]> Sergey Matveev's repositories - public-inbox.git/log
public-inbox.git
6 years agowatch: improve fairness during full rescans
Eric Wong [Sat, 24 Jun 2017 07:33:44 +0000 (07:33 +0000)]
watch: improve fairness during full rescans

We need to ensure new messages are being processed
fairly during full rescans, so have the ->scan subroutine
yield and reschedule itself.  Additionally, having a
long-running task inside the signal handler is dangerous
and subject to reentrancy bugs.

Due to the limitations of the Filesys::Notify::Simple interface,
we cannot rely on multiplexing I/O interfaces (select, IO::Poll,
Danga::Socket, etc...) for this.  Forking a separate process
was considered, but it is more expensive for a mostly-idle
process.

So, we use a variant of the "self-pipe trick" via inotify (or
whatever Filesys::Notify::Simple gives us).  Instead of writing
to our own pipe, we write to a file in our own temporary
directory watched by Filesys::Notify::Simple to trigger events
in signal handlers.

6 years agospamc: retry on EINTR
Eric Wong [Sat, 24 Jun 2017 00:52:10 +0000 (00:52 +0000)]
spamc: retry on EINTR

Signals can fire on us at any time if we're using blocking sysread.

6 years agowatch: ensure HUP causes the scanner to be reloaded
Eric Wong [Sat, 24 Jun 2017 00:00:04 +0000 (00:00 +0000)]
watch: ensure HUP causes the scanner to be reloaded

Otherwise the old watcher may run indefinitely

6 years agomda: set List-ID correctly according to RFC2919
Eric Wong [Mon, 26 Jun 2017 03:05:39 +0000 (03:05 +0000)]
mda: set List-ID correctly according to RFC2919

Oops, due to an old mistake , List-ID was set incorrectly
in the MDA.  This could cause some breakage w.r.t. mail filters.

6 years agolinkify: handle URLs in parenthesized statements
Eric Wong [Fri, 23 Jun 2017 22:42:34 +0000 (22:42 +0000)]
linkify: handle URLs in parenthesized statements

Sometimes, URLs exist at the end of parethesized statements,
and we shouldn't unnecessarily capture that.

(example: https://public-inbox.org/ruby-core/20170623032722.GA8124@dcvr/)

6 years agoallow admins to configure non-obfuscated addresses/domains
Eric Wong [Fri, 23 Jun 2017 20:23:07 +0000 (20:23 +0000)]
allow admins to configure non-obfuscated addresses/domains

We will also treat all known list addresses as non-obfuscated.

By setting publicinbox.noObfuscate in ~/.public-inbox/config,
this will allow users to disable address obfuscation on a
per-domain or per-address basis.

6 years agoconfig: assume lists have multiple addresses
Eric Wong [Fri, 23 Jun 2017 19:41:51 +0000 (19:41 +0000)]
config: assume lists have multiple addresses

This should simplify the rest of our code for handling
the do-not-obfuscate list.

6 years agoview: add newline before mailto: instructions in reply
Eric Wong [Fri, 23 Jun 2017 09:51:47 +0000 (09:51 +0000)]
view: add newline before mailto: instructions in reply

This is necessary to retain consistent spacing around bullet
points.

Fixes: 666844ae42b5b17f ("reply: handle address obfuscation :<")
6 years agombox: show application/mbox for obfuscated inboxes
Eric Wong [Fri, 23 Jun 2017 03:39:08 +0000 (03:39 +0000)]
mbox: show application/mbox for obfuscated inboxes

Sigh, yet another place to handle obfuscation for misguided
people who expect it.  Maybe this will do something to prevent
spammers from getting addresses, while still allowing the
"curl $URL | git am" use case to work.

6 years agoreply: handle address obfuscation :<
Eric Wong [Fri, 23 Jun 2017 02:25:33 +0000 (02:25 +0000)]
reply: handle address obfuscation :<

We can show users a lightly-obfuscated Bourne shell command
for invoking "git send-email" for address obfuscation.  However,
I'm not sure if the mailto: arg will work effectively since
URL encoding is probably too well-known to be effective.

6 years agosearchidx: fallback to lookup on pre-set article numbers
Eric Wong [Fri, 23 Jun 2017 01:43:21 +0000 (01:43 +0000)]
searchidx: fallback to lookup on pre-set article numbers

Yet another hiccup from reusing pre-set article numbers on
various ruby-lang.org mailing lists.  This was causing messages
to not appear to NNTP readers which use XOVER.

