* [PATCH 1/5] xt/eml_check_limits: check limits against an inbox
2020-05-10 22:37 [PATCH 0/5] scattered dev/CLI-oriented changes Eric Wong
@ 2020-05-10 22:37 ` Eric Wong
2020-05-10 22:37 ` [PATCH 2/5] rename "ContentId" to "ContentHash" Eric Wong
` (3 subsequent siblings)
4 siblings, 0 replies; 8+ messages in thread
From: Eric Wong @ 2020-05-10 22:37 UTC (permalink / raw)
To: meta
This allows maintainers to easily check limits against the
contents of existing inboxes. This script covers most of
the new limits enforced by PublicInbox::Eml.
Usage is similar to most xt/*.t scripts:
GIANT_INBOX_DIR=/path/to/inbox prove -bvw xt/eml_check_limits.t
Setting `TEST_CLASS=PublicInbox::MIME' allows us to check
performance and memory use against the old subclass of
Email::MIME.
---
MANIFEST | 1 +
xt/eml_check_limits.t | 76 +++++++++++++++++++++++++++++++++++++++++++
2 files changed, 77 insertions(+)
create mode 100644 xt/eml_check_limits.t
diff --git a/MANIFEST b/MANIFEST
index 9c804a0780e..b1512c7a919 100644
--- a/MANIFEST
+++ b/MANIFEST
@@ -333,6 +333,7 @@ t/x-unknown-alpine.eml
t/xcpdb-reshard.t
xt/cmp-msgstr.t
xt/cmp-msgview.t
+xt/eml_check_limits.t
xt/git-http-backend.t
xt/git_async_cmp.t
xt/mem-msgview.t
diff --git a/xt/eml_check_limits.t b/xt/eml_check_limits.t
new file mode 100644
index 00000000000..39de047645b
--- /dev/null
+++ b/xt/eml_check_limits.t
@@ -0,0 +1,76 @@
+#!perl -w
+# Copyright (C) 2020 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
+use Test::More;
+use PublicInbox::TestCommon;
+use PublicInbox::Eml;
+use PublicInbox::Inbox;
+use List::Util qw(max);
+use Benchmark qw(:all :hireswallclock);
+use PublicInbox::Spawn qw(popen_rd);
+use Carp ();
+require_git(2.19); # for --unordered
+require_mods(qw(BSD::Resource));
+BSD::Resource->import(qw(getrusage));
+my $cls = $ENV{TEST_CLASS};
+require_mods($cls) if $cls;
+$cls //= 'PublicInbox::Eml';
+my $inboxdir = $ENV{GIANT_INBOX_DIR};
+plan skip_all => "GIANT_INBOX_DIR not defined for $0" unless $inboxdir;
+local $PublicInbox::Eml::mime_nesting_limit = 0x7fffffff;
+local $PublicInbox::Eml::mime_parts_limit = 0x7fffffff;
+local $PublicInbox::Eml::header_size_limit = 0x7fffffff;
+my $ibx = PublicInbox::Inbox->new({ inboxdir => $inboxdir, name => 'x' });
+my $git = $ibx->git;
+my @cat = qw(cat-file --buffer --batch-check --batch-all-objects --unordered);
+my $fh = $git->popen(@cat);
+my ($m, $n);
+my $max_nest = [ 0, '' ]; # [ bytes, blob oid ]
+my $max_idx = [ 0, '' ];
+my $max_parts = [ 0, '' ];
+my $max_size = [ 0, '' ];
+my $max_hdr = [ 0, '' ];
+my $info = [ 0, '' ];
+my $each_part_cb = sub {
+ my ($p) = @_;
+ my ($part, $depth, $idx) = @$p;
+ $max_nest = [ $depth, $info->[1] ] if $depth > $max_nest->[0];
+ my $max = max(split(/\./, $idx));
+ $max_idx = [ $max, $info->[1] ] if $max > $max_idx->[0];
+ ++$info->[0];
+};
+
+my ($bref, $oid, $size);
+local $SIG{__WARN__} = sub { diag "$inboxdir $oid ", @_ };
+my $cat_cb = sub {
+ ($bref, $oid, undef, $size) = @_;
+ ++$m;
+ $info = [ 0, $oid ];
+ my $eml = $cls->new($bref);
+ my $hdr_len = length($eml->header_obj->as_string);
+ $max_hdr = [ $hdr_len, $oid ] if $hdr_len > $max_hdr->[0];
