Eric Wong [Sun, 25 Jul 2021 12:44:23 +0000 (12:44 +0000)]
extindex: support --jobs/-j properly on creation for shard count
This wasn't wired up properly, but Xapian appears to suffer from
I/O amplification problems as DB shards get larger:
https://lists.xapian.org/pipermail/xapian-discuss/2019-February/009727.html
<23640.32170.703368.841021@y.dockes.com>
Of course, we shouldn't have too many shards, either; because
performance problems with too many shards was the entire reason
extindex was created:
https://lists.xapian.org/pipermail/xapian-discuss/2020-August/009823.html
<
20200826064728.GA32239@dcvr>
Eric Wong [Sun, 25 Jul 2021 12:03:33 +0000 (12:03 +0000)]
doc: lei-{p2q,rediff}: note implicit --stdin
lei actually uses implicit --stdin everywhere, but I thing
these patch-related commands are the most common use of them.
Eric Wong [Sun, 25 Jul 2021 11:15:06 +0000 (11:15 +0000)]
t/lei-watch.t: improve test reliability
On single CPU (and overloaded SMP) systems, we can't rely on
inotify in lei-daemon firing before a "lei note-event done"
client hits it. So force in a single tick() to ensure the
scheduler can yield to lei-daemon and see the inotify wakeup
before "lei note-event done" to commit the write.
Eric Wong [Sun, 25 Jul 2021 10:40:17 +0000 (10:40 +0000)]
init: support git <2.30 for "-c KEY=VALUE" args
It turns out `--fixed-value' is a relatively new git-config(1)
feature in git 2.30+ (December 2020). So use the quotemeta
perlop for now since it seems compatible-enough for POSIX ERE
used by git.
Eric Wong [Sun, 25 Jul 2021 00:43:32 +0000 (00:43 +0000)]
lei_mail_sync: locations_for API uses oidbin for comparisons
Favor oidbin use internally to reduce internal memory traffic.
Eric Wong [Sun, 25 Jul 2021 00:43:31 +0000 (00:43 +0000)]
lei_inspect: fix typo
Not sure how this wasn't caught, earlier...
Eric Wong [Sun, 25 Jul 2021 00:43:30 +0000 (00:43 +0000)]
lei_search: favor binary OID comparisons
Reduce memory traffic and code, too.
Eric Wong [Sun, 25 Jul 2021 00:43:29 +0000 (00:43 +0000)]
extsearchidx: favor binary comparison in common case
We'll use 20-byte SHA-1 comparisons instead of 40-byte
hex representations for a minor reduction in memory
traffic.
Eric Wong [Sun, 25 Jul 2021 00:11:03 +0000 (00:11 +0000)]
extsearchidx: use more appropriate max for dedupe
The over.msgid table may contain ghost Message-IDs and also
Message-IDs of deleted spam messages, so over->max isn't a
good aproproximation of dedupe progress.
Eric Wong [Sun, 25 Jul 2021 00:11:02 +0000 (00:11 +0000)]
extindex: improve comment around git->async_wait_all
I found myself tempted to remove this, but it appears impossible
due to odd messages which have multiple Message-IDs.
Eric Wong [Sun, 25 Jul 2021 00:11:01 +0000 (00:11 +0000)]
extindex: support --dedupe[=MSGID]
Sometimes I just want to dedupe a single Message-ID to test
something, and this lets me do it.
This patch appears to do what its supposed to. But it also
appears to be finding duplicates that were previously missed.
That's a good thing, but I wish I understood what seems to be
fixed :x
I'm not sure why the previous ExtSearchIdx.pm (blob
357312b8)
was causing messages to be missed, even, and why this patch
seems to fix it... And it's not infinite looping, either.
Anyways, before this patch, "-extindex --dedupe" was taking ~5
min to no-op every message (after the initial full --dedupe run
which took over a day to run). No-op --dedupes now take just
under 2 hours to scan every single cross-posted message for a
no-op dedupe. The initial dedupe took nearly 44 hours on my
system for <https://yhbt.net/lore/all/> due to SATA-2 TLC SSD
latency on 3 gigantic Xapian shards.
Running --dedupe with this change seems to prevent
/BUG\?.*?not deduplicated properly/ stderr messages from being
triggered by View.pm. Current versions of -extindex do not
seem susceptible to introducing duplicates.
Eric Wong [Fri, 23 Jul 2021 10:56:12 +0000 (10:56 +0000)]
lei rm-watch: new command to support removing watches
Pretty trivial since it just invokes "git-config". It's mainly
intended to make shell completion easier.
