priorities for 0.7

priorities for 0.7

[Jameson Graef Rollins]

[-- Attachment #1: Type: text/plain, Size: 2986 bytes --]

Hey, folks.  Now that 0.6 is *finally* out the door, it's time to start
working on 0.7.  As Interim Distinguished co-Chief Vice Tyrant I have
been tasked with starting the discussion on 0.7 release priorities.

Below is just the beginning of list of things that have been discussed
on the list that we should consider merging or working on for 0.7.  It's
weighted towards stuff that I've worked on only because I'm more
familiar with the state of those patches.  Please send follow-ups to
this list to elaborate on certain things, or bring up other patches that
should be considered [0].  We can then push on Carl to merge these ahead
of the 0.7 freeze.

- notmuch-new atomicity:

  id:"1307822683-848-1-git-send-email-amdragon@mit.edu"

  I believe that the above is that latest version of this series; Austin
  can point us to the right one if it's not.  This is an important
  series that is long overdue to be merged.  It should definitely be a
  top priority.

- message/rfc822 part support:

  id:"1307120466-4980-1-git-send-email-jrollins@finestructure.net"
  id:"1307320169-29905-1-git-send-email-jrollins@finestructure.net"

  rfc822 part handling is a bit broken now after the notmuch-show
  rework.  The above patch series should have been included in 0.6 to
  fix a couple of regressions.  In the meantime Austin and I have
  started looking at further rework of notmuch-show to try to clean it
  up even more.  This rework might supersedes these patches, but it
  depends on how far we get.

- emacs: Add notmuch-show-refresh-view function

  id:"1306627784-3401-1-git-send-email-jrollins@finestructure.net"

- emacs: Add callback functions to crypto sigstatus button.

  id:"1306861633-3873-1-git-send-email-jrollins@finestructure.net"

- emacs: notmuch-hello improvements (Daniel Schoepe, Michal Sojka,
  Dmitry Kurochkin, Pieter Praet)

  These guys have been working on various improvements to the
  notmuch-hello page.  I'll let them comment on where they're at with
  that, and what should be merged.

- emacs: tagging race conditions (Austin Clements, Pieter Praet)

  Austin and Pieter have been looking at home to eliminate some race
  conditions when tagging from the emacs interface.  I'll let them
  comment on where that's at.

- emacs: default key bindings

  There's been a lot of discussion about default key bindings for the
  emacs UI.  This issue frankly terrifies me since I worry that it is
  basically impossible to converge on anything.  I think we'll need to
  see a specific full proposal before we can really consider what to do
  here.

- I know there are other old patches that are sinking in to the abyss.
  Hopefully any of those that are still relevant can be excavated.

jamie.

[0] It's maybe not mainline notmuch related, but we really need to get
some sort of tag sharing system worked out.  I can only imagine what
useful info is encoded in Carl's tags that would be invaluable for
everyone else to see, especially the release managers.

[-- Attachment #2: Type: application/pgp-signature, Size: 835 bytes --]

Re: priorities for 0.7

[Aneesh Kumar K.V]