+ $eml->each_part($each_part_cb, $info, 1);
+ $max_parts = $info if $info->[0] > $max_parts->[0];
+ $max_size = [ $size, $oid ] if $size > $max_size->[0];
+};
+
+my $t = timeit(1, sub {
+ $git->cat_async_begin;
+ my ($blob, $type);
+ while (<$fh>) {
+ ($blob, $type) = split / /;
+ next if $type ne 'blob';
+ ++$n;
+ $git->cat_async($blob, $cat_cb);
+ }
+ $git->cat_async_wait;
+});
+is($m, $n, 'scanned all messages');
+diag "$$ $inboxdir took ".timestr($t)." for $n <=> $m messages";
+diag "$$ max_nest $max_nest->[0] @ $max_nest->[1]";
+diag "$$ max_idx $max_idx->[0] @ $max_idx->[1]";
+diag "$$ max_parts $max_parts->[0] @ $max_parts->[1]";
+diag "$$ max_size $max_size->[0] @ $max_size->[1]";
+diag "$$ max_hdr $max_hdr->[0] @ $max_hdr->[1]";
+diag "$$ RSS ".getrusage()->maxrss. ' k';
+done_testing;
^ permalink raw reply related [flat|nested] 8+ messages in thread* [PATCH 2/5] rename "ContentId" to "ContentHash"
2020-05-10 22:37 [PATCH 0/5] scattered dev/CLI-oriented changes Eric Wong
2020-05-10 22:37 ` [PATCH 1/5] xt/eml_check_limits: check limits against an inbox Eric Wong
@ 2020-05-10 22:37 ` Eric Wong
2020-05-10 22:37 ` [PATCH 3/5] overidx: document the SQLite PRAGMA we use Eric Wong
` (2 subsequent siblings)
4 siblings, 0 replies; 8+ messages in thread
From: Eric Wong @ 2020-05-10 22:37 UTC (permalink / raw)
To: meta
The old name may be confused with "Content-ID" as described in
RFC 2392, so use an alternate name to avoid confusing future
readers.
---
Documentation/public-inbox-v2-format.pod | 12 ++---
MANIFEST | 4 +-
.../{ContentId.pm => ContentHash.pm} | 8 ++--
lib/PublicInbox/Import.pm | 2 +-
lib/PublicInbox/V2Writable.pm | 48 +++++++++----------
script/public-inbox-edit | 16 +++----
t/{content_id.t => content_hash.t} | 14 +++---
t/v1reindex.t | 2 +-
t/v2reindex.t | 2 +-
t/v2writable.t | 4 +-
10 files changed, 56 insertions(+), 56 deletions(-)
rename lib/PublicInbox/{ContentId.pm => ContentHash.pm} (93%)
rename t/{content_id.t => content_hash.t} (64%)
diff --git a/Documentation/public-inbox-v2-format.pod b/Documentation/public-inbox-v2-format.pod
index d87a717d40b..9e284a75431 100644
--- a/Documentation/public-inbox-v2-format.pod
+++ b/Documentation/public-inbox-v2-format.pod
@@ -159,7 +159,7 @@ top-level of the directory.
=head1 OBJECT IDENTIFIERS
-There are three distinct type of identifiers. content_id is the
+There are three distinct type of identifiers. content_hash is the
new one for v2 and should make message removal and deduplication
easier. object_id and Message-ID are already known.
@@ -179,11 +179,11 @@ The email header; duplicates allowed for archival purposes.
This remains a searchable field in Xapian. Note: it's possible
for emails to have multiple Message-ID headers (and L<git-send-email(1)>
had that bug for a bit); so we take all of them into account.
-In case of conflicts detected by content_id below, we generate a new
-Message-ID based on content_id; if the generated Message-ID still
+In case of conflicts detected by content_hash below, we generate a new
+Message-ID based on content_hash; if the generated Message-ID still
conflicts, a random one is generated.