Eric Wong [Fri, 23 Jul 2021 10:56:11 +0000 (10:56 +0000)]
lei: avoid SQLite COUNT() for dedupe
SQLite COUNT() is a slow operation that does a full table scan
with no conditions. There's no need for it, since lei dedupe
only needs to know if it's empty or not to decide between
new/ and cur/ for Maildir outputs.
Eric Wong [Fri, 23 Jul 2021 10:56:10 +0000 (10:56 +0000)]
t/lei*: check error messages on failures
I just hit an unreproducible failure in t/lei-p2q.t and
lacked $lei_err information to diagnose it. Hopefully
this helps track down odd failures in the future.
Eric Wong [Thu, 22 Jul 2021 08:37:36 +0000 (08:37 +0000)]
t/solver_git: use like() to improve error reporting
I hit a test failure here, but haven't been able to reproduce
it...
Eric Wong [Wed, 21 Jul 2021 14:07:06 +0000 (14:07 +0000)]
lei: auto-refresh watches in config, cancel missing
This makes behavior less surprising on restarts as we no longer
lose state on restarts, so there's no need to manually run "lei
add-watch" to re-enable watches. This also allows us to
transparently handle changes if somebody edits the lei config
file directly or via git-config(1).
Eric Wong [Mon, 19 Jul 2021 08:59:35 +0000 (08:59 +0000)]
lei: start implementing inotify Maildir support
This allows lei to automatically note keyword (message flag)
changes made to a Maildir and propagate it into lei/store:
lei add-watch --state=tag-ro /path/to/Maildir
This doesn't persist across restarts, yet. In the future,
it will be applied automatically to "lei q" output Maildirs
by default (with an option to disable it).
State values of tag-rw, index-<ro|rw>, import-<ro|rw> will all
be supported for Maildir.
This represents a fairly major internal change that's fairly
intrusive, but the whole daemon-oriented design was to
facilitate being able to automatically monitor (and propagate)
Maildir/IMAP flag changes.
Eric Wong [Wed, 21 Jul 2021 14:05:50 +0000 (14:05 +0000)]
init: allow arbitrary key-values via -c KEY=VALUE
This won't blindly append identical key=values, but
allows specifying multiple, different key=value pairs
as long as the values are different.
Eric Wong [Wed, 21 Jul 2021 14:05:49 +0000 (14:05 +0000)]
extsearch: support publicinbox.*.boost parameter
This behaves identically the lei external "boost" parameter in
prioritizing raw messages for extindex.
Relying exclusively on the config file order doesn't work well
for mirrors since it's impossible to guarantee config file
ordering via grokmirror hooks.
Config file ordering remains the default if boost is
unconfigured, or in case of ties.
Note: I chose the name "boost" rather than "priority" or "rank"
since I always get confused by whether higher or lower numbers
take precedence when it comes to kernel scheduling. "weight" is
also a part of Xapian API terminology, which we currently do not
expose to configuration (but may in the future).
Eric Wong [Tue, 20 Jul 2021 08:58:58 +0000 (08:58 +0000)]
httpd: fix SIGHUP by invalidating cache on reload
Since we require separate PublicInbox::HTTPD instances for each
listen socket address (in order to support {SERVER_<NAME|PORT>}
for PSGI env), the old cache needed to be invalidated on rare
app refreshes.
SIGHUP has always been broken in -httpd (but not -imapd or
-nntpd) due to this cache.
Update the daemon documentation and 5.10.1-ize some bits while
we're in the area.
Eric Wong [Tue, 13 Jul 2021 11:06:13 +0000 (11:06 +0000)]
config: s/_one_val/get_1/ for public use
We'll be using this in lei for watch configs.
Eric Wong [Thu, 8 Jul 2021 08:25:19 +0000 (08:25 +0000)]
extindex: dedupe: reduce SQLite contention and dirty data
Complex queries causes SQLite to block readers for longer than
their retry period. For dedupe, it was also preventing us from
making good use of checkpoints due to the query time.
With many deduplications, checkpoints are necessary to maintain
system health due to having too much data piled up.
Eric Wong [Wed, 7 Jul 2021 23:24:55 +0000 (23:24 +0000)]
extsearchidx: ignore Eml warnings across the board
There's nothing we can do about misformatted emails and headers
we get from untrusted sources. They're too noisy and those
messages already exist in public-inboxes, anyways, so just
keep things quiet so we can spot real problems more easily.
Eric Wong [Tue, 6 Jul 2021 12:42:03 +0000 (12:42 +0000)]
extindex: --gc: avoid SQLite lock conflict on shard cleanup
Xapian shard cleanup only requires read-only access to
over.sqlite3, so avoid opening it with read-write access since
create_tables will hit lock conflicts on "INSERT OR IGNORE"
statements.