On Wed, 06 Jul 2011 09:55:00 -0700, Jameson Graef Rollins <jrollins@finestructure.net> wrote:
Non-text part: multipart/mixed
Non-text part: multipart/signed
> Hey, folks.  Now that 0.6 is *finally* out the door, it's time to start
> working on 0.7.  As Interim Distinguished co-Chief Vice Tyrant I have
> been tasked with starting the discussion on 0.7 release priorities.
> 
> Below is just the beginning of list of things that have been discussed
> on the list that we should consider merging or working on for 0.7.  It's
> weighted towards stuff that I've worked on only because I'm more
> familiar with the state of those patches.  Please send follow-ups to
> this list to elaborate on certain things, or bring up other patches that
> should be considered [0].  We can then push on Carl to merge these ahead
> of the 0.7 freeze.
> 
> - notmuch-new atomicity:
> 
>   id:"1307822683-848-1-git-send-email-amdragon@mit.edu"
> 
>   I believe that the above is that latest version of this series; Austin
>   can point us to the right one if it's not.  This is an important
>   series that is long overdue to be merged.  It should definitely be a
>   top priority.
> 
> - message/rfc822 part support:
> 
>   id:"1307120466-4980-1-git-send-email-jrollins@finestructure.net"
>   id:"1307320169-29905-1-git-send-email-jrollins@finestructure.net"
> 
>   rfc822 part handling is a bit broken now after the notmuch-show
>   rework.  The above patch series should have been included in 0.6 to
>   fix a couple of regressions.  In the meantime Austin and I have
>   started looking at further rework of notmuch-show to try to clean it
>   up even more.  This rework might supersedes these patches, but it
>   depends on how far we get.
> 
> - emacs: Add notmuch-show-refresh-view function
> 
>   id:"1306627784-3401-1-git-send-email-jrollins@finestructure.net"
> 
> - emacs: Add callback functions to crypto sigstatus button.
> 
>   id:"1306861633-3873-1-git-send-email-jrollins@finestructure.net"
> 
> - emacs: notmuch-hello improvements (Daniel Schoepe, Michal Sojka,
>   Dmitry Kurochkin, Pieter Praet)
> 
>   These guys have been working on various improvements to the
>   notmuch-hello page.  I'll let them comment on where they're at with
>   that, and what should be merged.
> 
> - emacs: tagging race conditions (Austin Clements, Pieter Praet)
> 
>   Austin and Pieter have been looking at home to eliminate some race
>   conditions when tagging from the emacs interface.  I'll let them
>   comment on where that's at.
> 
> - emacs: default key bindings
> 
>   There's been a lot of discussion about default key bindings for the
>   emacs UI.  This issue frankly terrifies me since I worry that it is
>   basically impossible to converge on anything.  I think we'll need to
>   see a specific full proposal before we can really consider what to do
>   here.
> 
> - I know there are other old patches that are sinking in to the abyss.
>   Hopefully any of those that are still relevant can be excavated.


Don't overload same-window-regex

id:"1288200417-6584-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com"

Add reply-to header
id:"1288200417-6584-2-git-send-email-aneesh.kumar@linux.vnet.ibm.com"
I guess we want a generic form for this. I guess we want enable the
user to select what fields in the mail-header they would like to index

-aneesh

Re: priorities for 0.7

[Aneesh Kumar K.V]

On Wed, 06 Jul 2011 09:55:00 -0700, Jameson Graef Rollins <jrollins@finestructure.net> wrote:
Non-text part: multipart/mixed
Non-text part: multipart/signed
> Hey, folks.  Now that 0.6 is *finally* out the door, it's time to start
> working on 0.7.  As Interim Distinguished co-Chief Vice Tyrant I have
> been tasked with starting the discussion on 0.7 release priorities.
> 

We may also want to test with the trunk version of emacs. There are
multiple changes to html rendering and notmuch doesn't work with the
trunk version of emacs.

Currently emacs trunk is frozen for the next release. So we should have
a new emacs release soon.

-aneesh

Re: priorities for 0.7

[David Bremner]

On Wed, 06 Jul 2011 23:26:52 +0530, "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com> wrote:
> We may also want to test with the trunk version of emacs. There are
> multiple changes to html rendering and notmuch doesn't work with the
> trunk version of emacs.

As Aneesh knows [1], I hit those bugs too, so I think that is a great
idea, but I don't think it is likely that the (hypothetical) patches
will be ready in time for the 0.7 release, which is supposed to be at
the beginning of August.

David

[1]: because I bugged him about it previously.

Re: priorities for 0.7

[Daniel Schoepe]

[-- Attachment #1: Type: text/plain, Size: 953 bytes --]

On Wed, 06 Jul 2011 09:55:00 -0700, Jameson Graef Rollins <jrollins@finestructure.net> wrote:
> - emacs: notmuch-hello improvements (Daniel Schoepe, Michal Sojka,
>   Dmitry Kurochkin, Pieter Praet)
> 
>   These guys have been working on various improvements to the
>   notmuch-hello page.  I'll let them comment on where they're at with
>   that, and what should be merged.

My patch about user-defined sections mainly needs better documentation
and there's discussion reducing its flexibility a little, so it should
definitely be ready before the feature freeze, if that's still scheduled
for July 23.