-=item content_id
+=item content_hash
A hash of relevant headers and raw body content for
purging of unwanted content. This is not stored anywhere,
@@ -193,7 +193,7 @@ For now, the relevant headers are:
Subject, From, Date, References, In-Reply-To, To, Cc
-Received, List-Id, and similar headers are NOT part of content_id as
+Received, List-Id, and similar headers are NOT part of content_hash as
they differ across lists and we will want removal to be able to cross
lists.
@@ -203,7 +203,7 @@ raw body risks being broken by list signatures; but we can use
filters (e.g. PublicInbox::Filter::Vger) to clean the body for
imports.
-content_id is SHA-256 for now; but can be changed at any time
+content_hash is SHA-256 for now; but can be changed at any time
without making DB changes.
=back
diff --git a/MANIFEST b/MANIFEST
index b1512c7a919..7997bc9906c 100644
--- a/MANIFEST
+++ b/MANIFEST
@@ -99,7 +99,7 @@ lib/PublicInbox/AdminEdit.pm
lib/PublicInbox/AltId.pm
lib/PublicInbox/Cgit.pm
lib/PublicInbox/Config.pm
-lib/PublicInbox/ContentId.pm
+lib/PublicInbox/ContentHash.pm
lib/PublicInbox/DS.pm
lib/PublicInbox/DSKQXS.pm
lib/PublicInbox/DSPoll.pm
@@ -223,7 +223,7 @@ t/cgi.t
t/check-www-inbox.perl
t/config.t
t/config_limiter.t
-t/content_id.t
+t/content_hash.t
t/convert-compact.t
t/data/0001.patch
t/ds-kqxs.t
diff --git a/lib/PublicInbox/ContentId.pm b/lib/PublicInbox/ContentHash.pm
similarity index 93%
rename from lib/PublicInbox/ContentId.pm
rename to lib/PublicInbox/ContentHash.pm
index 8d77934f20a..420dc5e7c92 100644
--- a/lib/PublicInbox/ContentId.pm
+++ b/lib/PublicInbox/ContentHash.pm
@@ -6,11 +6,11 @@
# This is not stored in any database anywhere and may change
# as changes in duplicate detection are needed.
# See L<public-inbox-v2-format(5)> manpage for more details.
-package PublicInbox::ContentId;
+package PublicInbox::ContentHash;
use strict;
use warnings;
use base qw/Exporter/;
-our @EXPORT_OK = qw/content_id content_digest/;
+our @EXPORT_OK = qw/content_hash content_digest/;
use PublicInbox::MID qw(mids references);
use PublicInbox::MsgIter;
@@ -60,7 +60,7 @@ sub content_digest ($) {
# References: and In-Reply-To: get used interchangeably
# in some "duplicates" in LKML. We treat them the same
# in SearchIdx, so treat them the same for this:
- # do NOT consider the Message-ID as part of the content_id
+ # do NOT consider the Message-ID as part of the content_hash
# if we got here, we've already got Message-ID reuse
my %seen = map { $_ => 1 } @{mids($hdr)};
foreach my $mid (@{references($hdr)}) {
@@ -92,7 +92,7 @@ sub content_digest ($) {
$dig;
}
-sub content_id ($) {
+sub content_hash ($) {
content_digest($_[0])->digest;
}
diff --git a/lib/PublicInbox/Import.pm b/lib/PublicInbox/Import.pm
index 07d18599200..fc61d06207c 100644
--- a/lib/PublicInbox/Import.pm
+++ b/lib/PublicInbox/Import.pm
@@ -13,7 +13,7 @@ use PublicInbox::Spawn qw(spawn popen_rd);
use PublicInbox::MID qw(mids mid2path);
use PublicInbox::Address;
use PublicInbox::MsgTime qw(msg_timestamp msg_datestamp);
-use PublicInbox::ContentId qw(content_digest);
+use PublicInbox::ContentHash qw(content_digest);
use PublicInbox::MDA;
use PublicInbox::Eml;
use POSIX qw(strftime);
diff --git a/lib/PublicInbox/V2Writable.pm b/lib/PublicInbox/V2Writable.pm
