mu 1.7.18 & towards 1.8
51 views
Skip to first unread message
Dirk-Jan C. Binnema
May 8, 2022, 9:58:13โฏAM5/8/22
to mu-di...@googlegroups.com
Hi all,
I've pushed the changes for mu 1.7.18 to master (that is, the first big
batch on Fri plus some follow-ups); this finalizes a lot of the
code-cleanup that has started a while ago.
First, the code: one of the oldest sediments of mu code is formed by the
various mu-msg.* files. 1.7 replaces that with some brand new code under
lib/message; much of that landed earlier, and this is last part which
removes the last removes the last remaining mu-msg-* code.
Now, what does that mean in practice, apart from the general 'cleaner
code' bla bla?
- In mu4e, the server outputs message information s-expressions which
are then parsed by emacs; we were generating the expressions
on-the-fly, which takes a bit of time. As an optimization, we now
*cache* them as part of the message document in database. This speeds
up things quite a bit, mostly visible when there are many results,
i.e., as an extreme example:
- with 1.6:
,----
| time mu find "" --format=sexp > /dev/null
| mu find "" --format=sexp > /dev/null 37,41s user 6,47s system 99% cpu 44,143 total
`----
- and with 1.7.18:
,----
| time mu find "" --format=sexp > /dev/null
| mu find "" --format=sexp > /dev/null 0,33s user 0,33s system 99% cpu 0,664 total
`----
- a new command: 'mu fields', which shows details about the message
fields and flags. This is based on the same information that mu uses
itself, so this is always up to date and complete.
- a new message field "changed", which tracks the last time the message
was changed (the ctime).
mu find changed:3M..
shows message which were changed in the last 3 minutes
- we already had 'personal' contacts; now there's a new 'personal' flag
for messages; a message is considered 'personal' if at least one of
the contact fields contains a 'personal' address. And an address is
considered 'personal' if it matches the personal addresses passed to
'mu init (see 'mu info' for the values), i.e.
mu find hello flag:personal
- in mu4e not so much changes but, there are a few changes:
- by default, the "flags" column in the headers view now shows the
'list' and 'personal' flags as well.
- incompatible change: addresses are now represented as plists
(:name <name> :email <email>)
rather than cons-cells
(name . email)
so if you have code depending on the first, you need to update, ie.
(car contact) --> name
(cdr contact) --> address
now becomes:
(mu4e-contact-name contact) --> name
(mu4e-contact-email contact) --> address
sorry for the inconvenience; needed for some future plans.
- For more new things, please read NEWS.org, which should be fairly up to date.
Now, this *works for me*, but I expect a little trickle of changes on top
of this (some of which has already happened).
With this in place, I'm ...slowly... descending to the 1.8 release; no
complete feature freeze yet, but moving in that direction. Mostly,
bug-fixing, updating documentation, adding tests.
[ Note, one other item that I worked on was the 'mu4e-summary', as a
replacement for the headers buffer; I have no time now to finish it, so
that's something for post-1.8. ]
Kind regards,
Dirk.
--
Dirk-Jan C. Binnema Helsinki, Finland
e:dj...@djcbsoftware.nl w:www.djcbsoftware.nl
gpg: 6987 9CED 1745 9375 0F14 DA98 11DD FEA9 DCC4 A036
I've pushed the changes for mu 1.7.18 to master (that is, the first big
batch on Fri plus some follow-ups); this finalizes a lot of the
code-cleanup that has started a while ago.
First, the code: one of the oldest sediments of mu code is formed by the
various mu-msg.* files. 1.7 replaces that with some brand new code under
lib/message; much of that landed earlier, and this is last part which
removes the last removes the last remaining mu-msg-* code.
Now, what does that mean in practice, apart from the general 'cleaner
code' bla bla?
- In mu4e, the server outputs message information s-expressions which
are then parsed by emacs; we were generating the expressions
on-the-fly, which takes a bit of time. As an optimization, we now
*cache* them as part of the message document in database. This speeds
up things quite a bit, mostly visible when there are many results,
i.e., as an extreme example:
- with 1.6:
,----
| time mu find "" --format=sexp > /dev/null
| mu find "" --format=sexp > /dev/null 37,41s user 6,47s system 99% cpu 44,143 total
`----
- and with 1.7.18:
,----
| time mu find "" --format=sexp > /dev/null
| mu find "" --format=sexp > /dev/null 0,33s user 0,33s system 99% cpu 0,664 total
`----
- a new command: 'mu fields', which shows details about the message
fields and flags. This is based on the same information that mu uses
itself, so this is always up to date and complete.
