From: Eric Wong <e@80x24.org>
To: meta@public-inbox.org
Subject: [PATCH 2/7] lei q: support IMAP/IMAPS --output destinations
Date: Sun, 21 Feb 2021 07:41:29 +0000 [thread overview]
Message-ID: <20210221074134.15084-3-e@80x24.org> (raw)
In-Reply-To: <20210221074134.15084-1-e@80x24.org>
Augment (and dedupe) aren't parallel, yet, so its more sensitive to
high-latency networks.
---
lib/PublicInbox/LeiAuth.pm | 2 +-
lib/PublicInbox/LeiOverview.pm | 7 +-
lib/PublicInbox/LeiQuery.pm | 18 ++++-
lib/PublicInbox/LeiToMail.pm | 56 +++++++++++++++-
lib/PublicInbox/NetReader.pm | 7 +-
lib/PublicInbox/NetWriter.pm | 12 ++++
xt/net_writer-imap.t | 118 ++++++++++++++++++++++++++++++---
7 files changed, 202 insertions(+), 18 deletions(-)
diff --git a/lib/PublicInbox/LeiAuth.pm b/lib/PublicInbox/LeiAuth.pm
index 7acb9900..bf0110ed 100644
--- a/lib/PublicInbox/LeiAuth.pm
+++ b/lib/PublicInbox/LeiAuth.pm
@@ -63,7 +63,7 @@ sub ipc_atfork_child {
}
sub new {
- my ($cls, $nrd) = @_;
+ my ($cls, $nrd) = @_; # nrd may be NetReader or descendant (NetWriter)
bless { nrd => $nrd }, $cls;
}
diff --git a/lib/PublicInbox/LeiOverview.pm b/lib/PublicInbox/LeiOverview.pm
index 3169bae6..4db1d8c8 100644
--- a/lib/PublicInbox/LeiOverview.pm
+++ b/lib/PublicInbox/LeiOverview.pm
@@ -59,7 +59,12 @@ sub new {
my $fmt = $opt->{$ofmt_key};
$fmt = lc($fmt) if defined $fmt;
- if ($dst =~ s/\A([a-z0-9]+)://is) { # e.g. Maildir:/home/user/Mail/
+ if ($dst =~ m!\A([a-z0-9\+]+)://!is) {
+ defined($fmt) and return $lei->fail(<<"");
+--$ofmt_key=$fmt invalid with URL $dst
+
+ $fmt = lc $1;
+ } elsif ($dst =~ s/\A([a-z0-9]+)://is) { # e.g. Maildir:/home/user/Mail/
my $ofmt = lc $1;
$fmt //= $ofmt;
return $lei->fail(<<"") if $fmt ne $ofmt;
diff --git a/lib/PublicInbox/LeiQuery.pm b/lib/PublicInbox/LeiQuery.pm
index f71beae6..eaf91f2e 100644
--- a/lib/PublicInbox/LeiQuery.pm
+++ b/lib/PublicInbox/LeiQuery.pm
@@ -11,14 +11,26 @@ sub prep_ext { # externals_each callback
$lxs->prepare_external($loc) unless $exclude->{$loc};
}
-sub qstr_add { # for --stdin
+sub _start_query {
+ my ($self) = @_;
+ if (my $nwr = $self->{nwr}) {
+ require PublicInbox::LeiAuth;
+ my $auth = $self->{auth} = PublicInbox::LeiAuth->new($nwr);
+ my $lxs = $self->{lxs};
+ $auth->auth_start($self, $lxs->can('do_query'), $lxs, $self);
+ } else {
+ $self->{lxs}->do_query($self);
+ }
+}
+
+sub qstr_add { # PublicInbox::InputPipe::consume callback for --stdin
my ($self) = @_; # $_[1] = $rbuf
if (defined($_[1])) {
$_[1] eq '' and return eval {
my $lse = delete $self->{lse};
$lse->query_approxidate($lse->git,
$self->{mset_opt}->{qstr});
- $self->{lxs}->do_query($self);
+ _start_query($self);
};
$self->{mset_opt}->{qstr} .= $_[1];
} else {