Eric Wong [Tue, 6 Jul 2021 12:42:02 +0000 (12:42 +0000)]
extindex: implement --dedupe to fix old extindices
This is intended to fix older indices that had deduplication
bugs for matching content. It'll also make dealing with
future changes to ContentHash easier since that's never
guaranteed stable.
It also supports --dry-run to print changes only without
making them.
Eric Wong [Tue, 6 Jul 2021 12:42:01 +0000 (12:42 +0000)]
eml: relax warn_ignore regexps for current Email::Address::XS
These seem needed with the data I'm currently working on, but I
haven't changed my version of Email::Address::XS since my last
Debian stable upgrade (to buster).
Eric Wong [Sun, 4 Jul 2021 19:35:54 +0000 (19:35 +0000)]
lei: drop workers on EOF from clients
Sometimes a user will be bored waiting for a command to finish,
so ensure we drop disconnect workers in this case.
Eric Wong [Fri, 2 Jul 2021 21:02:23 +0000 (21:02 +0000)]
lei import: increase flags search batch size, display progress
IMAP flag-only synchronization doesn't fetch entire messages,
so we can safely bump the batch size iff a user specified one
for full messages to 10000 times that.
Since I sometimes wonder why nothing happens for several seconds
after starting "lei import $URL", we'll also show some progress
during the flag synchronization phase.
Eric Wong [Fri, 2 Jul 2021 20:42:24 +0000 (20:42 +0000)]
lei inspect: help+completion for --dir option
It's the most generic name I could find for it since it can
mean so many things...
Eric Wong [Fri, 2 Jul 2021 20:42:09 +0000 (20:42 +0000)]
extsearchidx: extra assertions for deduplication flow
I haven't found any bugs from this (still looking for missed
deduplication bugs), and it's a bit shorter and more likely to
catch future bugs. Clean up an unnecessary ->{mid} array copy
while we're at it, too.
Eric Wong [Thu, 1 Jul 2021 11:31:33 +0000 (11:31 +0000)]
lei inspect: support "mid:" (and "m:") prefix
Using this to track down deduplication failures in -extindex...
Eric Wong [Thu, 1 Jul 2021 11:31:32 +0000 (11:31 +0000)]
lei inspect: support automatic pager in output
All commands which output non-trivial amounts of data to
the terminal should support this.
Eric Wong [Wed, 30 Jun 2021 22:00:20 +0000 (22:00 +0000)]
extsearchidx: lock before writing multi-pack-index
This avoids errors from git in case -extindex gets invoked in
parallel.
Eric Wong [Wed, 30 Jun 2021 19:14:48 +0000 (19:14 +0000)]
extsearchidx: symlink .rev and .bitmap files into ALL.git
It's possible for these to exist and git can (or may eventually)
take advantage of them to speed up functionality which affects
us.
Eric Wong [Wed, 30 Jun 2021 17:58:54 +0000 (17:58 +0000)]
searchidx: default BATCH_BYTES to 8MB on 64-bit systems
This default seems closer to reasonable on 64-bit systems which
are the norm these days. 32-bit systems gain 48K so it's an
even 1 MB, but we need to keep 32-bit systems from using too
much since there's still some ancient systems out there with
small inboxes.
Eric Wong [Mon, 28 Jun 2021 21:59:06 +0000 (21:59 +0000)]
www: fix manifest.js.gz for default publicInbox.grokManifest
ManifestJsGz->response was not invoking the new "url_filter"
method properly. Furthermore, fix url_filter for returning 404
responses.
Reported-by: Kyle Meyer <kyle@kyleam.com>
Link: https://public-inbox.org/meta/87fsx3128a.fsf@kyleam.com/
Fixes: 520be116e8a686cb ("www_listing: start updating for pagination + search")
Eric Wong [Fri, 25 Jun 2021 01:06:39 +0000 (01:06 +0000)]
extindex: maintain pack symlinks and use "git multi-pack-index"
This is a fair amount of complexity, but it speeds up
"git cat-file --batch" startup by 3-4% with 50K packfiles
with a hot kernel cache.
This appears extremely sensitive to RAM available to
the kernel page cache with my SATA 2 SSD. Faster storage
and more RAM can bring loading pack.
2.60s vs 2.69s were the best cases on my workstation with and
without the multi-pack-index, however times could be all over
the place (even in the minutes) with more activity on my
workstation.
Getting sub-minute times requires a git patch to speed up
alt_odb_usable():
<https://lore.kernel.org/
20210624005806.12079-1-e@80x24.org/>
Otherwise, prepare to wait several minutes.