Another thing I'd like to see in 0.7 is Pieter's patch about dividing
customize-options into different subgroups:
id:"1309890780-8214-1-git-send-email-pieter@praet.org"

Another candidate for inclusion would possibly be my commit that adds
tab-completion to notmuch-search:
id:"87lixevs2e.fsf@gilead.invalid"

Cheers,
Daniel

[-- Attachment #2: Type: application/pgp-signature, Size: 835 bytes --]

Re: priorities for 0.7

[David Bremner]

On Thu, 07 Jul 2011 20:34:21 +0200, Daniel Schoepe <daniel.schoepe@googlemail.com> wrote:

> My patch about user-defined sections mainly needs better documentation
> and there's discussion reducing its flexibility a little, so it should
> definitely be ready before the feature freeze, if that's still scheduled
> for July 23.

More or less. Of course, we need features merged into master before the
freeze, but that's partly why we're having this discussion.

d

Re: priorities for 0.7

[David Bremner]

[-- Attachment #1: Type: text/plain, Size: 471 bytes --]

On Wed, 06 Jul 2011 09:55:00 -0700, Jameson Graef Rollins <jrollins@finestructure.net> wrote:
> Hey, folks.  Now that 0.6 is *finally* out the door, it's time to start
> working on 0.7.  As Interim Distinguished co-Chief Vice Tyrant I have
> been tasked with starting the discussion on 0.7 release priorities.

By the way, it would be great if people either making or promoting patch
series that provide new features or notable bug fixes could provide a
patch to NEWS

d

[-- Attachment #2: Type: application/pgp-signature, Size: 315 bytes --]

Re: priorities for 0.7

[Daniel Schoepe]

[-- Attachment #1: Type: text/plain, Size: 494 bytes --]

On Thu, 07 Jul 2011 16:55:26 -0300, David Bremner <david@tethera.net> wrote:
> More or less. Of course, we need features merged into master before the
> freeze, but that's partly why we're having this discussion.

I think the latest version of the patch is now ready for consideration
for 0.7, thanks to many helpful suggestions and code by Michal Sojka.

id:"1310234598-15143-1-git-send-email-daniel.schoepe@googlemail.com"

(I'll write a patch updating the NEWS file in the next few days).



[-- Attachment #2: Type: application/pgp-signature, Size: 835 bytes --]

Re: priorities for 0.7

[Pieter Praet]

On Wed, 06 Jul 2011 09:55:00 -0700, Jameson Graef Rollins <jrollins@finestructure.net> wrote:
> Hey, folks.  Now that 0.6 is *finally* out the door, it's time to start
> working on 0.7.  As Interim Distinguished co-Chief Vice Tyrant I have
> been tasked with starting the discussion on 0.7 release priorities.
> 
> Below is just the beginning of list of things that have been discussed
> on the list that we should consider merging or working on for 0.7.  It's
> weighted towards stuff that I've worked on only because I'm more
> familiar with the state of those patches.  Please send follow-ups to
> this list to elaborate on certain things, or bring up other patches that
> should be considered [0].  We can then push on Carl to merge these ahead
> of the 0.7 freeze.
> 
> - notmuch-new atomicity:
> 
>   id:"1307822683-848-1-git-send-email-amdragon@mit.edu"
> 
>   I believe that the above is that latest version of this series; Austin
>   can point us to the right one if it's not.  This is an important
>   series that is long overdue to be merged.  It should definitely be a
>   top priority.

+1

> - message/rfc822 part support:
> 
>   id:"1307120466-4980-1-git-send-email-jrollins@finestructure.net"
>   id:"1307320169-29905-1-git-send-email-jrollins@finestructure.net"
> 
>   rfc822 part handling is a bit broken now after the notmuch-show
>   rework.  The above patch series should have been included in 0.6 to
>   fix a couple of regressions.  In the meantime Austin and I have
>   started looking at further rework of notmuch-show to try to clean it
>   up even more.  This rework might supersedes these patches, but it
>   depends on how far we get.
> 
> - emacs: Add notmuch-show-refresh-view function
> 
>   id:"1306627784-3401-1-git-send-email-jrollins@finestructure.net"
> 
> - emacs: Add callback functions to crypto sigstatus button.
> 
>   id:"1306861633-3873-1-git-send-email-jrollins@finestructure.net"
> 
> - emacs: notmuch-hello improvements (Daniel Schoepe, Michal Sojka,
>   Dmitry Kurochkin, Pieter Praet)
> 
>   These guys have been working on various improvements to the
>   notmuch-hello page.  I'll let them comment on where they're at with
>   that, and what should be merged.

