Date | Commit message (Collapse) |
|
All the augment and deduplication stuff seems to be working
based on unit tests. OpPipe is a nice general addition that
will probably make future state machines easier.
|
|
Most writes to stdout aren't atomic and we need locking to
prevent workers from interleaving and corrupting JSON output.
The one case stdout won't require locking is if it's pointed
to a regular file with O_APPEND; as POSIX O_APPEND semantics
guarantees atomicity.
|
|
The new test ensures consistency between oneshot and
client/daemon users. Cancelling an in-progress result now also
stops xsearch workers to avoid wasted CPU and I/O.
Note the lei->atfork_child_wq usage changes, it is to workaround
a bug in Perl 5: http://nntp.perl.org/group/perl.perl5.porters/258784
<CAHhgV8hPbcmkzWizp6Vijw921M5BOXixj4+zTh3nRS9vRBYk8w@mail.gmail.com>
This switches the internal protocol to use SOCK_SEQPACKET
AF_UNIX sockets to prevent merging messages from the daemon to
client to run pager and kill/exit the client script.
|
|
This internal API is better suited for fork-friendliness (but
locking + dedupe still needs to be re-added).
Normal "json" is the default, though stream-friendly "concatjson"
and "jsonl" (AKA "ndjson" AKA "ldjson") all seem working
(though tests aren't working, yet).
For normal "json", the biggest downside is the necessity of a
trailing "null" element at the end of the array because of
parallel processes, since (AFAIK) regular JSON doesn't allow
trailing commas, unlike JavaScript.
|
|
It's easier to make the code more generic by transferring
all four FDs (std(in|out|err) + socket) instead of omitting
stdin.
We'll be reading from stdin on some imports, and possibly
outputting to stdout, so omitting stdin now would needlessly
complicate things.
The differences with IO::FDPass "1" code paths and the "4"
code paths used by Inline::C and Socket::MsgHdr are far too
much to support and test at the moment.
|
|
For another step in in syscall reduction, we'll support
transferring 3 FDs and a buffer with a single sendmsg/recvmsg
syscall using Socket::MsgHdr if available.
Beyond script/lei itself, this will be used for internal IPC
between search backends (perhaps with SOCK_SEQPACKET). There's
a chance this could make it to the public-facing daemons, too.
This adds an optional dependency on the Socket::MsgHdr package,
available as libsocket-msghdr-perl on Debian-based distros
(but not CentOS 7.x and FreeBSD 11.x, at least).
Our Inline::C version in PublicInbox::Spawn remains the last
choice for script/lei due to the high startup time, and
IO::FDPass remains supported for non-Debian distros.
Since the socket name prefix changes from 3 to 4, we'll also
take this opportunity to make the argv+env buffer transfer less
error-prone by relying on argc instead of designated delimiters.
|
|
Parallelism and interactivity with pager + SIGPIPE needs work;
but results are shown and phrase search works without shell
users having to apply Xapian quoting rules on top of standard
shell quoting.
|
|
The words "extinbox" and "extindex" are too close and easy to
confuse with the other. Rename "extinbox" to "external", since
these could be IMAP, JMAP or other non-public-inbox search APIs.
Link: https://public-inbox.org/meta/20201226112649.GB6226@dcvr/
|
|
I intend to use this with LeiStore when importing from multiple
slow sources at once (e.g. curl, IMAP, etc). This is because
over.sqlite3 can only have a single writer, and we'll have
several slow readers running in parallel.
Watch and SearchIdxShard should also be able to use this code
in the future, but this will be proven with LeiStore, first.
|
|
For writing mboxes and Maildirs, users may wish to use
stricter or looser deduplication strategies. This
gives them more control.
|
|
This is only lightly-tested against stuff LeiToMail generates
and will need real-world tests to validate.
|
|
This is intended for maintaining Maildir states, mbox message
deduplication, but may be useful for other purposes...
|
|
No Maildir, support, yet, but it'll come.
|
|
* origin/master: (58 commits)
ds: flatten + reuse @events, epoll_wait style fixes
ds: simplify EventLoop implementation
check defined return value for localized slurp errors
import: check for git->qx errors, clearer return values
git: qx: avoid extra "local" for scalar context case
search: remove {mset} option for ->mset method
search: remove pointless {relevance} setting
miscsearch: take reopen from Search and use it
extsearch: unconditionally reopen on access
extindex: allow using --all without EXTINDEX_DIR
extindex: add undocumented --no-scan switch
extindex: enable autoflush on STDOUT/STDERR
extindex: various --watch signal handling fixes
extindex: --watch for inotify-based updates
eml: fix undefined vars on <Perl 5.28
t/config: test --get-urlmatch for git <2.26
default to CORE::warn in $SIG{__WARN__} handlers
inbox: name variable for values loop iterator
inboxidle: avoid needless syscalls on refresh
inboxidle: clue users into resolving ENOSPC from inotify
...
|
|
While a single extindex combines multiple inboxes into a single
search index, extindex still requires up-front indexing on items
which can be searched. XSearch has no on-disk footprint itself
and uses Xapian DBs of existing publicinbox and extindex
("extinbox") exclusively.