6 years agomsgmap: ignore duplicates instead of dying
Eric Wong [Fri, 23 Jun 2017 01:19:46 +0000 (01:19 +0000)]
msgmap: ignore duplicates instead of dying

This prevents public-inbox-watch from dying when reloading
(and thus rescanning) already-imported directories.

6 years agowatchmaildir: deal with rejected (100) messages
Eric Wong [Fri, 23 Jun 2017 00:47:23 +0000 (00:47 +0000)]
watchmaildir: deal with rejected (100) messages

The RubyLang filter is strict about what messages it rejects, so
the spam learning path will not auto-train or remove messages
missing X-Mail-Count headers.

6 years agotest for PublicInbox::Filter::RubyLang
Eric Wong [Thu, 22 Jun 2017 21:25:39 +0000 (21:25 +0000)]
test for PublicInbox::Filter::RubyLang

This will make it easier to prevent breakage in the future.

6 years agofilter/rubylang: reuse altid entry from inbox object
Eric Wong [Thu, 22 Jun 2017 21:04:53 +0000 (21:04 +0000)]
filter/rubylang: reuse altid entry from inbox object

This allows users to DRY up their config a bit and avoid
specifying altid twice when reusing the NNTP-centric msgmap
for [ruby-*:\d+] serial numbers.

My current work-in-progress ~/.public-inbox/config entry
for the ruby-core list is:

------8<-------
[publicinbox "ruby-core"]
address = ruby-core@ruby-lang.org
url = //public-inbox.org/ruby-core
mainrepo = /path/to/ruby-core.git
newsgroup = inbox.comp.lang.ruby.core
watchheader = List-Id:<ruby-core.ruby-lang.org>
altid = serial:ruby-core:file=msgmap.sqlite3
watch = maildir:/path/to/Maildir/.INBOX.ruby
filter = PublicInbox::Filter::RubyLang

6 years agomsgmap: mid_insert ignores duplicates instead of die-ing
Eric Wong [Thu, 22 Jun 2017 19:51:23 +0000 (19:51 +0000)]
msgmap: mid_insert ignores duplicates instead of die-ing

This will allow smoother imports as occasional Message-ID
duplicates happen and the best we can do is ignore the
second one.

6 years agoadd filter for RubyLang lists
Eric Wong [Wed, 21 Jun 2017 23:33:49 +0000 (23:33 +0000)]
add filter for RubyLang lists

Unfortunately, it appears we have to reject this and instead add
support filtering at View time(*), due to DKIM signatures in
messages from ruby-lang.org.

(*) which may not be worth it

6 years agoimport: fix encoding issues from weird "raw" emails
Eric Wong [Tue, 20 Jun 2017 22:06:54 +0000 (22:06 +0000)]
import: fix encoding issues from weird "raw" emails

This seems to allow weirdly-encoded "raw" emails in
  blade.nagaokaut.ac.jp/ruby/ruby-core/*
to be handled without difficulties.

6 years agoview: implement optional address obfuscation
Eric Wong [Fri, 16 Jun 2017 02:03:32 +0000 (02:03 +0000)]
view: implement optional address obfuscation

This is lightly-tested and seems to work.  I'm still
hesitant to support this, but the alternative of receiving death
threats for displaying unobfuscated addresses seems to
be not worth it.

6 years agoreply: support Reply-To
Eric Wong [Wed, 14 Jun 2017 00:10:53 +0000 (00:10 +0000)]
reply: support Reply-To

Reply-To is common and probably should've been supported,
since day one, but we won't omit other addresses, either.

6 years agoreplyto parameter support
Eric Wong [Wed, 14 Jun 2017 00:10:52 +0000 (00:10 +0000)]
replyto parameter support

This allows us to support centralized mailing lists (which suck,
but better than no mailing list at all).

6 years agoview: split out reply logic into its own module
Eric Wong [Wed, 14 Jun 2017 00:10:51 +0000 (00:10 +0000)]
view: split out reply logic into its own module

We'll be adding more reply options for centralized mailing
lists.  So split out the logic so it's easy-to-find.
Organizing code is hard :<

6 years agosearchidx: remove messages correctly from Xapian index
Eric Wong [Thu, 15 Jun 2017 23:07:58 +0000 (23:07 +0000)]
searchidx: remove messages correctly from Xapian index

This fixes a bug introduced in
commit 7eeadcb62729b0efbcb53cd9b7b181897c92cf9a
("search: remove unnecessary abstractions and functionality")

6 years agosearch: allow searching within mail diffs
Eric Wong [Wed, 14 Jun 2017 00:14:48 +0000 (00:14 +0000)]
search: allow searching within mail diffs

This can be tied into a repository browser to browse
in-flight topics on a mailing list.