I assume you're referring to Dmitry's search widget patch ?

The only role I played in it (apart from signing off on it) was
that of commit message theft provoker :D.

AFAIC, it's ripe for inclusion:
- id:"1309768024-14584-1-git-send-email-dmitry.kurochkin@gmail.com"

> - emacs: tagging race conditions (Austin Clements, Pieter Praet)
> 
>   Austin and Pieter have been looking at home to eliminate some race
>   conditions when tagging from the emacs interface.  I'll let them
>   comment on where that's at.

It's coming along nicely, mostly a case of dotting the i's,
though I'm holding off on sending my new patch series until my
bad attempt at a regexp is fixed:

- id:"1310307099-25197-1-git-send-email-pieter@praet.org"

Everyone's welcome to come point and laugh *but* the entrance
fee to the party is a *consistently* working regexp. ;)

Also, this patch series is a prerequisite:

- id:"1310313335-4159-1-git-send-email-pieter@praet.org"

> - emacs: default key bindings
> 
>   There's been a lot of discussion about default key bindings for the
>   emacs UI.  This issue frankly terrifies me since I worry that it is
>   basically impossible to converge on anything.  I think we'll need to
>   see a specific full proposal before we can really consider what to do
>   here.
> 
> - I know there are other old patches that are sinking in to the abyss.
>   Hopefully any of those that are still relevant can be excavated.

I'll be taking a look at the rest when I find some more time.

Meanwhile, I'd like to push these excellent patches for inclusion:

- [PATCH 4/5] Add notmuch-hook. Called when notmuch is started, before notmuch-hello buffer is created.
  - id:"1278000507-22290-5-git-send-email-dmitry.kurochkin@gmail.com"
- [PATCH 5/5] Add notmuch-hello-hook. Called every time notmuch-hello buffer is updated.
  - id:"1278000507-22290-6-git-send-email-dmitry.kurochkin@gmail.com"
- [PATCH v2] emacs: Turn id:"<message-id>" elements into buttons for notmuch searches
  - id:"1309961881-28333-2-git-send-email-daniel.schoepe@googlemail.com"

> jamie.
> 
> [0] It's maybe not mainline notmuch related, but we really need to get
> some sort of tag sharing system worked out.  I can only imagine what
> useful info is encoded in Carl's tags that would be invaluable for
> everyone else to see, especially the release managers.
> _______________________________________________
> notmuch mailing list
> notmuch@notmuchmail.org
> http://notmuchmail.org/mailman/listinfo/notmuch


Peace

-- 
Pieter

Re: priorities for 0.7

[Austin Clements]

On Wed, Jul 6, 2011 at 12:55 PM, Jameson Graef Rollins
<jrollins@finestructure.net> wrote:
> - notmuch-new atomicity:
>
>  id:"1307822683-848-1-git-send-email-amdragon@mit.edu"
>
>  I believe that the above is that latest version of this series; Austin
>  can point us to the right one if it's not.  This is an important
>  series that is long overdue to be merged.  It should definitely be a
>  top priority.

id:"CAH-f9WvAFuVp-6u6tFgbcPEGW304wkGy=uhSOGhbNXd0m7H3mw@mail.gmail.com"
points to a rebased and slightly conflict-resolved update of this.

>
> - message/rfc822 part support:
>
>  id:"1307120466-4980-1-git-send-email-jrollins@finestructure.net"
>  id:"1307320169-29905-1-git-send-email-jrollins@finestructure.net"
>
>  rfc822 part handling is a bit broken now after the notmuch-show
>  rework.  The above patch series should have been included in 0.6 to
>  fix a couple of regressions.  In the meantime Austin and I have
>  started looking at further rework of notmuch-show to try to clean it
>  up even more.  This rework might supersedes these patches, but it
>  depends on how far we get.

For purely practical reasons, I'm not sure how to best proceed with
the further rework.  It almost completely rewrites notmuch-show.c, so
I would hate to have to maintain the patch series out of tree for any
real length of time.