index f599e0a03d8..bf5a0df947a 100644
--- a/lib/PublicInbox/V2Writable.pm
+++ b/lib/PublicInbox/V2Writable.pm
@@ -13,7 +13,7 @@ use PublicInbox::Eml;
use PublicInbox::Git;
use PublicInbox::Import;
use PublicInbox::MID qw(mids references);
-use PublicInbox::ContentId qw(content_id content_digest);
+use PublicInbox::ContentHash qw(content_hash content_digest);
use PublicInbox::Inbox;
use PublicInbox::OverIdx;
use PublicInbox::Msgmap;
@@ -353,23 +353,23 @@ sub _replace_oids ($$$) {
$rewrites;
}
-sub content_ids ($) {
+sub content_hashes ($) {
my ($mime) = @_;
- my @cids = ( content_id($mime) );
+ my @chashes = ( content_hash($mime) );
# We still support Email::MIME, here, and
# Email::MIME->as_string doesn't always round-trip, so we may
- # use a second content_id
- my $rt = content_id(PublicInbox::Eml->new(\($mime->as_string)));
- push @cids, $rt if $cids[0] ne $rt;
- \@cids;
+ # use a second content_hash
+ my $rt = content_hash(PublicInbox::Eml->new(\($mime->as_string)));
+ push @chashes, $rt if $chashes[0] ne $rt;
+ \@chashes;
}
sub content_matches ($$) {
- my ($cids, $existing) = @_;
- my $cid = content_id($existing);
- foreach (@$cids) {
- return 1 if $_ eq $cid
+ my ($chashes, $existing) = @_;
+ my $chash = content_hash($existing);
+ foreach (@$chashes) {
+ return 1 if $_ eq $chash
}
0
}
@@ -386,13 +386,13 @@ sub rewrite_internal ($$;$$$) {
$im = $self->importer;
}
my $over = $self->{over};
- my $cids = content_ids($old_mime);
+ my $chashes = content_hashes($old_mime);
my @removed;
my $mids = mids($old_mime->header_obj);
# We avoid introducing new blobs into git since the raw content
# can be slightly different, so we do not need the user-supplied
- # message now that we have the mids and content_id
+ # message now that we have the mids and content_hash
$old_mime = undef;
my $mark;
@@ -407,7 +407,7 @@ sub rewrite_internal ($$;$$$) {
}
my $orig = $$msg;
my $cur = PublicInbox::Eml->new($msg);
- if (content_matches($cids, $cur)) {
+ if (content_matches($chashes, $cur)) {
$gone{$smsg->{num}} = [ $smsg, $cur, \$orig ];
}
}
@@ -835,7 +835,7 @@ sub get_blob ($$) {
sub content_exists ($$$) {
my ($self, $mime, $mid) = @_;
my $over = $self->{over};
- my $cids = content_ids($mime);
+ my $chashes = content_hashes($mime);
my ($id, $prev);
while (my $smsg = $over->next_by_mid($mid, \$id, \$prev)) {
my $msg = get_blob($self, $smsg);
@@ -844,7 +844,7 @@ sub content_exists ($$$) {
next;
}
my $cur = PublicInbox::Eml->new($msg);
- return 1 if content_matches($cids, $cur);
+ return 1 if content_matches($chashes, $cur);
# XXX DEBUG_DIFF is experimental and may be removed
diff($mid, $cur, $mime) if $ENV{DEBUG_DIFF};
@@ -873,9 +873,9 @@ sub mark_deleted ($$$$) {
my $msgref = $git->cat_file($oid);
my $mime = PublicInbox::Eml->new($$msgref);
my $mids = mids($mime->header_obj);
- my $cid = content_id($mime);
+ my $chash = content_hash($mime);
foreach my $mid (@$mids) {
- $sync->{D}->{"$mid\0$cid"} = $oid;
+ $sync->{D}->{"$mid\0$chash"} = $oid;
}
}
@@ -904,11 +904,11 @@ sub reindex_oid_m ($$$$;$) {
my $msgref = $git->cat_file($oid, \$len);
my $mime = PublicInbox::Eml->new($$msgref);
my $mids = mids($mime->header_obj);
- my $cid = content_id($mime);
+ my $chash = content_hash($mime);