6 years agosearchidx: switch to accounting by message bytes
Eric Wong [Wed, 14 Jun 2017 00:14:47 +0000 (00:14 +0000)]
searchidx: switch to accounting by message bytes

Xapian memory usage is tied to the size of the indexed
text, so take the raw message size into account when
deciding when to flush Xapian data.

More importantly, we now flush Xapian before we have it
buffer beyond our maximum; and we do it unconditionally
to prevent even high priority processes from OOM-ing.

6 years agosearch: remove unnecessary abstractions and functionality
Eric Wong [Wed, 14 Jun 2017 00:14:46 +0000 (00:14 +0000)]
search: remove unnecessary abstractions and functionality

This simplifies the code a bit and reduces the translation
overhead for looking directly at data from tools shipped
with Xapian.

While we're at it, fix thread-all.t :)

6 years agofilter/subjecttag: account for missing Subject: header
Eric Wong [Fri, 12 May 2017 18:49:32 +0000 (18:49 +0000)]
filter/subjecttag: account for missing Subject: header

This is a high indicator of spam (but out-of-scope for this
particular module) but sometimes it is not, and people
legitimately forget to set a Subject: header at all.

6 years agoimport: reset :raw mode for commit title (subject)
Eric Wong [Thu, 25 May 2017 02:24:16 +0000 (02:24 +0000)]
import: reset :raw mode for commit title (subject)

This was necessary for the presence of the 0xa0 byte(*)
in the Subject: of the message at:
http://blade.nagaokaut.ac.jp/ruby/ruby-core/3220

(*) That is 0xa0, not 0x0a ("\n"), so I wonder if the
    nibbles got swapped somehow.

6 years agosearchview: retry queries if uri_unescape-able
Eric Wong [Tue, 23 May 2017 23:07:24 +0000 (23:07 +0000)]
searchview: retry queries if uri_unescape-able

It is possible to have double-escaped queries when copy and
pasting into browsers, so try to help users work around this
common error by automatically retrying after unescaping once.

Of course, we must inform the user when doing this results in
success, in case they really meant to search for a
double-escaped term which resulted in nothing.

Reported-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
  https://public-inbox.org/meta/CACBZZX5Gnow08r=0A1J_kt3a=zpGyMfvsqu8nAN7kacNnDm+dg@mail.gmail.com/

6 years agowww: do not mangle characters from search queries
Eric Wong [Tue, 23 May 2017 21:53:57 +0000 (21:53 +0000)]
www: do not mangle characters from search queries

Reported-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
  https://public-inbox.org/meta/CACBZZX5Gnow08r=0A1J_kt3a=zpGyMfvsqu8nAN7kacNnDm+dg@mail.gmail.com/

6 years agowww: avoid undefined warnings for query string parsing
Eric Wong [Tue, 9 May 2017 20:43:33 +0000 (20:43 +0000)]
www: avoid undefined warnings for query string parsing

Sometimes bots generate malformed queries with sequential
"&" and ";" characters.

6 years agowatchmaildir: show $@ in warning message
Eric Wong [Tue, 9 May 2017 06:30:42 +0000 (06:30 +0000)]
watchmaildir: show $@ in warning message

It should be helpful to know what error happened.

6 years agosearchidx: use cached local $@ copy
Eric Wong [Tue, 9 May 2017 06:30:41 +0000 (06:30 +0000)]
searchidx: use cached local $@ copy

umask should never fail and set $@, but use the cached local
to be more explicit just in case.

6 years agospamassassin: update example ~/.spamassassin/user_prefs file
Eric Wong [Sun, 7 May 2017 00:46:46 +0000 (00:46 +0000)]
spamassassin: update example ~/.spamassassin/user_prefs file

This is closer to what I run on the public-inbox.org servers.

6 years agosearchidx: fix ghost root vivification
Eric Wong [Sun, 7 May 2017 10:49:00 +0000 (10:49 +0000)]
searchidx: fix ghost root vivification

Due to the asynchronous nature of SMTP, it is possible for the
root message of a thread (with no References/In-Reply-To)
to arrive last in a series.  We must preserve the thread_id
of the ghost message in this case, as we do when vivifiying
non-root ghosts.