- a new message field "changed", which tracks the last time the message
was changed (the ctime).
mu find changed:3M..
shows message which were changed in the last 3 minutes
- we already had 'personal' contacts; now there's a new 'personal' flag
for messages; a message is considered 'personal' if at least one of
the contact fields contains a 'personal' address. And an address is
considered 'personal' if it matches the personal addresses passed to
'mu init (see 'mu info' for the values), i.e.
mu find hello flag:personal
- in mu4e not so much changes but, there are a few changes:
- by default, the "flags" column in the headers view now shows the
'list' and 'personal' flags as well.
- incompatible change: addresses are now represented as plists
(:name <name> :email <email>)
rather than cons-cells
(name . email)
so if you have code depending on the first, you need to update, ie.
(car contact) --> name
(cdr contact) --> address
now becomes:
(mu4e-contact-name contact) --> name
(mu4e-contact-email contact) --> address
sorry for the inconvenience; needed for some future plans.
- For more new things, please read NEWS.org, which should be fairly up to date.
Now, this *works for me*, but I expect a little trickle of changes on top
of this (some of which has already happened).
With this in place, I'm ...slowly... descending to the 1.8 release; no
complete feature freeze yet, but moving in that direction. Mostly,
bug-fixing, updating documentation, adding tests.
[ Note, one other item that I worked on was the 'mu4e-summary', as a
replacement for the headers buffer; I have no time now to finish it, so
that's something for post-1.8. ]
Kind regards,
Dirk.
--
Dirk-Jan C. Binnema Helsinki, Finland
e:dj...@djcbsoftware.nl w:www.djcbsoftware.nl
gpg: 6987 9CED 1745 9375 0F14 DA98 11DD FEA9 DCC4 A036
Tassilo Horn
May 9, 2022, 7:26:17โฏAM5/9/22
to mu-di...@googlegroups.com, Dirk-Jan C. Binnema
Hi Dirk,
I've found out that `mu4e--contacts-hash' is missing many contacts I've
recently had conversations with. The result is that I can't complete
them after the keywords to/from/contacts at the query prompt.
In the log buffer, I can see entries such as:
--8<---------------cut here---------------start------------->8---
2022-05-09 13:08:25.540 -> (contacts :personal nil :after "2018-01-01" :tstamp "16771599551" :maxnum 2000)
2022-05-09 13:08:25.550 * Received 49 byte(s)
2022-05-09 13:08:25.550 <- (:contacts nil :tstamp 23370351411)
--8<---------------cut here---------------end--------------->8---
The :personal nil comes from the fact that
mu4e-compose-complete-only-personal is nil.
The mu query
mu cfind --after=`date +%s -d 2022-01-01`
finds those contacts.
Interestingly, the contacts are also missing in
mu cfind --after=`date +%s -d 2022-01-01` --personal
although I've had a conversation with those contacts within the last
week where To/Cc/From contained them and it hasn't been on a
mailinglist.
Bye,
Tassilo
I've found out that `mu4e--contacts-hash' is missing many contacts I've
recently had conversations with. The result is that I can't complete
them after the keywords to/from/contacts at the query prompt.
In the log buffer, I can see entries such as:
--8<---------------cut here---------------start------------->8---
2022-05-09 13:08:25.540 -> (contacts :personal nil :after "2018-01-01" :tstamp "16771599551" :maxnum 2000)
2022-05-09 13:08:25.550 * Received 49 byte(s)
2022-05-09 13:08:25.550 <- (:contacts nil :tstamp 23370351411)
--8<---------------cut here---------------end--------------->8---
The :personal nil comes from the fact that
mu4e-compose-complete-only-personal is nil.
The mu query
mu cfind --after=`date +%s -d 2022-01-01`
finds those contacts.
Interestingly, the contacts are also missing in
mu cfind --after=`date +%s -d 2022-01-01` --personal
although I've had a conversation with those contacts within the last
week where To/Cc/From contained them and it hasn't been on a
mailinglist.
Bye,
Tassilo
Dirk-Jan C. Binnema
May 9, 2022, 8:02:55โฏAM5/9/22
to mu-di...@googlegroups.com
On Monday May 09 2022, Tassilo Horn wrote:
> Hi Dirk,
>
> I've found out that `mu4e--contacts-hash' is missing many contacts I've
> recently had conversations with. The result is that I can't complete
> them after the keywords to/from/contacts at the query prompt.
