| Date | Commit message (Collapse) |
|---|
|
IMAP requires either the Email::Address::XS or Mail::Address
package (part of perl-MailTools RPM or libmailtools-perl deb);
and Email::Address::XS is not officially packaged for some older
distros, most notably CentOS 7.x.
|
|
For v1 inboxes (and possibly v2 in the future, for VACUUM),
public-inbox-compact replaces over.sqlite3 with a new file.
This currently doesn't need an extra inotify watch descriptor
(or FD for kevent) at the moment, so it can coexist nicely for
systems w/o IO::KQueue or Linux::Inotify2.
|
|
Since we limit our mailboxes slices to 50K and can guarantee a
contiguous UID space for those mailboxes, we can store a mapping
of "UID offsets" (not full UIDs) to Message Sequence Numbers as
an array of 16-bit unsigned integers in a 100K scalar.
For UID-only FETCH responses, we can momentarily unpack the
compact 100K representation to a ~1.6M Perl array of IV/UV
elements for a slight speedup.
Furthermore, we can (ab)use hash key deduplication in Perl5 to
deduplicate this 100K scalar across all clients with the same
mailbox slice open.
Technically we can increase our slice size to 64K w/o increasing
our storage overhead, but I suspect humans are more accustomed
to slices easily divisible by 10.
|
|
Supporting MSNs in long-lived connections beyond the lifetime of
a single request/response cycle is not scalable to a C10K
scenario. It's probably not needed, since most clients seem to
use UIDs.
A somewhat efficient implementation I can come up uses
pack("S*" ...) (AKA "uint16_t mapping[50000]") has an overhead
of 100K per-client socket on a mailbox with 50K messages. The
100K is a contiguous scalar, so it could be swapped out for
idle clients on most architectures if THP is disabled.
An alternative could be to use a tempfile as an allocator
partitioned into 100K chunks (or SQLite); but I'll only do that
if somebody presents a compelling case to support MSN SEARCH.
|
|
The sort was unstable on my test instance anyways, and
clients don't seem to mind. So stop wasting CPU cycles.
|
|
Simple queries work, more complex queries involving parentheses,
"OR", "NOT" don't work, yet.
Tested with "=b", "=B", and "=H" search and limits in mutt
on both v1 and v2 with multiple Xapian shards.
|
|
We can get exact values for EXISTS, UIDNEXT using SQLite
rather than calculating off $ibx->mm->max ourselves.
Furthermore, $ibx->mm is less useful than $ibx->over for IMAP
(and for our read-only daemons in general) so do not depend on
$ibx->mm outside of startup/reload to save FDs and reduce kernel
page cache footprint.
|
|
This appears to significantly improve header caching behavior
with mutt. With the current public-inbox.org/git mirror(*),
mutt will only re-FETCH the last ~300 or so messages in the
final "inbox.comp.version-control.git.7" mailbox, instead of
~49,000 messages every time.
It's not perfect, but a 500ms query is better than a >10s query
and mutt itself spends as much time loading its header cache.
(*) there are many gaps in NNTP article numbers (UIDs) due to
spam removal from public-inbox-learn.
|
|
This speeds up requests from mutt for HEADER.FIELDS by around 10%
since we don't waste time doing CRLF conversion on large message
bodies that get discarded, anyways.
|
|
While selecting a mailbox is done case-insensitively, "INBOX" is
special for the LIST command, according to RFC 3501 6.3.8:
> The special name INBOX is included in the output from LIST, if
> INBOX is supported by this server for this user and if the
> uppercase string "INBOX" matches the interpreted reference and
> mailbox name arguments with wildcards as described above. The
> criteria for omitting INBOX is whether SELECT INBOX will
> return failure; it is not relevant whether the user's real
> INBOX resides on this or some other server.
Thus, the existing news.public-inbox.org convention of naming
newsgroups starting with "inbox." needs to be special-cased to
not confuse clients.
While we're at it, do not create ".0" for dummy newsgroups if
they're selected, either.
|
|
NNTP and IMAP both require CRLF conversions on the wire.
They're also the only components which care about
$smsg->{bytes}, so store the CRLF-adjusted value in over.sqlite3
and Xapian DBs..
This will allow us to optimize RFC822.SIZE fetch item in IMAP
without triggering size mismatch errors in some clients' default
configurations (e.g. Mail::IMAPClient), but not most others.