Otherwise, this causes threads to be broken when the root
arrives last.

7 years agosearch: fix help message for searching within quotes
Eric Wong [Tue, 11 Apr 2017 23:39:54 +0000 (23:39 +0000)]
search: fix help message for searching within quotes

I'm not sure if people use either and it's not in mairix
(where we base our abbreviations off of).  Lets go
with the shorter prefix since it's easier-to-type.

7 years agolearn: scan all inboxes when learning spam
Eric Wong [Wed, 5 Apr 2017 01:41:28 +0000 (01:41 +0000)]
learn: scan all inboxes when learning spam

This matches the behavior of the -watch daemon since
6d534038285ddd760709ba76ea007f9108200097
("watch: watchspam affects all configured inboxes")

7 years agowatchmaildir: do not reject lowercase flags on Maildir files
Eric Wong [Tue, 4 Apr 2017 18:25:47 +0000 (18:25 +0000)]
watchmaildir: do not reject lowercase flags on Maildir files

Dovecot uses 'a'..'z' (lowercase) to designate keywords
in Maildir flags.  This was preventing certain messages
from being marked as spam.

https://wiki2.dovecot.org/MailboxFormat/Maildir

7 years agosearchview: show full (&x=t) messages in ascending chronlogical order
Eric Wong [Fri, 24 Mar 2017 01:41:11 +0000 (01:41 +0000)]
searchview: show full (&x=t) messages in ascending chronlogical order

When displaying search results with full messages, it makes
more sense to show them in ascending chronological order when
going by date.  Reverse chronological order makes more sense
for search results which only show the subject.

7 years agosearchview: add "t" id to link to thread overview
Eric Wong [Fri, 24 Mar 2017 00:15:08 +0000 (00:15 +0000)]
searchview: add "t" id to link to thread overview

At least for the thread view (&x=t); this will make it
easy to link to the overview.

7 years agoextmsg: use updated mail-archive.com URL
Eric Wong [Wed, 22 Mar 2017 02:14:19 +0000 (02:14 +0000)]
extmsg: use updated mail-archive.com URL

Apparently mid.mail-archive.com does not support HTTPS,
and the HTTP version redirects to the search query, anyways.

7 years agoview: escape HTML description name
Eric Wong [Tue, 14 Mar 2017 21:23:39 +0000 (21:23 +0000)]
view: escape HTML description name

Otherwise funky filenames can cause HTML injection
vulnerabilities (hope you have JavaScript disabled!)

7 years agowww: do not unescape PATH_INFO twice
Eric Wong [Tue, 14 Feb 2017 22:45:15 +0000 (22:45 +0000)]
www: do not unescape PATH_INFO twice

PSGI specs already require PATH_INFO to be unescaped;
so our tests were wrong, too.

7 years agot/mime: quiet warnings for old versions of Email::Simple
Eric Wong [Sun, 12 Feb 2017 02:41:22 +0000 (02:41 +0000)]
t/mime: quiet warnings for old versions of Email::Simple

This is fixed in the newest versions of Email::Simple,
but not the version in Debian jessie (2.203)

7 years agohandle repeated References and In-Reply-To headers
Eric Wong [Sat, 11 Feb 2017 23:54:48 +0000 (23:54 +0000)]
handle repeated References and In-Reply-To headers

It seems possible for git-send-email(1) to generate repeated
repeated instances of References and In-Reply-To headers,
as evidenced in:

https://public-inbox.org/git/20161111124541.8216-17-vascomalmeida@sapo.pt/raw

This causes a mismatch between how our search indexer threads
and how our HTML view handles threading.  In the future, View.pm
will use the smsg-parsed {references} field and avoid redoing
Email::MIME header parsing.

We will still need to figure out a way to deal with messages
with repeated Message-IDs, at some point, too.

7 years agoconfig: do not slurp lines into memory
Eric Wong [Wed, 8 Feb 2017 21:41:38 +0000 (21:41 +0000)]
config: do not slurp lines into memory

There's no need to hold everything in memory, here,
since apparently "foreach" will read everything at
once in array context

(for some reason, I thought Perl5 was smart enough
 to avoid creating a temporary array, here...)

7 years agoTODO: several updates
Eric Wong [Tue, 7 Feb 2017 22:27:52 +0000 (22:27 +0000)]
TODO: several updates

Always plenty to do while working on this...