@@ -115,7 +127,7 @@ no query allowed on command-line with --stdin
return;
}
$mset_opt{qstr} = $lse->query_argv_to_string($lse->git, \@argv);
- $lxs->do_query($self);
+ _start_query($self);
}
# shell completion helper called by lei__complete
diff --git a/lib/PublicInbox/LeiToMail.pm b/lib/PublicInbox/LeiToMail.pm
index e89cca71..0e0b0a43 100644
--- a/lib/PublicInbox/LeiToMail.pm
+++ b/lib/PublicInbox/LeiToMail.pm
@@ -331,9 +331,31 @@ sub _maildir_write_cb ($$) {
}
}
+sub _imap_write_cb ($$) {
+ my ($self, $lei) = @_;
+ my $dedupe = $lei->{dedupe};
+ $dedupe->prepare_dedupe if $dedupe;
+ my $imap_append = $lei->{nwr}->can('imap_append');
+ my $mic = $lei->{nwr}->mic_get($lei->{ovv}->{dst});
+ my $folder = $self->{uri}->mailbox;
+ sub { # for git_to_mail
+ my ($bref, $smsg, $eml) = @_;
+ $mic // return $lei->fail; # dst may be undef-ed in last run
+ if ($dedupe) {
+ $eml //= PublicInbox::Eml->new($$bref); # copy bref
+ return if $dedupe->is_dup($eml, $smsg->{blob});
+ }
+ eval { $imap_append->($mic, $folder, $bref, $smsg, $eml) };
+ if (my $err = $@) {
+ undef $mic;
+ die $err;
+ }
+ }
+}
+
sub write_cb { # returns a callback for git_to_mail
my ($self, $lei) = @_;
- # _mbox_write_cb or _maildir_write_cb
+ # _mbox_write_cb, _maildir_write_cb or _imap_write_cb
my $m = "_$self->{base_type}_write_cb";
$self->$m($lei);
}
@@ -360,6 +382,18 @@ sub new {
"$dst exists and is not a writable file\n";
$self->can("eml2$fmt") or die "bad mbox format: $fmt\n";
$self->{base_type} = 'mbox';
+ } elsif ($fmt =~ /\Aimaps?\z/) { # TODO .onion support
+ require PublicInbox::NetWriter;
+ my $nwr = PublicInbox::NetWriter->new;
+ $nwr->add_url($dst);
+ $nwr->{quiet} = $lei->{opt}->{quiet};
+ my $err = $nwr->errors($dst);
+ return $lei->fail($err) if $err;
+ require PublicInbox::URIimap; # TODO: URI cast early
+ $self->{uri} = PublicInbox::URIimap->new($dst);
+ $self->{uri}->mailbox or die "No mailbox: $dst";
+ $lei->{nwr} = $nwr;
+ $self->{base_type} = 'imap';
} else {
die "bad mail --format=$fmt\n";
}
@@ -394,6 +428,26 @@ sub _do_augment_maildir {
}
}
+sub _augment_imap { # PublicInbox::NetReader::imap_each cb
+ my ($url, $uid, $kw, $eml, $lei) = @_;
+ _augment($eml, $lei);
+}
+
+sub _do_augment_imap {
+ my ($self, $lei) = @_;
+ my $dst = $lei->{ovv}->{dst};
+ my $nwr = $lei->{nwr};
+ if ($lei->{opt}->{augment}) {
+ my $dedupe = $lei->{dedupe};
+ if ($dedupe && $dedupe->prepare_dedupe) {
+ $nwr->imap_each($dst, \&_augment_imap, $lei);
+ $dedupe->pause_dedupe;
+ }
+ } else { # clobber existing IMAP folder
+ $nwr->imap_delete_all($dst);
+ }
+}
+
sub _pre_augment_mbox {
my ($self, $lei) = @_;
my $dst = $lei->{ovv}->{dst};
diff --git a/lib/PublicInbox/NetReader.pm b/lib/PublicInbox/NetReader.pm
index 92d004bc..541094a0 100644