Eric Wong [Thu, 24 Jun 2021 12:59:09 +0000 (12:59 +0000)]
www_listing: fix manifest.js.gz generation with extindex "all"
WwwListing and ManifestJsGz may be too different nowadays to
be worth the code sharing between them.
Update some comments and note we still needs better tests :x
Fixes: 520be116e8a686cb ("www_listing: start updating for pagination + search")
Eric Wong [Thu, 24 Jun 2021 08:41:56 +0000 (08:41 +0000)]
v2writable: avoid spawning "git hash-object"
We have git_sha() nowadays that's used everywhere, so avoid
process spawning overhead for "git hash-object".
Eric Wong [Thu, 24 Jun 2021 08:32:05 +0000 (08:32 +0000)]
doc: tuning: add a note about Linux sys.vm.max_map_count
git tends to die when mmap(2) fails on this limit, so let
users know about it. Perhaps git could gracefully fallback.
Eric Wong [Thu, 24 Jun 2021 05:50:04 +0000 (05:50 +0000)]
favor git(1) rather than libgit2 for ExtSearch
While both git and libgit2 take around 16 minutes to load 100K
alternates there's already a proposed patch to make git faster:
<https://lore.kernel.org/git/
20210624005806.12079-1-e@80x24.org/>
It's also easier to patch and install git locally since the
git.git build system defaults to prefix=$HOME and dealing with
dynamic linking with libgit2 is more difficult for end users
relying on Inline::C.
libgit2 remains in use for the non-ALL.git case, but maybe it's
not necessary (libgit2 is significantly slower than git in
Debian 10 due to SHA-1 collision checking).
Eric Wong [Wed, 23 Jun 2021 11:14:22 +0000 (07:14 -0400)]
www: do not warn on blank query parameters
Sometimes users (or bots) may lead queries with '&' and
trigger uninitialized variable warnings, just ignore them
and give consumers a $ctx->{qp}->{''} entry.
While we're in the area, pass a regexp rather than scalar string
to the `split' perlop to prevent Perl from recompiling the
regexp on every call.
Eric Wong [Wed, 23 Jun 2021 11:14:21 +0000 (07:14 -0400)]
www_listing: start updating for pagination + search
When dealing with thousands of inboxes, displaying all of
them on a single page isn't going to work. So steal some
pagination and search results code from the message search
to generate some basic HTML output that looks good in w3m.
Eric Wong [Wed, 23 Jun 2021 11:14:20 +0000 (07:14 -0400)]
search: make xap_terms easier-to-use and use it more
This allows us to simplify callers throughout, and exceptions are
can no longer be silently hidden. MiscSearch now uses xap_terms
for looking up eidx_key terms for a code reduction.
We also simplify LeiStore->_msg_kw for runtime use by moving the
MsetIterator handling into t/lei_store.t test case.
Alyssa Ross [Tue, 22 Jun 2021 12:09:20 +0000 (12:09 +0000)]
t/hl_mod: accept "make" or "makefile" for Makefile
Version 4.0 of highlight has renamed the "make" language to
"makefile". So just check the string starts with "make", to handle
both 3.x and 4.x.
I tested that public-inbox does actually work with highlight 4 -- it
can highlight my Makefile fine. :)
Eric Wong [Tue, 22 Jun 2021 10:04:36 +0000 (10:04 +0000)]
lei: use open() perlop for -C (chdir)
This is for consistency with the open() at initial accept, in
case we hit a code path which expects Perl directory handles
rather than "file handles". Both work with the chdir() perlop
(fchdir(2), in our case).
Eric Wong [Sun, 20 Jun 2021 08:39:47 +0000 (08:39 +0000)]
lei import: help + completion for --new-only
I've found it's very helpful for large IMAP folders.
Eric Wong [Sun, 20 Jun 2021 04:33:19 +0000 (04:33 +0000)]
lei sucks: don't warn or error out on missing dependencies
%INC can hold undef. This can be hit on a Linux machine missing
Linux::Inotify2. Loading PublicInbox::KQNotify is attempted and
PublicInbox/KQNotify.pm always exists, causing the `undef' entry
in %INC when it fails to load IO::KQueue.
$ perl -MData::Dumper -I lib \
-E 'eval { require PublicInbox::KQNotify }; say Dumper(\%INC)'
Eric Wong [Sat, 19 Jun 2021 03:22:28 +0000 (03:22 +0000)]
view: extra check to for redundant messages in HTML view
There appears to be some cases of duplicates appearing due to
-extindex. I haven't nailed down the cause of it, yet, but
this should make things easier for readers using the PSGI
HTML interface in the meantime.