It could also fix hypothetical problems with NNTP clients that
report discrepancies between overview and article data.
|
|
This is just a hair faster and cacheable in the future, if we
need it. Most notably, this avoids doing PublicInbox::Eml->new
for simple "RFC822", "BODY[]", and "RFC822.SIZE" requests.
|
|
Dummy messages make for bad user experience with MUAs which
still use sequence numbers. Not being able to fetch a message
doesn't seem fatal in mutt, so just ignore (sometimes large)
gaps.
|
|
We'll need to support searching UID ranges for IMAP,
so make sure it's indexed, too.
|
|
Searching for messages smaller than a certain size is allowed by
offlineimap(1), mbsync(1), and possibly other tools. Maybe
public-inbox-watch will support it, too.
I don't see a reason to expose searching by size via WWW search
right now (but maybe in the future, I could be convinced to).
Note: we only store the byte-size of the message in git,
this is typically LF-only and we won't have the correct
size after CRLF conversion for NNTP or IMAP.
|
|
Since it seems somewhat common for IMAP clients to limit
searches by sent Date: or INTERNALDATE, we can rely on
the NNTP/WWW-optimized overview DB.
For other queries, we'll have to depend on the Xapian DB.
|
|
None of the new cases are wired up, yet, but existing cases
still work.
|
|
No point in spewing "uninitialized" warnings into logs when
the cat jumps on the Enter key.
|
|
We'll use the xqx() to avoid losing too much performance
compared to normal `backtick` (qx) when testing using
"make check-run" + Inline::C.
|
|
Having two large numbers separated by a dash can make visual
comparisons difficult when numbers are in the 3,000,000 range
for LKML. So avoid the $UID_END value, since it can be
calculated from $UID_MIN. And we can avoid large values of
$UID_MIN, too, by instead storing the block index and just
multiplying it by 50000 (and adding 1) on the server side.
Of course, LKML still goes up to 72, at the moment.
|
|
I'm not sure this matters, and it could be a waste of
CPU cycles if no real clients care. However, it does
make debugging over telnet or s_client a bit easier.
|
|
Finish up the IMAP-only portion of iterative config reloading,
which allows us to create all sub-ranges of an inbox up front.
The InboxIdler still uses ->each_inbox which will struggle with
100K inboxes.
Having messages in the top-level newsgroup name of an inbox will
still waste bandwidth for clients which want to do full syncs
once there's a rollover to a new 50K range. So instead, make
every inbox accessible exclusively via 50K slices in the form of
"$NEWSGROUP.$UID_MIN-$UID_END".
This introduces the DummyInbox, which makes $NEWSGROUP
and every parent component a selectable, empty inbox.
This aids navigation with mutt and possibly other MUAs.
Finally, the xt/perf-imap-list maintainer test is broken, now,
so remove it. The grep perlfunc is already proven effective,
and we'll have separate tests for mocking out ~100k inboxes.
|
|
IMAP RFC 3501 stipulates case-insensitive comparisons, and so
does RFC 977 (NNTP). However, INN (nnrpd) uses case-sensitive
comparisons, so we've always used case-sensitive comparisons for
NNTP to match nnrpd behavior.
Unfortunately, some IMAP clients insist on sending "INBOX" with
caps, which causes problems for us. Since NNTP group names are
typically all lowercase anyways, just force all comparisons to
lowercase for IMAP and warn admins if uppercase-containing
newsgroups won't be accessible over IMAP.
This ensures our existing -nntpd behavior remains unchanged
while being compatible with the expectations of real-world IMAP
clients.
|
|
"$UID_START:*" needs to return at least one message according
to RFC 3501 section 6.4.8.
While we're in the area, coerce ranges to (unsigned) integers by
adding zero ("+ 0") to reduce memory overhead.
|
|
We'll be using this wrapper class to workaround some upstream
bugs in Mail::IMAPClient. There may also be experiments with
new APIs for more performance.
|
|
This matches the behavior of the existing synchronous ->cat_file
method. In fact, ->cat_file now becomes a small wrapper around
the ->cat_async method.
|
|
This will make it easier to implement the retries on
alternates_changed() of the synchronous ->cat_file API.
|
|
The RFC 3501 `sequence-set' definition allows comma-delimited
ranges, so we'll support it in case clients send them.
Coalescing overlapping ranges isn't required, so we won't
support it as such an attempt to save bandwidth would waste
memory on the server, instead.
|
|
This ought to improve overall performance with multiple clients.
Single client performance suffers a tiny bit due to extra
syscall overhead from epoll.
This also makes the existing async interface easier-to-use,
since calling cat_async_begin is no longer required.
|
|
While we can't memoize the regexp forever like we do with other
Eml users, we can still benefit from caching regexp compilation
on a per-request basis.