XSearch still suffers from the multi-shard Xapian scalability
problems which led to the creation of extindex, but I expect the
number of shards to remain relatively low.
I envision users hosting public-inbox instances on their
workstations will only have two extindex combined by this, one
read-only extindex for serving public archives, and one
read-write extindex managed by LeiStore for private mail.
|
|
I've been using something like this to mock out thousands
of inboxes for testing.
|
|
They need to be indexed by MiscIdx, but MiscIdx
still needs more work to support faster config
loading when dealing with ~100K data sources.
|
|
This could've been done ages ago, but I rarely invoked
public-inbox-* commands from an interactive terminal
like I would with lei.
|
|
Much work still needs to be done, but that goes for this
entire project :P
|
|
This is a localized version of the process-wide END{}, but runs
at the end of variable scope. A subroutine ref and arguments
may be passed, which allows us to avoid anonymous subs and
problems they cause.
It's similar to `defer' or `ensure' in other languages; Perl can
rely on deterministic destructors due to refcounting.
|
|
"LEI" is an acronym, and ALL CAPS is consistent with existing
PublicInbox::{IMAP,HTTP,NNTP,WWW} naming for top-level modules,
3 of 4 old ones which deal directly with sockets and requests.
|
|
We can use the same "local $ENV{FOO}" hack we do with
t/nntpd-v2.t to test the oneshot code path without imposing
an extra script in the users' $PATH.
|
|
Still unstable, this builds off the equally unstable extindex :P
This will be used for caching/memoization of traditional mail
stores (IMAP, Maildir, etc) while providing indexing via Xapian,
along with compression, and checksumming from git.
Most notably, this adds the ability to add/remove per-message
keywords (draft, seen, flagged, answered) as described in the
JMAP specification (RFC 8621 section 4.1.1).
We'll use `.' (a single period) as an $eidx_key since it's an
invalid {inboxdir} or {newsgroup} name.
|
|
The start of lei, a Local Email Interface. It'll support a
daemon via FD passing to avoid startup time penalties if
IO::FDPass is installed, but fall back to a slow one-shot mode
if not.
Compared to traditional socket daemon, FD passing should allow
us to eventually do stuff like run "git show" and still have
proper terminal support for pager and color.
|
|
The CLI tool still needs usability work, and "misc" is still in
flux, but the core message indexing part is stable (since it's
stolen from v2 :P).
|
|
Using "eidx_key:" boolean prefix to limit results to a given
inbox, we can use ->ALL to emulate and replace per-Inbox
xap15/[0-9] search indices.
With this change, the presence of "extindex.all.topdir" in the
$PI_CONFIG will cause the WWW code to use that extindex and
ignore per-inbox Xapian DBs in xap15/[0-9].
Unfortunately IMAP search still requires old per-inbox indices,
for now. Mapping extindex Xapian docids to per-Inbox UIDs and
vice-versa is proving tricky. Fortunately, IMAP search is
rarely used and optional. The RFCs don't specify expensive
phrase search, either, so `indexlevel=medium' can be used in
per-inbox Xapian indices to save space.
For primarily WWW (and future JMAP) users; this should result in
significant disk space, FD, and page cache footprint savings for
large instances with many inboxes and many cross-posted
messages.
|
|
We must use the result of link_refs() since it can trigger
merge_threads() and invalidate $old_tid. In case
merge_threads() isn't triggered, link_refs() will return
$old_tid anyways.
When rethreading and allocating new {tid}, we also must update
the row where the now-expired {tid} came from to ensure only the
new {tid} is seen when reindexing subsequent messages in
history. Otherwise, every subsequently reindexed+rethreaded
message could end up getting a new {tid}.
Reported-by: Kyle Meyer <kyle@kyleam.com>
Link: https://public-inbox.org/meta/87360nlc44.fsf@kyleam.com/
|
|
This will be used to index and search Inbox objects and perhaps
individual git repositories/epochs for grokmirror manifest.js.gz
generation. There is no sharding planned for this at the moment
since inbox count should remain low (~100K to 1M) compared to
message count.
Folding this into the existing sharded DBs could be possible;
but would likely increase query and maintenance costs, as well
as development complexity. So we'll use a few more inodes and
FDs at runtime, instead.
|
|
Upon "eindex" rhymes with "reindex", which could be confusing;
so name the command and config prefix to use "extindex" which
is hopefully less confusing.
|
|
Not documented, yet, but it runs...
|
|
It compiles...
|
|
This will provide a similar API to PublicInbox::Inbox for
read-only WWW, -imapd, and -nntpd interfaces.
|
|
If there's no body ({bdy} field), ->each_part set the {bdy}
field to "\n" and the ->as_string result afterwards is one
extra "\n" byte longer than the original.