The raw mboxrd remains undeduplicated for now, and the
correct fix/workaround would be some fsck-like mode for
public-inbox-extindex.
Eric Wong [Fri, 18 Jun 2021 21:44:38 +0000 (18:44 -0300)]
scripts: add syscall-list tool for development
We'll be supporting inotify directly as we do with epoll so so
Linux users won't have to deal with XS, extra DSOs or install
Linux::Inotify2 (and common::sense) modules.
Eric Wong [Fri, 18 Jun 2021 19:28:13 +0000 (10:28 -0900)]
t/sigfd: add diagnostic for occasional FreeBSD failure
Not 100% sure what's going on, here...
Eric Wong [Fri, 18 Jun 2021 19:20:31 +0000 (19:20 +0000)]
lei/store: do not put NULL into over.num column
Simplify oid2docid and filter out undefined docids in ->add_eml,
instead. This avoids SQLite "datatype mismatch" errors in
OverIdx->add_over
Fixes: d1052f03ea85d4af ("lei/store: cull redundant docids based on blob OID")
Eric Wong [Thu, 17 Jun 2021 22:00:47 +0000 (22:00 +0000)]
lei/store: cull redundant docids based on blob OID
I'm not sure how this happened (only once for me in March), but
it should not happen... In any case, we'll operate on the
lowest numbered docid and cull redundant index entries when
lei/store is open for read-write.
This also fixes the normal lei/store removal path to clean up
the xref3 table (since it's not done automatically for
public-facing -eidx due to the multi-list nature of it).
Eric Wong [Thu, 17 Jun 2021 22:00:46 +0000 (22:00 +0000)]
lei_input: prefix bare Maildir paths w/ "maildir:"
This will simplify upcoming code for watches.
Eric Wong [Thu, 17 Jun 2021 22:00:45 +0000 (22:00 +0000)]
lei inspect: learn "num:" and "docid:" prefixes
"num:" is useful for inspecting Inbox-ish directories, while
"docid:" can be used for any Xapian DB (not just stuff managed
by our code).
Eric Wong [Sun, 13 Jun 2021 18:12:06 +0000 (18:12 +0000)]
lei index+import: reject keywords from R/O IMAP
Since users can't set IMAP flags in read-only IMAP folders,
we won't clobber local flags when importing from IMAP. This
also enables the local_blob fallback used for lei-index to
be used for index deduplication.
Eric Wong [Sun, 13 Jun 2021 18:12:05 +0000 (18:12 +0000)]
lei_input: allow keywords when importing 1 file from Maildir
This will eventually be useful for supporting inotify watches
on Maildir. It will also allow users to script their own FS
watchers more easily.
Eric Wong [Sat, 12 Jun 2021 00:10:45 +0000 (00:10 +0000)]
net_reader: canonicalize URL args on add_url
This fixes cases when users specify an IMAP or NNTP URL
with standard port numbers explicitly.
In other words, this allows users to use
"lei ls-mail-source nntps://public-inbox.org:563/" and
"lei ls-mail-source imaps://public-inbox.org:993/"
without hitting "BUG:" errors.
Eric Wong [Sat, 12 Jun 2021 00:10:44 +0000 (00:10 +0000)]
lei import: use url_folder_cache for completion
And fix "lei index" completion while we're at it.
Eric Wong [Sat, 12 Jun 2021 00:10:43 +0000 (00:10 +0000)]
t/lei-import-http: quiet unnecessary diag message
Leftover while writing the test.
Eric Wong [Sat, 12 Jun 2021 00:10:42 +0000 (00:10 +0000)]
lei ls-mail-source: write through to URL folder cache
We'll be able to use this for shell completion for
lei import, lcat, tag, etc..
This also adds --url support for scripting purposes.
Eric Wong [Sat, 12 Jun 2021 00:10:41 +0000 (00:10 +0000)]
lei: stop pager early on exit
This is necessary when using "ls-mail-source" on an unreachable
IMAP server.
Eric Wong [Fri, 11 Jun 2021 09:42:40 +0000 (09:42 +0000)]
lei ls-mail-source: list IMAP folders and NNTP groups
While other tools can provide the same functionality, having
integration with git-credential is convenient, here. Caching
and completion will be implemented separately.
Eric Wong [Wed, 9 Jun 2021 23:27:50 +0000 (20:27 -0300)]
lei tag: less confusing warning about unimported messages
"unimported" is more meaningful than "missing", here. And
instead of having every worker spew about unimported messages,
we'll accumulate and only print one warning line. This
necessitated alterating ->DESTROY behavior and persisting
the client socket within the $lei object itself, not just
the PktOp consumer object.