die "BUG: reindex_oid_m called for <=1 mids" if scalar(@$mids) <= 1;
for my $mid (reverse @$mids) {
- delete($sync->{D}->{"$mid\0$cid"}) and
+ delete($sync->{D}->{"$mid\0$chash"}) and
die "BUG: reindex_oid should handle <$mid> delete";
}
my $over = $self->{over};
@@ -1002,7 +1002,7 @@ sub reindex_oid ($$$$) {
return if $len == 0; # purged
my $mime = PublicInbox::Eml->new($$msgref);
my $mids = mids($mime->header_obj);
- my $cid = content_id($mime);
+ my $chash = content_hash($mime);
if (scalar(@$mids) == 0) {
warn "E: $oid has no Message-ID, skipping\n";
@@ -1011,7 +1011,7 @@ sub reindex_oid ($$$$) {
my $mid = $mids->[0];
# was the file previously marked as deleted?, skip if so
- if (delete($sync->{D}->{"$mid\0$cid"})) {
+ if (delete($sync->{D}->{"$mid\0$chash"})) {
if (!$sync->{reindex}) {
$num = $sync->{regen}--;
$self->{mm}->num_highwater($num);
@@ -1036,7 +1036,7 @@ sub reindex_oid ($$$$) {
} else { # multiple MIDs are a weird case:
my $del = 0;
for (@$mids) {
- $del += delete($sync->{D}->{"$_\0$cid"}) // 0;
+ $del += delete($sync->{D}->{"$_\0$chash"}) // 0;
}
if ($del) {
unindex_oid_remote($self, $oid, $_) for @$mids;
@@ -1309,7 +1309,7 @@ sub index_sync {
return unless defined $latest;
$self->idx_init($opt); # acquire lock
my $sync = {
- D => {}, # "$mid\0$cid" => $oid
+ D => {}, # "$mid\0$chash" => $oid
unindex_range => {}, # EPOCH => oid_old..oid_new
reindex => $opt->{reindex},
-opt => $opt
diff --git a/script/public-inbox-edit b/script/public-inbox-edit
index e895a228386..d8e511b2ee4 100755
--- a/script/public-inbox-edit
+++ b/script/public-inbox-edit
@@ -9,7 +9,7 @@ use warnings;
use Getopt::Long qw(:config gnu_getopt no_ignore_case auto_abbrev);
use PublicInbox::AdminEdit;
use File::Temp 0.19 (); # 0.19 for TMPDIR
-use PublicInbox::ContentId qw(content_id);
+use PublicInbox::ContentHash qw(content_hash);
use PublicInbox::MID qw(mid_clean mids);
PublicInbox::Admin::check_require('-index');
use PublicInbox::Eml;
@@ -43,7 +43,7 @@ if (defined $mid && defined $file) {
my @ibxs = PublicInbox::Admin::resolve_inboxes(\@ARGV, $opt, $cfg);
PublicInbox::AdminEdit::check_editable(\@ibxs);
-my $found = {}; # cid => [ [ibx, smsg] [, [ibx, smsg] ] ]
+my $found = {}; # chash => [ [ibx, smsg] [, [ibx, smsg] ] ]
sub find_mid ($$$) {
my ($found, $mid, $ibxs) = @_;
@@ -53,9 +53,9 @@ sub find_mid ($$$) {
while (my $smsg = $over->next_by_mid($mid, \$id, \$prev)) {
my $ref = $ibx->msg_by_smsg($smsg);
my $mime = PublicInbox::Eml->new($ref);
- my $cid = content_id($mime);
+ my $chash = content_hash($mime);
my $tuple = [ $ibx, $smsg ];
- push @{$found->{$cid} ||= []}, $tuple
+ push @{$found->{$chash} ||= []}, $tuple
}
PublicInbox::InboxWritable::cleanup($ibx);
}
@@ -96,8 +96,8 @@ Multiple messages with different content found matching
die "open($file) failed: $!";
my $mids = mids($mime->header_obj);
find_mid($found, $_, \@ibxs) for (@$mids); # populates $found
- my $cid = content_id($mime);
- my $to_edit = $found->{$cid};
+ my $chash = content_hash($mime);
+ my $to_edit = $found->{$chash};
unless ($to_edit) {
my $nr = scalar(keys %$found);
if ($nr > 0) {
@@ -115,7 +115,7 @@ $mids
}
exit 1;
}
- $found = { $cid => $to_edit };
+ $found = { $chash => $to_edit };
}
my %tmpopt = (