>
> In the log buffer, I can see entries such as:
>
> 2022-05-09 13:08:25.540 -> (contacts :personal nil :after "2018-01-01" :tstamp "16771599551" :maxnum 2000)
> 2022-05-09 13:08:25.550 * Received 49 byte(s)
> 2022-05-09 13:08:25.550 <- (:contacts nil :tstamp 23370351411)
>
> 2022-05-09 13:08:25.550 * Received 49 byte(s)
> 2022-05-09 13:08:25.550 <- (:contacts nil :tstamp 23370351411)
>
> The :personal nil comes from the fact that
> mu4e-compose-complete-only-personal is nil.
>
> The mu query
>
> mu cfind --after=`date +%s -d 2022-01-01`
>
> finds those contacts.
>
> Interestingly, the contacts are also missing in
>
> mu cfind --after=`date +%s -d 2022-01-01` --personal
>
> although I've had a conversation with those contacts within the last
> week where To/Cc/From contained them and it hasn't been on a
> mailinglist.
Ack. I'll fix it shortly.
> mu4e-compose-complete-only-personal is nil.
>
> The mu query
>
> mu cfind --after=`date +%s -d 2022-01-01`
>
> finds those contacts.
>
> Interestingly, the contacts are also missing in
>
> mu cfind --after=`date +%s -d 2022-01-01` --personal
>
> although I've had a conversation with those contacts within the last
> week where To/Cc/From contained them and it hasn't been on a
> mailinglist.
Dirk-Jan C. Binnema
May 22, 2022, 1:22:24โฏAM5/22/22
to mu-di...@googlegroups.com
Hi all,
On Wednesday May 04 2022, Dirk-Jan C. Binnema wrote:
> Hi all,
>
> I've pushed the changes for mu 1.7.18 to master (that is, the first big
> batch on Fri plus some follow-ups); this finalizes a lot of the
> code-cleanup that has started a while ago.
>
<snip>
We're a little further with the mu 1.7.x releases.
- a number of issues reported against 1.7.18 have been fixed, in
particular with the contacts-cache / contacts completion in mu4e
- fixed an exception for certain messages that claimed to have
sub-parts, but hadn't
- a fixed for the mu4e keybindings (so the '-' binding works)
- fixes for a number of assumptions that are not true on bare-bones
system such as alpine/musl regarding timezones and locales. Handle
those, and update the tests accordingly
- fix for a race condition with indexing that broke our MacOS CI.
- mu4e's query autocompletion learned about the the 'personal' flag.
- better error reporting (i.e. 'no matches' for a mu find/mu-cfind query
is only as 'soft error')
- (re-) add support for forwarding attachments
- removed the 'toys'
- re-enabled guile
Oh, as visible new thing, "related" messages are now shown using
'mu4e-related-face`, by default shown with italics.
So now I'll spend some time with tests, docs, and possible bugs.
On Wednesday May 04 2022, Dirk-Jan C. Binnema wrote:
> Hi all,
>
> I've pushed the changes for mu 1.7.18 to master (that is, the first big
> batch on Fri plus some follow-ups); this finalizes a lot of the
> code-cleanup that has started a while ago.
>
We're a little further with the mu 1.7.x releases.
- a number of issues reported against 1.7.18 have been fixed, in
particular with the contacts-cache / contacts completion in mu4e
- fixed an exception for certain messages that claimed to have
sub-parts, but hadn't
- a fixed for the mu4e keybindings (so the '-' binding works)
- fixes for a number of assumptions that are not true on bare-bones
system such as alpine/musl regarding timezones and locales. Handle
those, and update the tests accordingly
- fix for a race condition with indexing that broke our MacOS CI.
- mu4e's query autocompletion learned about the the 'personal' flag.
- better error reporting (i.e. 'no matches' for a mu find/mu-cfind query
is only as 'soft error')
- (re-) add support for forwarding attachments
- removed the 'toys'
- re-enabled guile
Oh, as visible new thing, "related" messages are now shown using
'mu4e-related-face`, by default shown with italics.
So now I'll spend some time with tests, docs, and possible bugs.
Dirk-Jan C. Binnema
Jun 5, 2022, 4:14:32โฏAM6/5/22
to mu-di...@googlegroups.com
Hi all!
We've reached 1.7.26, and we're getting close to 1.8.