Eric Wong [Wed, 9 Jun 2021 22:39:24 +0000 (22:39 +0000)]
lei import: support --new-only for IMAP
Taking ~40s to synchronize a ~75K message IMAP folder is
still a lot of time, so support an option to only touch
new messages.
This is similar to "offlineimap -q" (quick) or "mbsync --new"
switches, but lei already accepts "-q" as a shortcut for
--quiet. "--new" could work, but "--new-only" might be more
descriptive (or "--only-new"?), since the default fetches
also fetches new messages.
v2: warn for non-IMAP sources, I'm not sure it's worth it for
Maildir or other sources, yet. It will also make sense
for MH and JMAP once we support them.
Eric Wong [Wed, 9 Jun 2021 07:47:51 +0000 (07:47 +0000)]
lei prune-mail-sync: new command to prune invalid sync data
This will be invoked automatically by "lei import" eventually,
but it may make sense to expose as a separate command.
Eric Wong [Wed, 9 Jun 2021 07:47:50 +0000 (07:47 +0000)]
lei_mail_sync: hoist out --all handling from export-kw
We'll be reusing it in other commands, too.
Eric Wong [Wed, 9 Jun 2021 07:47:49 +0000 (07:47 +0000)]
lei tag: parallelize Maildir access
Since Maildir isn't guaranteed to have any sort of order, we
can parallelize inputs, here. On a 4-core system, this reduced
one of my tag invocations from 5.5 to 1.4s.
Eric Wong [Wed, 9 Jun 2021 07:47:48 +0000 (07:47 +0000)]
mdir_reader: maildir_each_file: pass flags, skip Trash
This is a slight behavior change for "lei q": Trashed
(but not-yet-expunged) messages no longer get unlinked
when --output is used without --augment.
Eric Wong [Wed, 9 Jun 2021 07:47:47 +0000 (07:47 +0000)]
inbox_writable: fix import_maildir
I'm not sure if anybody uses this, but it exists. It'll likely
be dropped in the future.
Fixes: fa3f0cbcd1af5008 ("use MdirReader in -watch and InboxWritable")
Eric Wong [Wed, 9 Jun 2021 10:03:05 +0000 (10:03 +0000)]
lei/store: do eidx_init before creating R/W lms dbh
Sharing lms->{dbh} with eidx shards appears to be the cause of
the "Issuing rollback() due to DESTROY without explicit
disconnect() of DBD::SQLite::db handle" messages I've been
seeing from "lei up".
Eric Wong [Wed, 9 Jun 2021 00:11:11 +0000 (00:11 +0000)]
lei edit-search: fix and add a (weak) test
This broke recently and lacked an automated test, so rely on
EDITOR=cat to ensure we have some coverage.
Fixes: d2670108f71b1eff ("pkt_op: make pkt_do an OO method")
Eric Wong [Tue, 8 Jun 2021 23:56:13 +0000 (23:56 +0000)]
lei pmdir: fix nproc for <= 4 CPUs
I forgot my FreeBSD VM has 8 cores, actually, and tweaked the
nproc detection on that machine before finalizing commit
10b523eb017162240b1ac3647f8dcbbf2be348a7
("lei import: speed up repeated Maildir imports")
Fixes: 10b523eb01716224 ("lei import: speed up repeated Maildir imports")
Eric Wong [Tue, 8 Jun 2021 09:50:21 +0000 (09:50 +0000)]
lei import: speed up repeated Maildir imports
On a 4-core CPU, this speeds up "lei import" on a largish
Maildir inbox with 75K messages from ~8 minutes down to ~40s.
Parallelizing alone did not bring any improvement and may
even hurt performance slightly, depending on CPU availability.
However, creating the index on the "fid" and "name" columns in
blob2name yields us the same speedup we got.
Parallelizing IMAP makes more sense due to the fact most IMAP
stores are non-local and subject to network latency.
Followup-to: bdecd7ed8e0dcf0b45491b947cd737ba8cfe38a3 ("lei import: speed up kw updates for old IMAP messages")
Eric Wong [Tue, 8 Jun 2021 09:50:20 +0000 (09:50 +0000)]
lei: generalize auxiliary WQ handling
op_wait_event is now more lei-specific since we no longer have
to care about oneshot and use a synchronous loop.
{ikw} (import-keywords) started a trend, but LeiPmdir (parallel
Maildir) is an upcoming WQ class that will follow this idea.
Eventually, {l2m} usage may be updated to follow this, too.