7 years agosearch: schema version bump for empty References/In-Reply-To
Eric Wong [Mon, 6 Feb 2017 21:39:45 +0000 (21:39 +0000)]
search: schema version bump for empty References/In-Reply-To

We cannot distinguish between legitimate ghosts and mis-threaded
messages before commit 83425ef12e4b65cdcecd11ddcb38175d4a91d5a0
("searchidx: deal with empty In-Reply-To and References headers")
so we must rebuild the index in parallel to fix it.

7 years agoRevert "searchidx: reindex clobbers old thread IDs"
Eric Wong [Mon, 6 Feb 2017 21:37:26 +0000 (21:37 +0000)]
Revert "searchidx: reindex clobbers old thread IDs"

Oops, that's broken, too.  I guess the only way to reindex
after fixing the thread detection is to start from scratch.

This reverts commit 5d91adedf5f33ef1cb87df2a86306ddf370b4f8d.

7 years agosearchidx: reindex clobbers old thread IDs
Eric Wong [Mon, 6 Feb 2017 21:08:13 +0000 (21:08 +0000)]
searchidx: reindex clobbers old thread IDs

We cannot always reuse thread IDs since our threading
logic may change as bugs are fixed.

7 years agosearchidx: deal with empty In-Reply-To and References headers
Eric Wong [Mon, 6 Feb 2017 19:54:25 +0000 (19:54 +0000)]
searchidx: deal with empty In-Reply-To and References headers

In some messages, these headers exist, but have empty values.
Do not let empty values throw off our search indexer to tie
threads together, as it can make non-sensical threads grouped
to a Message-Id of "" (empty string).

See
<https://public-inbox.org/git/11340844841342-git-send-email-mailing-lists.git@rawuncut.elitemail.org/raw>
for an example of such a message.

Thanks-to: Johannes Schindelin <Johannes.Schindelin@gmx.de>
  <https://public-inbox.org/git/alpine.DEB.2.20.1702041206130.3496@virtualbox/>

7 years agosearchview: increase limit for displaying search results
Eric Wong [Mon, 6 Feb 2017 02:38:37 +0000 (02:38 +0000)]
searchview: increase limit for displaying search results

We are in no danger of excessive buffering or OOM-ing,
the main page for every inbox already loads 200 results;
and thread page views even load 1000!  Increase this to
200 for now.

7 years agosearchview: clarify numeric summary at bottom
Eric Wong [Mon, 6 Feb 2017 02:07:24 +0000 (02:07 +0000)]
searchview: clarify numeric summary at bottom

Xapian can only give estimated results when a result limit is
given to it, so make clear it is an estimate to avoid showing
non-sensical ranges when no results are returned.

7 years agoadd filter for Subject: tags
Eric Wong [Thu, 26 Jan 2017 02:09:36 +0000 (02:09 +0000)]
add filter for Subject: tags

Some mailing lists add annoying tags into the Subject line which
discourages readers from doing proper mail organization on the
client side.  They also waste precious screen space and
attention span.

Remove them from our archives to reduce clutter.

7 years agowatchmaildir: allow arguments for filters
Eric Wong [Wed, 25 Jan 2017 21:39:06 +0000 (21:39 +0000)]
watchmaildir: allow arguments for filters

We'll want to allow some degree of configuration for
various mailing lists.

7 years agowatchmaildir: limit live importer processes
Eric Wong [Wed, 18 Jan 2017 19:13:09 +0000 (19:13 +0000)]
watchmaildir: limit live importer processes

We don't want to be triggering OOM or swapping on weaker
systems when we have dozens of inboxes as potential targets.

7 years agolearn: implement "rm" only functionality
Eric Wong [Thu, 19 Jan 2017 00:31:30 +0000 (00:31 +0000)]
learn: implement "rm" only functionality

Do not consider this interface stable, but I just needed a
way to remove mis-imported multipart messages so
public-inbox-watch could pick them up again from my Maildir.

7 years agomime: avoid SUPER usage in Email::MIME subclass
Eric Wong [Wed, 18 Jan 2017 23:50:57 +0000 (23:50 +0000)]
mime: avoid SUPER usage in Email::MIME subclass

We must call Email::Simple methods directly in our monkey patch
for Email::MIME to call the intended method.  Using SUPER in our
subclass would instead hit a different, unintended method in
Email::MIME.