@@ -218,7 +218,7 @@ W: possible message boundary splitting error
my $nhdr = $new_mime->header_obj;
my $ohdr = $old_mime->header_obj;
if (($nhdr->as_string eq $ohdr->as_string) &&
- (content_id($new_mime) eq content_id($old_mime))) {
+ (content_hash($new_mime) eq content_hash($old_mime))) {
warn "No change detected to:\n", show_cmd($ibx, $smsg);
next unless $opt->{verbose};
diff --git a/t/content_id.t b/t/content_hash.t
similarity index 64%
rename from t/content_id.t
rename to t/content_hash.t
index 9df81aa8293..646aab07c9a 100644
--- a/t/content_id.t
+++ b/t/content_hash.t
@@ -3,7 +3,7 @@
use strict;
use warnings;
use Test::More;
-use PublicInbox::ContentId qw(content_id);
+use PublicInbox::ContentHash qw(content_hash);
use PublicInbox::Eml;
my $mime = PublicInbox::Eml->new(<<'EOF');
@@ -16,17 +16,17 @@ Date: Fri, 02 Oct 1993 00:00:00 +0000
hello world
EOF
-my $orig = content_id($mime);
-my $reload = content_id(PublicInbox::Eml->new($mime->as_string));
-is($orig, $reload, 'content_id matches after serialization');
+my $orig = content_hash($mime);
+my $reload = content_hash(PublicInbox::Eml->new($mime->as_string));
+is($orig, $reload, 'content_hash matches after serialization');
foreach my $h (qw(From To Cc)) {
my $n = q("Quoted N'Ame" <foo@EXAMPLE.com>);
$mime->header_set($h, "$n");
- my $q = content_id($mime);
- is($mime->header($h), $n, "content_id does not mutate $h:");
+ my $q = content_hash($mime);
+ is($mime->header($h), $n, "content_hash does not mutate $h:");
$mime->header_set($h, 'Quoted N\'Ame <foo@example.com>');
- my $nq = content_id($mime);
+ my $nq = content_hash($mime);
is($nq, $q, "quotes ignored in $h:");
}
diff --git a/t/v1reindex.t b/t/v1reindex.t
index 13605f8bd6c..9f23ef01e56 100644
--- a/t/v1reindex.t
+++ b/t/v1reindex.t
@@ -3,7 +3,7 @@
use strict;
use warnings;
use Test::More;
-use PublicInbox::ContentId qw(content_digest);
+use PublicInbox::ContentHash qw(content_digest);
use File::Path qw(remove_tree);
use PublicInbox::TestCommon;
use PublicInbox::Eml;
diff --git a/t/v2reindex.t b/t/v2reindex.t
index f16a0b0d81c..b99106d0fe7 100644
--- a/t/v2reindex.t
+++ b/t/v2reindex.t
@@ -4,7 +4,7 @@ use strict;
use warnings;
use Test::More;
use PublicInbox::Eml;
-use PublicInbox::ContentId qw(content_digest);
+use PublicInbox::ContentHash qw(content_digest);
use File::Path qw(remove_tree);
use PublicInbox::TestCommon;
require_git(2.6);
diff --git a/t/v2writable.t b/t/v2writable.t
index e5a565cea23..fa5c786e151 100644
--- a/t/v2writable.t
+++ b/t/v2writable.t
@@ -4,7 +4,7 @@ use strict;
use warnings;
use Test::More;
use PublicInbox::Eml;
-use PublicInbox::ContentId qw(content_digest content_id);
+use PublicInbox::ContentHash qw(content_digest content_hash);
use PublicInbox::TestCommon;
use Cwd qw(abs_path);
require_git(2.6);
@@ -215,7 +215,7 @@ EOF
$im = PublicInbox::V2Writable->new($ibx, {nproc => 2});
is($im->{shards}, 1, 'detected single shard from previous');
my ($mark, $rm_mime, $smsg) = $im->remove($mime, 'test removal');
- is(content_id($rm_mime), content_id($mime),
+ is(content_hash($rm_mime), content_hash($mime),
'removed object returned matches');
ok(defined($mark), 'mark set');
$im->done;
^ permalink raw reply related [flat|nested] 8+ messages in thread