- Emacs bookmark support got restored, so you can use the emacs' builtin
bookmarks for messages once again (`M-x bookmark-set` etc.)
- Workaround for the mu server getting stuck on MacOS
- Fixed a few bug(let)s and warnings during compilation
- Work around for some shenanigans with compiling guile support on MacOS
- Work around for running on bare-bones systems (like Alpine/Musl)
without proper locale support
- Added more tests
There was one bigger issue (#2268); development versions 1.7.18 -
17.7.25 had a bug where the mail file names sometimes got misnamed (with
some extra ':2,').
This can be restored with something like (tested with GNU sed):
$ find ~/Maildir -name '*:2,*:*' | sed "s/\(\([^:]*\)\(:2,\)\{1,\}\(:2,.*$\)\)/mv '\0' '\2\4'/" > rename.sh
(replace 'Maildir' with the path to your maildir)
once this is done, do check the generated 'rename.sh' and after
convincing yourself it does the right thing, do
$ sh rename.sh
and after that, re-index.
Sorry for the inconvenience!
Kind regards,
Dirk.
We've reached 1.7.26, and we're getting close to 1.8.
- Emacs bookmark support got restored, so you can use the emacs' builtin
bookmarks for messages once again (`M-x bookmark-set` etc.)
- Workaround for the mu server getting stuck on MacOS
- Fixed a few bug(let)s and warnings during compilation
- Work around for some shenanigans with compiling guile support on MacOS
- Work around for running on bare-bones systems (like Alpine/Musl)
without proper locale support
- Added more tests
There was one bigger issue (#2268); development versions 1.7.18 -
17.7.25 had a bug where the mail file names sometimes got misnamed (with
some extra ':2,').
This can be restored with something like (tested with GNU sed):
$ find ~/Maildir -name '*:2,*:*' | sed "s/\(\([^:]*\)\(:2,\)\{1,\}\(:2,.*$\)\)/mv '\0' '\2\4'/" > rename.sh
(replace 'Maildir' with the path to your maildir)
once this is done, do check the generated 'rename.sh' and after
convincing yourself it does the right thing, do
$ sh rename.sh
and after that, re-index.
Sorry for the inconvenience!
Kind regards,
Dirk.
Dirk-Jan C. Binnema
Jun 18, 2022, 8:44:07โฏAM6/18/22
to mu-di...@googlegroups.com
Hi All,
I just pushed 1.7.28; a bunch of changes since 1.7.26 and even some tiny
new features:
- there's a new flag 'calendar' for message with calendar invitations;
and there's a marker for that in the headers-view as well.
So if you want to be really fancy and have the needed fonts you can use
e.g.
--8<---------------cut here---------------start------------->8---
(setq
mu4e-use-fancy-chars t
mu4e-headers-draft-mark '("D" . "๐")
mu4e-headers-flagged-mark '("F" . "๐")
mu4e-headers-new-mark '("N" . "๐ฅ")
mu4e-headers-passed-mark '("P" . "โฏ")
mu4e-headers-replied-mark '("R" . "โฎ")
mu4e-headers-seen-mark '("S" . "โ")
mu4e-headers-trashed-mark '("T" . "๐")
mu4e-headers-attach-mark '("a" . "๐")
mu4e-headers-encrypted-mark '("x" . "๐")
mu4e-headers-signed-mark '("s" . "๐")
mu4e-headers-unread-mark '("u" . "โ")
mu4e-headers-list-mark '("s" . "๐")
mu4e-headers-personal-mark '("p" . "๐จ")
mu4e-headers-calendar-mark '("c" . "๐ "))
--8<---------------cut here---------------end--------------->8---
anyway, there's not too much left to do for 1.8 it seems, as in "small
things to fix"; so that'll be something for post-1.8.
So if there is some catastrophic bug still, now is a good time to speak
up, thanks!
I just pushed 1.7.28; a bunch of changes since 1.7.26 and even some tiny
new features:
- there's a new flag 'calendar' for message with calendar invitations;
and there's a marker for that in the headers-view as well.
So if you want to be really fancy and have the needed fonts you can use
e.g.