It's not worth extra cycles in common ->each_part calls to
ensure 100% round-trip matches of header-only messages (which
are likely spam), especially when the only difference is a
trailing "\n".
|
|
Unlike Email::MIME, PublicInbox::Eml::as_string should be able
to round trip from the Perl object to a raw scalar and back
without changes.
|
|
I should've dropped "PENDING" notes before the 1.6 release;
they're dropped now, and a note is added to remind my future
self to drop them before 1.7.
|
|
It seems easiest to have a singleton Gcf2Client client object
per daemon worker for all inboxes to use. This reduces overall
FD usage from pipes.
The `public-inbox-gcf2' command + manpage are gone and a `$^X'
one-liner is used, instead. This saves inodes for internal
commands and hopefully makes it easier to avoid mismatched
PERL5LIB include paths (as noticed during development :x).
We'll also make the existing cat-file process management
infrastructure more resilient to BOFHs on process killing
sprees (or in case our libgit2-based code fails on us).
(Rare) PublicInbox::WWW PSGI users NOT using public-inbox-httpd
won't automatically benefit from this change, and extra
configuration will be required (to be documented later).
|
|
This should be able to replace multiple `git cat-file' for blob
retrieval, but adjustments may be needed.
|
|
Having tens of thousands of inboxes and associated git processes
won't work well, so we'll use libgit2 to access the object DB
directly. We only care about OID lookups and won't need to rely
on per-repo revision names or paths.
The Git::Raw XS package won't be used since its manpages don't
promise a stable API. Since we already use Inline::C and have
experience with I::C when it comes to compatibility, this only
introduces libgit2 itself as a source of new incompatibilities.
This also provides an excuse for me to writev(2) to reduce
syscalls, but liburing is on the horizon for next year.
|
|
We will need to allow simultaneous iterators on the same
config object, since we'll need this for ExtMsg, NNTPD,
WwwListing, NewsWWW, and other places.
|
|
It's still as slow as before with hundreds/thousands of inboxes,
but at least it's fair. Future changes will allow it to be
cached and memoized with persistent HTTP servers.
|
|
While it's not a known problem, our deduplicating logic may
change in the future; or a BOFH could be manually injecting
duplicate messages directly into the git epoch repositories.
Ensure indexing in mirrors doesn't break when there's
duplicates. This is in preparation for detached indices
for multi-inbox search.
|
|
ParentPipe was a subset of EOFpipe, except EOFpipe correctly
accounts for theoretical(*) spurious wakeups on the pipe.
(*) AFAIK, spurious wakeups are/were more likely on TCP sockets
due to checksum failures, something that's not a problem on
local pipes. We're also not sharing pipes like we do with
listen sockets on accept(2), so there's no chance of another
process grabbing bytes (unless we have bugs in our code).
|
|
It's a bit inefficient to use a pipe, here. However, using
dwaitpid() on a process that's not expected to exit soon is
also inefficient as it causes excessive wakeups as most of
our inbox-writing code expects synchronous waitpid().
This only affects -watch instances configured for NNTP and IMAP
clients.
|
|
This is no longer limited to Maildirs now that IMAP and NNTP
support exist; so give it a shorter name.
|
|
We've got examples for all the other daemons, too!
|
|
Since this was already a separate package, split it off
into its own file since SearchView may not handle inbox
groups.
|
|
Determining storage device speed and latencies doesn't
seem portable or even possible with the wide variety
of storage layers in use.
This means we need to write a tuning document and hope
users read and improve on it :P
|
|
No need to waste resources when doing minimal work. With
PI_TEST_VERSION=2, this fixes a test failure where
Net::NNTP::DESTROY was getting called in the shard process.
We'll also get rid of an unnecessary use_ok under v2, too.
|
|
SQLite and Xapian files are written randomly, thus they become
fragmented under btrfs with copy-on-write. This leads to
noticeable performance problems (and probably ENOSPC) as these
files get big.
lore/git (v2, <1GB) indexes around 20% faster with this on an
ancient SSD. lore/lkml seems to be taking forever and I'll
probably cancel it to save wear on my SSD.
Unfortunately, disabling CoW also means disabling checksumming
(and compression), so we'll be careful to only set the No_COW
attribute on regeneratable data. We want to keep CoW (and
checksums+compression) on git storage because current ref
storage is neither checksummed nor compressed, and git streams
pack output.
|
|
This avoids pinning a potentially large chunk of memory from
`git-log --reverse' into RAM (or triggering less predictable
swap behavior). Instead it uses a contiguous temporary file
with a fixed-size record for every blob we'll need to index.
|