. From what I remember when I last > > looked at hacking a feature into it (#736327), archivemail uses the > > older of two different APIs provided by the python "mailbox" library, > > and only the newer one was carried forward to Python 3. So moving away > > from that older API is a big part of the work. > > You were right. This was harder than I expected. :) Mainly it is > exactly as you described - the rfc822.Message class (which doesn't exist > in Python 3) does not map exactly to the email.message.Message class. I'm > stuck at the moment with figuring out how to calculate the message size. > In rfc822.Message, you could access the file handle directly and get the > size that way. do you have any plan on completing this port? I'm not a user of archivemail but it looks like it should be removed, not salvaged: * no new upstream releases since 2011 (!) * last upload to debian in 2014 * retired from fedora: https://bugzilla.redhat.com/show_bug.cgi?id=1777616 maybe it's time to let it go? If i dont hear otherwise in a week, i'll file for its removal Regards, -- Sandro "morph" Tosi My website: http://sandrotosi.me/ Me at Debian: http://wiki.debian.org/SandroTosi Twitter: https://twitter.com/sandrotosi
Bug#936146: archivemail - Python 3 porting
On Tue, 26 May 2020, Jonathan Dowland wrote: archivemail seems to be a good candidate to RM due to dead upstream. However, it still has a relatively high popcon, so people seem to be using it. I'm willing to take a stab at porting to Python 3 if anyone is available to test it? The port effort doesn't look that bad at first glance, but I don't use this package. I'm happy to test anything you produce, but I'd warn you that I think it's quite a significant piece of work. From what I remember when I last looked at hacking a feature into it (#736327), archivemail uses the older of two different APIs provided by the python "mailbox" library, and only the newer one was carried forward to Python 3. So moving away from that older API is a big part of the work. You were right. This was harder than I expected. :) Mainly it is exactly as you described - the rfc822.Message class (which doesn't exist in Python 3) does not map exactly to the email.message.Message class. I'm stuck at the moment with figuring out how to calculate the message size. In rfc822.Message, you could access the file handle directly and get the size that way. Scott
Bug#936146: archivemail - Python 3 porting
On Thu, May 14, 2020 at 03:31:31PM -0400, Scott Talbert wrote: archivemail seems to be a good candidate to RM due to dead upstream. However, it still has a relatively high popcon, so people seem to be using it. I'm willing to take a stab at porting to Python 3 if anyone is available to test it? The port effort doesn't look that bad at first glance, but I don't use this package. I'm happy to test anything you produce, but I'd warn you that I think it's quite a significant piece of work. From what I remember when I last looked at hacking a feature into it (#736327), archivemail uses the older of two different APIs provided by the python "mailbox" library, and only the newer one was carried forward to Python 3. So moving away from that older API is a big part of the work. -- 👱🏻 Jonathan Dowland ✎j...@debian.org 🔗 https://jmtd.net
Bug#936146: archivemail - Python 3 porting
archivemail seems to be a good candidate to RM due to dead upstream. However, it still has a relatively high popcon, so people seem to be using it. I'm willing to take a stab at porting to Python 3 if anyone is available to test it? The port effort doesn't look that bad at first glance, but I don't use this package. Scott