--8<---------------cut here---------------start------------->8---
(setq
mu4e-use-fancy-chars t
mu4e-headers-draft-mark '("D" . "๐")
mu4e-headers-flagged-mark '("F" . "๐")
mu4e-headers-new-mark '("N" . "๐ฅ")
mu4e-headers-passed-mark '("P" . "โฏ")
mu4e-headers-replied-mark '("R" . "โฎ")
mu4e-headers-seen-mark '("S" . "โ")
mu4e-headers-trashed-mark '("T" . "๐")
mu4e-headers-attach-mark '("a" . "๐")
mu4e-headers-encrypted-mark '("x" . "๐")
mu4e-headers-signed-mark '("s" . "๐")
mu4e-headers-unread-mark '("u" . "โ")
mu4e-headers-list-mark '("s" . "๐")
mu4e-headers-personal-mark '("p" . "๐จ")
mu4e-headers-calendar-mark '("c" . "๐ "))
--8<---------------cut here---------------end--------------->8---
anyway, there's not too much left to do for 1.8 it seems, as in "small
things to fix"; so that'll be something for post-1.8.
So if there is some catastrophic bug still, now is a good time to speak
up, thanks!
Tim Cross
Jun 18, 2022, 7:32:17โฏPM6/18/22
to mu-di...@googlegroups.com
"Dirk-Jan C. Binnema" <dj...@djcbsoftware.nl> writes:
> Hi All,
>
> I just pushed 1.7.28; a bunch of changes since 1.7.26 and even some tiny
> new features:
>
> - there's a new flag 'calendar' for message with calendar invitations;
> and there's a marker for that in the headers-view as well.
>
> So if you want to be really fancy and have the needed fonts you can use
> e.g.
>
> (setq
> mu4e-use-fancy-chars t
> mu4e-headers-draft-mark '("D" . "๐")
> mu4e-headers-flagged-mark '("F" . "๐")
> mu4e-headers-new-mark '("N" . "๐ฅ")
> mu4e-headers-passed-mark '("P" . "โฏ")
> mu4e-headers-replied-mark '("R" . "โฎ")
> mu4e-headers-seen-mark '("S" . "โ")
> mu4e-headers-trashed-mark '("T" . "๐")
> mu4e-headers-attach-mark '("a" . "๐")
> mu4e-headers-encrypted-mark '("x" . "๐")
> mu4e-headers-signed-mark '("s" . "๐")
> mu4e-headers-unread-mark '("u" . "โ")
> mu4e-headers-list-mark '("s" . "๐")
> mu4e-headers-personal-mark '("p" . "๐จ")
> mu4e-headers-calendar-mark '("c" . "๐ "))
>
> mu4e-use-fancy-chars t
> mu4e-headers-draft-mark '("D" . "๐")
> mu4e-headers-flagged-mark '("F" . "๐")
> mu4e-headers-new-mark '("N" . "๐ฅ")
> mu4e-headers-passed-mark '("P" . "โฏ")
> mu4e-headers-replied-mark '("R" . "โฎ")
> mu4e-headers-seen-mark '("S" . "โ")
> mu4e-headers-trashed-mark '("T" . "๐")
> mu4e-headers-attach-mark '("a" . "๐")
> mu4e-headers-encrypted-mark '("x" . "๐")
> mu4e-headers-signed-mark '("s" . "๐")
> mu4e-headers-unread-mark '("u" . "โ")
> mu4e-headers-list-mark '("s" . "๐")
> mu4e-headers-personal-mark '("p" . "๐จ")
> mu4e-headers-calendar-mark '("c" . "๐ "))
>
> anyway, there's not too much left to do for 1.8 it seems, as in "small
> things to fix"; so that'll be something for post-1.8.
>
> So if there is some catastrophic bug still, now is a good time to speak
> up, thanks!
>
Thanks Dirk. RUnning it now and it looks good. Appreciate all your hard
> things to fix"; so that'll be something for post-1.8.
>
> So if there is some catastrophic bug still, now is a good time to speak
> up, thanks!
>
work with this.
Eduardo Mercovich
Jun 25, 2022, 11:02:15โฏAM6/25/22
to mu-di...@googlegroups.com, Dirk-Jan C. Binnema
Hi Dirk.
> I just pushed 1.7.28 [...]
As always, just a big thank you for sharing all your great work. :)
Warm regards...
--
Eduardo Mercovich
Donde se cruzan tus talentos
con las necesidades del mundo,
ahรญ estรก tu vocaciรณn.
(Anรณnimo)
> I just pushed 1.7.28 [...]
As always, just a big thank you for sharing all your great work. :)
Warm regards...
--
Eduardo Mercovich
Donde se cruzan tus talentos
con las necesidades del mundo,
ahรญ estรก tu vocaciรณn.
(Anรณnimo)
Reply all
Reply to author
Forward
0 new messages