--- a/lib/PublicInbox/NetReader.pm
+++ b/lib/PublicInbox/NetReader.pm
@@ -422,8 +422,13 @@ sub _imap_fetch_all ($$$) {
# uses cached auth info prepared by mic_for
sub mic_get {
my ($self, $sec) = @_;
- my $mic_arg = $self->{mic_arg}->{$sec} or
+ my $mic_arg = $self->{mic_arg}->{$sec};
+ unless ($mic_arg) {
+ my $uri = PublicInbox::URIimap->new($sec);
+ $sec = uri_section($uri);
+ $mic_arg = $self->{mic_arg}->{$sec} or
die "BUG: no Mail::IMAPClient->new arg for $sec";
+ }
if (defined(my $cb_name = $mic_arg->{Authcallback})) {
if (ref($cb_name) ne 'CODE') {
$mic_arg->{Authcallback} = $self->can($cb_name);
diff --git a/lib/PublicInbox/NetWriter.pm b/lib/PublicInbox/NetWriter.pm
index 6f0a0b94..89f8662e 100644
--- a/lib/PublicInbox/NetWriter.pm
+++ b/lib/PublicInbox/NetWriter.pm
@@ -23,4 +23,16 @@ sub imap_append {
die "APPEND $folder: $@";
}
+sub imap_delete_all {
+ my ($self, $url) = @_;
+ my $uri = PublicInbox::URIimap->new($url);
+ my $sec = $self->can('uri_section')->($uri);
+ local $0 = $uri->mailbox." $sec";
+ my $mic = $self->mic_get($sec) or die "E: not connected: $@";
+ $mic->select($uri->mailbox) or return; # non-existent
+ if ($mic->delete_message('1:*')) {
+ $mic->expunge;
+ }
+}
+
1;
diff --git a/xt/net_writer-imap.t b/xt/net_writer-imap.t
index dfd765be..4832245a 100644
--- a/xt/net_writer-imap.t
+++ b/xt/net_writer-imap.t
@@ -7,6 +7,7 @@ use POSIX qw(strftime);
use PublicInbox::OnDestroy;
use PublicInbox::URIimap;
use PublicInbox::Config;
+use Fcntl qw(O_EXCL O_WRONLY O_CREAT);
my $imap_url = $ENV{TEST_IMAP_WRITE_URL} or
plan skip_all => 'TEST_IMAP_WRITE_URL unset';
my $uri = PublicInbox::URIimap->new($imap_url);
@@ -19,30 +20,125 @@ my ($base) = ($0 =~ m!\b([^/]+)\.[^\.]+\z!);
my $folder = "INBOX.$base-$host-".strftime('%Y%m%d%H%M%S', gmtime(time)).
"-$$-".sprintf('%x', int(rand(0xffffffff)));
my $nwr = PublicInbox::NetWriter->new;
-$imap_url .= '/' unless substr($imap_url, -1) eq '/';
+chop($imap_url) if substr($imap_url, -1) eq '/';
my $folder_uri = PublicInbox::URIimap->new("$imap_url/$folder");
is($folder_uri->mailbox, $folder, 'folder correct') or
BAIL_OUT "BUG: bad $$uri";
$nwr->add_url($$folder_uri);
is($nwr->errors, undef, 'no errors');
$nwr->{pi_cfg} = bless {}, 'PublicInbox::Config';
-my $mics = $nwr->imap_common_init;
+
+my $set_cred_helper = sub {
+ my ($f, $cred_set) = @_;
+ sysopen(my $fh, $f, O_CREAT|O_EXCL|O_WRONLY) or BAIL_OUT "open $f: $!";
+ print $fh <<EOF or BAIL_OUT "print $f: $!";
+[credential]
+ helper = $cred_set
+EOF
+ close $fh or BAIL_OUT "close $f: $!";
+};
+
+# allow testers with git-credential-store configured to reuse
+# stored credentials inside test_lei(sub {...}) when $ENV{HOME}
+# is overridden and localized.