Eric Wong [Tue, 8 Jun 2021 09:50:19 +0000 (09:50 +0000)]
lei: safety fix for multiple WQ classes
For commands utilizing multiple workers, this simple change
generalizes the persistence mechanism and and prevents
lei->dclose from causing script/lei to exit if there are
still in-flight workers.
This ougth to prevent read-after-write consistency problems that
occasionally manifest in scripts (e.g. test cases) but usually
go unnoticed in normal use.
Eric Wong [Mon, 7 Jun 2021 19:06:30 +0000 (19:06 +0000)]
lei/store: checkpoint commits mail_sync.sqlite3
We mainly rely on ->done with lei/store, but moving to
->checkpoint probably makes sense. Note: over, msgmap, and
mail_sync all have slightly different transacation behavior;
perhaps they can be unified in the future.
Eric Wong [Sat, 5 Jun 2021 21:06:06 +0000 (21:06 +0000)]
lei: don't drop WQ workers on normal exit
This is dangerous and causes race conditions on commands
which utilize multiple workqueues.
Eric Wong [Sat, 5 Jun 2021 21:04:50 +0000 (21:04 +0000)]
INSTALL: note about lei metadata storage
Since lei is for personal mailboxes, I don't think lei needs to
keep keyword and label changes in history. And fix a minor
wording problem ("or" => "nor") while we're at it.
Eric Wong [Thu, 3 Jun 2021 09:52:06 +0000 (09:52 +0000)]
pkt_op: make pkt_do an OO method
This will make it easier to use for internal use such as
managing Maildir and IMAP IDLE watches.
Eric Wong [Thu, 3 Jun 2021 09:01:24 +0000 (09:01 +0000)]
pkt_op: remove blocking I/O support
Since lei-daemon is guaranteed to be running, there's no
need to keep blocking I/O support around (and we can
get it back via git if we need it).
Followup-to: 1d6e1f9a6a66a42d ("lei: require Socket::MsgHdr or Inline::C, drop oneshot")
Eric Wong [Thu, 3 Jun 2021 01:05:20 +0000 (01:05 +0000)]
lei import: speed up kw updates for old IMAP messages
On a 4-core CPU, this speeds up "lei import" on a largish IMAP
inbox with 75K messages from ~21 minutes down to 40s.
Parallelizing with the new LeiImportKw WQ worker class gives a
near-linear speedup and brought the runtime down to ~5:40.
The new idx_fid_uid index on the "fid" and "uid" columns of
blob2num in mail_sync.sqlite3 brought us the final speedup.
An additional index on over.sqlite3#xref3(oidbin) did not help,
since idx_nntp already exists and speeds up the new ->oidbin_exists
internal API.
I initially experimented with a separate "lei import-kw" command
but decided against it since it's useless outside of IMAP+JMAP
and would require extra cognitive overhead for both users and
hackers. So LeiImportKw is just a WQ worker used by "lei import"
and not its own user-visible command.
v2: fix ikw_done_wait arg handling (ugh, confusing API :x)
Eric Wong [Wed, 2 Jun 2021 10:03:25 +0000 (10:03 +0000)]
lei export-kw: do not write directly to mail_sync.sqlite3
Only the lei/store process should be writing to files/DBs
in lei/store.
Eric Wong [Tue, 1 Jun 2021 19:20:16 +0000 (19:20 +0000)]
lei: remove "forget" (old name for "rm")
"rm" is probably the better name for it, since it matches
"public-inbox-learn rm"
Eric Wong [Tue, 1 Jun 2021 19:14:28 +0000 (19:14 +0000)]
lei_mail_sync: more debug info for uncommitted txn
I'm not actually sure if I hit an uncommitted transaction just
now, it doesn't seem like it.
Eric Wong [Mon, 31 May 2021 10:20:06 +0000 (10:20 +0000)]
lei import: reduce writes to lei/store on IMAP sync
We don't need to write VMD changes to lei/store if local
keywords are unchanged.
Eric Wong [Sun, 30 May 2021 11:45:44 +0000 (11:45 +0000)]
lei import: import IMAP flag changes from old messages
This makes "lei import" behavior with IMAP folders more
consistent with that with Maildir.
Opening IMAP folders read-write with "SELECT" (instead of
read-only with "EXAMINE") was necessary, since it lets an IMAP
server communicate to us as to whether or not it's worth
refetching IMAP flags of previously imported messages.
Fetching UID+FLAGS only is one of the fastest IMAP operations
with dovecot, our -imapd and presumably other common IMAP servers.
It is issued by common MUAs such as mutt after every SELECT.
Users may now rely on "lei import" exclusively to merge mail and
keywords into lei/store, and "lei export-kw" to propagate
keyword changes back to IMAP servers.