Reported-by: Junio C Hamano <gitster@pobox.com>
<xmqq4m0wb43w.fsf@gitster.mtv.corp.google.com>

7 years agoinbox: reinstate periodic cleanup of Xapian and SQLite objects
Eric Wong [Wed, 11 Jan 2017 10:13:00 +0000 (10:13 +0000)]
inbox: reinstate periodic cleanup of Xapian and SQLite objects

We may need to do this even more aggressively, since the
Xapian database does not always give the latest results.
This time, we'll do it without relying on weak references,
and instead check refcounts.

7 years agointroduce PublicInbox::MIME wrapper class
Eric Wong [Tue, 10 Jan 2017 21:40:37 +0000 (21:40 +0000)]
introduce PublicInbox::MIME wrapper class

This should fix problems with multipart messages where
text/plain parts lack a header.

cf. git clone --mirror https://github.com/rjbs/Email-MIME.git
    refs/pull/28/head

In the future, we may still introduce as streaming
interface to reduce memory usage on large emails.

7 years agoinbox: properly register cleanup timer for git processes
Eric Wong [Sat, 7 Jan 2017 02:10:23 +0000 (02:10 +0000)]
inbox: properly register cleanup timer for git processes

We still need to cleanup git processes occasionally, since
"git cat-file --batch" does not release old packs (and
git processes are fairly expensive).

For SQLite and Xapian file handles, they should be capable
of managing themselves without too much trouble, so lets
try keeping them for the lifetime of a process.

7 years agosearch: remove subject_summary
Eric Wong [Sat, 7 Jan 2017 01:44:52 +0000 (01:44 +0000)]
search: remove subject_summary

Apparently it never actually got used, and the world seems
fine without it, so we can drop it.

While we're at it, consider removing our subject_path
usage from existence, too.  We are not using fancy subject-line
based URLs, here.

7 years agosearchmsg: favor direct hash access over accessor methods
Eric Wong [Sat, 7 Jan 2017 01:44:51 +0000 (01:44 +0000)]
searchmsg: favor direct hash access over accessor methods

This is faster, smaller, and more straighforward to me with
fewer layers of indirection.

7 years agoremove incorrect comment about strftime + locales
Eric Wong [Sat, 7 Jan 2017 01:44:50 +0000 (01:44 +0000)]
remove incorrect comment about strftime + locales

We only need strftime to be locale-independent when generating
dates for email and HTTP headers.  Purely numeric dates can
use strftime for ease-of-readability.

7 years agoconfig: allow per-inbox nntpserver
Eric Wong [Sat, 7 Jan 2017 01:44:49 +0000 (01:44 +0000)]
config: allow per-inbox nntpserver

This allows certain inboxes to override the global nntpserver
(perhaps under a different domain).

7 years agoinbox: eliminate weaken usage entirely
Eric Wong [Sat, 7 Jan 2017 01:44:48 +0000 (01:44 +0000)]
inbox: eliminate weaken usage entirely

We can do a better job initializing the data structure
so we no longer need to rely on weak references to cleanup
when we ditch the config on reload.

7 years agoinbox: describe the full key name
Eric Wong [Sat, 7 Jan 2017 01:44:47 +0000 (01:44 +0000)]
inbox: describe the full key name

Hopefully make this easier for future generations to understand.

7 years agoconfig: remove unused get() method
Eric Wong [Sat, 7 Jan 2017 01:44:46 +0000 (01:44 +0000)]
config: remove unused get() method

This seems like an unnecessary abstraction, or an abstraction
on the wrong level.

7 years agoconfig: always use namespaced "publicinboxlimiter"
Eric Wong [Sat, 7 Jan 2017 01:44:45 +0000 (01:44 +0000)]
config: always use namespaced "publicinboxlimiter"

I'm not sure if we'll ever support sharing a config file
with other tools, but maybe we will, and "limiter" is
too generic.

7 years agoqspawn: prepare to support runtime reloading of Limiter
Eric Wong [Sat, 7 Jan 2017 01:44:44 +0000 (01:44 +0000)]
qspawn: prepare to support runtime reloading of Limiter

We may allow the {max} value of a limiter to be changed
in the future, so lets start accounting for it before we
spawn followup processes.

7 years agohttp: remove weaken usage, reduce anonsub capture scope
Eric Wong [Wed, 4 Jan 2017 11:20:51 +0000 (11:20 +0000)]
http: remove weaken usage, reduce anonsub capture scope

Avoiding weaken here is no more dangerous than the existing
circular refs (e.g. psgix.io) we create and manage throughout
the lifetime of the connection.  So, trust ourselves to maintain
the data structure properly and avoid triggering extra memory
usage.

While we're at it, avoid having anonymous subroutines capture
more variables than necessary to simplify reference auditing.