A FETCH request from mutt on a 4K message inbox is around 8%
faster after this. Since regexp compilation via qr// isn't
unbearably slow, a shared cache probably isn't worth the
trouble of implementing. A per-request cache seems enough.
|
|
It seems worthless to support CLOSE for read-only inboxes, but
mutt sends it, so don't return a BAD error with proper use.
|
|
They're not specified in RFC 3501 for responses, and at least
mutt fails to handle it.
|
|
We'll return dummy messages for now when sequence numbers go
missing, in case clients can't handle missing messages.
|
|
We must keep the contents of {-partial} around when handling
a request to fetch multiple messages.
|
|
This makes the test code easier-to-manage and allows us to run
faster unit tests which don't involve loading Mail::IMAPClient.
|
|
IMAP supports a high level of granularity when it comes to
fetching, but fortunately Perl makes it fairly easy to support.
|
|
Instead of counts starting at 0, we start the single-part
message at 1 like we do with subparts of a multipart message.
This will make it easier to map offsets for "BODY[$SECTION]"
when using IMAP FETCH, since $SECTION must contain non-zero
numbers according to RFC 3501.
This doesn't make any difference for WWW URLs, since single part
messages cannot have downloadable attachments.
|
|
I'm not sure which clients use these, but it could be useful
down the line.
|
|
We can fill in some missing pieces from the emulation APIs
to enable IMAP IDLE tests on non-Linux platforms.
|
|
We'll optimize for the common case of: $TAG LIST "" *
and rely on the grep perlfunc to handle trickier cases.
|
|
I'm not sure if there's much use for this command, but it's
part of RFC3501 and works read-only.
|
|
InboxIdle should not be holding onto Inbox objects after the
Config object they came from expires, and Config objects may
expire on SIGHUP.
Old Inbox objects still persist due to IMAP clients holding onto
them, but that's a concern we'll deal with at another time, or
not at all, since all clients expire, eventually.
Regardless, stale inotify watch descriptors should not be left
hanging after SIGHUP refreshes.
|
|
It seems to be working as far as Mail::IMAPClient is concerned.
|
|
This will be used to implement IMAP IDLE, first.
Eventually, it may be used to trigger other things:
* incremental internal updates for manifest.js.gz
* restart `git cat-file' processes on pack index unlink
* IMAP IDLE-like long-polling HTTP endpoint
And maybe more things we haven't thought of, yet.
It uses Linux::Inotify2 or IO::KQueue depending on what packages
are installed and what the kernel supports. It falls back to
nanosecond-aware Time::HiRes::stat() (available with Perl 5.10.0+)
on systems lacking Linux::Inotify2 and IO::KQueue.
In the future, a pure Perl alternative to Linux::Inotify2 may be
supplied for users of architectures we already support signalfd
and epoll on.
v2 changes:
- avoid O_TRUNC on lock file
- change ctime on Linux systems w/o inotify
- fix naming of comments and fields
|
|
It shares a bit of code with NNTP. It's copy+pasted for now
since this provides new ground to experiment with APIs for
dealing with slow storage and many inboxes.
|
|
InboxWritable should only set $v2w->{parallel} if the $parallel
flag is defined to 0 or 1. We want indexing a new inbox to
utilize SMP, just like --reindex.
-index once again allows -j0/--jobs=0 to force single-process
use, and we'll be ensuring that works in tests to maintain
performance on small systems.
Fixes: 61a2fff5b34a3e32 ("admin: move index_inbox over")
|
|
We'll continue to favor simpler data models that can be
used directly rather than wasting time and memory with
accessor APIs.
The ->from, ->to, -cc, ->mid, ->subject, >references methods can
all be trivially replaced by hash lookups since all their values
are stored in doc_data. Most remaining callers of those methods
were test cases, anyways.
->from_name is only used in the PSGI code, so we can just
use ->psgi_cull to take care of populating the {from_name}
field.
|
|
To further simplify callers and avoid embarrasing memory
explosions[1], we can finally eliminate this method in
favor of smsg_eml.
[1] commit 7d02b9e64455831d3bda20cd2e64e0c15dc07df5
("view: stop storing all MIME objects on large threads")
fixed a huge memory blowup.
|
|
This will eventually replace the __hdr() calling methods and
eradicate {mime} usage from Smsg. For now, we can eliminate
PublicInbox::Smsg->new since most callers already rely on an
open `bless' to avoid the old {mime} arg.
|
|
gitweb does the same with $GIT_DIR/description and gitweb.owner.
Allowing UTF-8 description should not cause problems when used
in responses for to the NNTP "LIST NEWSGROUPS" request, either,
since RFC 3977 section 7.6.6 recommends the description be UTF-8
(but does not require it).
Link: https://public-inbox.org/meta/20200528151216.l7vmnmrs4ojw372g@sourcephile.fr/
|