A sticks-and-stones workflow for personal mailboxes is currently:
lei import imaps://$MY_PERSONAL_INBOX
lei q --mua=$MUA -o /tmp/results SEARCH TERMS...
# do stuff from within $MUA to /tmp/results
lei import /tmp/results # read keyword changes from MUA
lei export-kw imaps://$MY_PERSONAL_INBOX
# repeat when new stuff shows up in personal inbox
The next goal is to automate repeated imports + export-kw
commands with with inotify and IMAP IDLE.
Eric Wong [Sun, 30 May 2021 06:33:58 +0000 (06:33 +0000)]
lei: support implicit stdin by default
This adds implicit stdin suppport for p2q and lcat,
while rm and rediff no longer need explicit support
for it.
Eric Wong [Sun, 30 May 2021 06:33:57 +0000 (06:33 +0000)]
lei lcat: support maildir: paths, too
This could be helpful in case when a Maildir is on a slow
or unmounted filesystem and lei/store is on fast storage.
Eric Wong [Sun, 30 May 2021 06:33:56 +0000 (06:33 +0000)]
lei lcat: allow IMAP folder URLs w/o UIDVALIDITY
Requiring UIDVALIDITY on the command-line is of course
unreasonable.
Eric Wong [Sun, 30 May 2021 06:33:55 +0000 (06:33 +0000)]
lei lcat+inspect: start wiring up completion
Colons and other delimiters still cause problems for our bash
completion, but some completion is better than no completion.
Eric Wong [Sat, 29 May 2021 20:20:39 +0000 (20:20 +0000)]
lei q: --sort and --save|v2 are incompatible
Saved searches rely on (reverse) docid ordering for efficient
incremental results, and sorting any other way prevents that.
Update comment description in LeiQuery while we're at it:
"ls-query" and "rm-query" are "ls-search" and "forget-search",
respectively, and "mv-query" is implicit with "edit-search"
Eric Wong [Sat, 29 May 2021 20:20:38 +0000 (20:20 +0000)]
lei import|lcat: improve+fix single message IMAP support
lcat can now dump the memoized contents of entire IMAP folders,
not just a single UID. It's now parallelized and pipelined for
multiple lei2mail workers.
Furthemore, various forms of JSON output work consistently
with blob-only output, now.
While working on this, I noticed NetReader was passing UID URLs
to imap_each callbacks, which was causing mail_sync.sqlite3 to
store UIDs in `folders' and clearly wrong so it's now fixed.
Eric Wong [Sat, 29 May 2021 20:20:37 +0000 (20:20 +0000)]
TODO: add pipelining note
Network latency is painful to deal with.
Eric Wong [Sat, 29 May 2021 00:57:19 +0000 (00:57 +0000)]
lei_to_mail: use abs_path for Maildir in mail_sync.sqlite3
lei->rel2abs doesn't resolve symlinks, which could cause
synchronization problems with export-kw or other commands.
Eric Wong [Fri, 28 May 2021 22:37:21 +0000 (22:37 +0000)]
lei q|up: support v2:/path/to/inboxdir destination
This allows "lei-managed pseudo mailing lists" as described
by Konstantin.
Alternates use is optional and can be enables via --shared.
This doesn't manage or edit ~/.public-inbox/config; presumably
there'll need to be some tweaking of search parameters before
finalizing and making the inbox publicly accessible via HTTP/NNTP.
Link: https://public-inbox.org/meta/20210426164454.5zd5kgugfhfwfkpo@nitro.local/T/
Eric Wong [Fri, 28 May 2021 19:47:23 +0000 (19:47 +0000)]
lei: retry_reopen on read-only Xapian access
Xapian DBs may be modified by a parallel process while we're
reading it, and Xapian's MVCC model places the burden on readers
to retry operations.
We'll also have retry_reopen croak instead of die on errors,
which ought to help us track down some "Document not found"
errors I've occasionally seen when using "lei <q|up>".
Eric Wong [Fri, 28 May 2021 09:45:21 +0000 (09:45 +0000)]
script/lei: drop leftover message about fallback
Non-daemon lei isn't implemented, anymore.
Eric Wong [Fri, 28 May 2021 09:45:20 +0000 (09:45 +0000)]
lei: restore working directory in more places
Every tick of the event loop can change the working directory,
so we need to restore it for every client if they operate
in different directories.
This would be easier if we had openat(2) and friends in Perl;
but Inline::C is practically required for lei, now.
Eric Wong [Fri, 28 May 2021 09:45:19 +0000 (09:45 +0000)]
t/lei-*: better diagnostics for occasional failures
Some of these have been failing occasionally, not sure
how, yet...