7 years agohttpd/async: remove weaken usage
Eric Wong [Wed, 4 Jan 2017 11:20:50 +0000 (11:20 +0000)]
httpd/async: remove weaken usage

We do not need to use weaken() here, so avoid it to simplify our
interactions with Perl; as weaken requires additional storage
and (it seems) time complexity.

7 years agohttp: fix spelling error
Eric Wong [Wed, 4 Jan 2017 11:20:49 +0000 (11:20 +0000)]
http: fix spelling error

Oops.  And we'll be fixing circular references from now...

7 years agowatch: watchspam affects all configured inboxes
Eric Wong [Mon, 2 Jan 2017 13:16:15 +0000 (13:16 +0000)]
watch: watchspam affects all configured inboxes

If a message is spam in one mailbox, it is spam in all others a
particular user/group will care about.

7 years agodoc: minor updates to design notes
Eric Wong [Mon, 26 Dec 2016 21:41:15 +0000 (21:41 +0000)]
doc: minor updates to design notes

ssoma is not worth marketing, but perhaps our mirror of
the git mailing list archives is...

7 years agoevcleanup: ensure deferred close from timers are handled ASAP
Eric Wong [Mon, 26 Dec 2016 03:05:15 +0000 (03:05 +0000)]
evcleanup: ensure deferred close from timers are handled ASAP

Danga::Socket defers close() syscalls until the end of the event
loop to avoid FD recycling.  Unfortunately, this is dependent on
IO events firing and waking the process up from
poll/kevent/epoll_wait.

Without any I/O activity, a socket could remain in the
@Danga::Socket::ToClose array indefinitely.  Thus, we will
trigger a fake IO event after running all timers to trigger
the deferred close in Danga::Socket::PostEventLoop.

7 years agohttpd/async: improve variable naming
Eric Wong [Sun, 25 Dec 2016 08:09:48 +0000 (08:09 +0000)]
httpd/async: improve variable naming

We only refer to PublicInbox::HTTP objects here, so '$io'
was a bad name.

7 years agogithttpbackend: minor cleanups to improve readability
Eric Wong [Sun, 25 Dec 2016 07:33:02 +0000 (07:33 +0000)]
githttpbackend: minor cleanups to improve readability

Fewer returns improves readability and the diffstat agrees.

7 years agogithttpbackend: simplify compatibility code
Eric Wong [Sun, 25 Dec 2016 06:52:03 +0000 (06:52 +0000)]
githttpbackend: simplify compatibility code

Fewer conditionals means theres fewer code paths to test
and makes things easier-to-read.

7 years agogithttpbackend: minor readability improvement
Eric Wong [Sun, 25 Dec 2016 06:39:13 +0000 (06:39 +0000)]
githttpbackend: minor readability improvement

Use a more meaningful variable name for the Qspawn
object, since this module is the reference for its
use.

7 years agohttp: fix clobbering of $null_io
Eric Wong [Sun, 25 Dec 2016 09:40:25 +0000 (09:40 +0000)]
http: fix clobbering of $null_io

Oops, this would be disatrous if we started handling
bigger request bodies or slow clients.

Fixes: c008654229a9 ("avoid IO::File for anonymous temporary files")
7 years agolinkify: modify argument in place
Eric Wong [Sat, 24 Dec 2016 11:52:44 +0000 (11:52 +0000)]
linkify: modify argument in place

This results in over 1% speedup doing $MESSAGE_ID/T/ HTML
generation for a 368-message thread.

7 years agoview: do not modify array during iteration
Eric Wong [Sat, 24 Dec 2016 11:52:43 +0000 (11:52 +0000)]
view: do not modify array during iteration

This results in a half percent speedup or so doing
$MESSAGE_ID/T/ HTML generation for a 368 message thread.

7 years agoview: stop chomping off whitespace at ends of messages
Eric Wong [Sat, 24 Dec 2016 11:52:42 +0000 (11:52 +0000)]
view: stop chomping off whitespace at ends of messages

This allows a 3-4% speedup in $MESSAGE_ID/T/ page generation
speed for a 368+ message thread.  It also more faithfully
preserves the message as intended; even if the it makes the
sender look like a space-wasting slob :P

7 years agoview: remove unused parameter
Eric Wong [Sat, 24 Dec 2016 11:52:41 +0000 (11:52 +0000)]
view: remove unused parameter

And add a comment about it to remind our future selves.

7 years agosearch: lookup_mail handles modified DBs
Eric Wong [Thu, 22 Dec 2016 08:00:26 +0000 (08:00 +0000)]
search: lookup_mail handles modified DBs

We call lookup_mail all over the place, be sure we can handle
database modifications in those cases.

7 years agodoc: various comments on async handling
Eric Wong [Thu, 22 Dec 2016 07:29:17 +0000 (07:29 +0000)]
doc: various comments on async handling

Notes for future developers (myself included) since we
can't assume people can read my mind.

7 years agosearchthread: simplify API and remove needless OO
Eric Wong [Tue, 20 Dec 2016 23:42:36 +0000 (23:42 +0000)]
searchthread: simplify API and remove needless OO

This simplifies callers to prevent errors and avoids
needless object-orientation in favor of a single procedure
call to handle threading and ordering.

7 years agosearchthread: update comment about loop prevention
Eric Wong [Tue, 20 Dec 2016 23:42:35 +0000 (23:42 +0000)]
searchthread: update comment about loop prevention

It definitely is necessary to prevent looping with the
%seen hash.

7 years agosearchmsg: remove ensure_metadata
Eric Wong [Tue, 20 Dec 2016 03:03:57 +0000 (03:03 +0000)]
searchmsg: remove ensure_metadata

Instead, only preload the ->mid field for threading,
as we only need ->thread and ->path once in Search->get_thread
(but we will need the ->mid field repeatedly).

This more than doubles View->load_results performance on
according to thread-all on an inbox with over 300K messages.

7 years agotests: add thread-all testing for benchmarking
Eric Wong [Tue, 20 Dec 2016 03:03:56 +0000 (03:03 +0000)]
tests: add thread-all testing for benchmarking

I'll be using this to improve message threading performance.

7 years agosearchmsg: do not memoize {date} field
Eric Wong [Sat, 17 Dec 2016 12:04:11 +0000 (12:04 +0000)]
searchmsg: do not memoize {date} field

We only generate the ->date once in NNTP, so creating
the hash entry is a waste.

7 years agosearchmsg: remove locale-dependency for ->date
Eric Wong [Sat, 17 Dec 2016 12:04:10 +0000 (12:04 +0000)]
searchmsg: remove locale-dependency for ->date

strftime is locale-dependent, which can cause surprising
failures for some users.

7 years agot/config.t: fix feedmax default
Eric Wong [Sat, 17 Dec 2016 05:50:30 +0000 (05:50 +0000)]
t/config.t: fix feedmax default

Oops :x

7 years agowwwtext: link to RFC4685 (Atom Threading)
Eric Wong [Wed, 14 Dec 2016 21:00:13 +0000 (21:00 +0000)]
wwwtext: link to RFC4685 (Atom Threading)

This should give this feature some more visibility.

7 years agoatom: implement message threading per RFC 4685
Eric Wong [Tue, 13 Dec 2016 02:33:30 +0000 (02:33 +0000)]
atom: implement message threading per RFC 4685

This will allows certain feed readers to render a message thread
as described in <https://www.jwz.org/doc/threading.html>.

Feed readers with knowledge of of RFC 4685 are unknown to us at
this time, but perhaps this will encourage future implementations.

Existing feed readers I've tested (newsbeuter, feed2imap) seem
to ignore these tags gracefully without degradation.

7 years agofeed: support publicinbox.<name>.feedmax
Eric Wong [Sat, 17 Dec 2016 04:27:52 +0000 (04:27 +0000)]
feed: support publicinbox.<name>.feedmax

This allows users to customize by using smaller or larger Atom
feeds than the default value of 25 entries.

7 years agoTODO: note IO::KQueue for the ticket
Eric Wong [Wed, 14 Dec 2016 23:53:06 +0000 (23:53 +0000)]
TODO: note IO::KQueue for the ticket

Do not require users to have network access to know what
the link refers to.

7 years agot/thread-cycle: no need for Xapian to run this test
Eric Wong [Wed, 14 Dec 2016 19:28:53 +0000 (19:28 +0000)]
t/thread-cycle: no need for Xapian to run this test

We don't actually use anything from SearchMsg,
just the class name.

7 years agowwwtext: remove outdated comment
Eric Wong [Wed, 14 Dec 2016 20:58:00 +0000 (20:58 +0000)]
wwwtext: remove outdated comment

I originally envisioned wwwtext being more flexible and able to
serve arbitrary blobs; but at this point I consider it redundant
and public-inbox is not wiki software.