+my ($cred_set, @cred_link, $tmpdir, $for_destroy);
+chomp(my $cred_helper = `git config credential.helper 2>/dev/null`);
+if ($cred_helper eq 'store') {
+ my $config = $ENV{XDG_CONFIG_HOME} // "$ENV{HOME}/.config";
+ for my $f ("$ENV{HOME}/.git-credentials", "$config/git/credentials") {
+ next unless -f $f;
+ @cred_link = ($f, '/.git-credentials');
+ last;
+ }
+ $cred_set = qq("$cred_helper");
+} elsif ($cred_helper =~ /\Acache(?:[ \t]|\z)/) {
+ my $cache = $ENV{XDG_CACHE_HOME} // "$ENV{HOME}/.cache";
+ for my $d ("$ENV{HOME}/.git-credential-cache",
+ "$cache/git/credential") {
+ next unless -d $d;
+ @cred_link = ($d, '/.git-credential-cache');
+ $cred_set = qq("$cred_helper");
+ last;
+ }
+} elsif (!$cred_helper) { # make the test less painful if no creds configured
+ ($tmpdir, $for_destroy) = tmpdir;
+ my $d = "$tmpdir/.git-credential-cache";
+ mkdir($d, 0700) or BAIL_OUT $!;
+ $cred_set = "cache --timeout=60";
+ @cred_link = ($d, '/.git-credential-cache');
+} else {
+ diag "credential.helper=$cred_helper will not be used for this test";
+}
+
+my $mics = do {
+ local $ENV{HOME} = $tmpdir // $ENV{HOME};
+ if ($tmpdir && $cred_set) {
+ $set_cred_helper->("$ENV{HOME}/.gitconfig", $cred_set)
+ }
+ $nwr->imap_common_init;
+};
my $mic = (values %$mics)[0];
-my $cleanup = PublicInbox::OnDestroy->new(sub {
+my $cleanup = PublicInbox::OnDestroy->new($$, sub {
+ my $mic = $nwr->mic_get($imap_url);
$mic->delete($folder) or fail "delete $folder <$folder_uri>: $@";
+ if ($tmpdir && -f "$tmpdir/.gitconfig") {
+ local $ENV{HOME} = $tmpdir;
+ system(qw(git credential-cache exit));
+ }
});
my $imap_append = $nwr->can('imap_append');
my $smsg = bless { kw => [ 'seen' ] }, 'PublicInbox::Smsg';
$imap_append->($mic, $folder, undef, $smsg, eml_load('t/plack-qp.eml'));
-my @res;
$nwr->{quiet} = 1;
-$nwr->imap_each($$folder_uri, sub {
- my ($u, $uid, $kw, $eml, $arg) = @_;
- push @res, [ $kw, $eml ];
-});
-is(scalar(@res), 1, 'got appended message');
-is_deeply(\@res, [ [ [ 'seen' ], eml_load('t/plack-qp.eml') ] ],
+my $imap_slurp_all = sub {
+ my ($u, $uid, $kw, $eml, $res) = @_;
+ push @$res, [ $kw, $eml ];
+};
+$nwr->imap_each($$folder_uri, $imap_slurp_all, my $res = []);
+is(scalar(@$res), 1, 'got appended message');
+my $plack_qp_eml = eml_load('t/plack-qp.eml');
+is_deeply($res, [ [ [ 'seen' ], $plack_qp_eml ] ],
'uploaded message read back');
+$res = $mic = $mics = undef;
+
+test_lei(sub {
+ my ($ro_home, $cfg_path) = setup_public_inboxes;
+ my $cfg = PublicInbox::Config->new($cfg_path);
+ $cfg->each_inbox(sub {
+ my ($ibx) = @_;
+ lei_ok qw(add-external -q), $ibx->{inboxdir} or BAIL_OUT;
+ });
+
+ # cred_link[0] may be on a different (hopefully encrypted) FS,
+ # we only symlink to it here, so we don't copy any sensitive data
+ # into the temporary directory
+ if (@cred_link && !symlink($cred_link[0], $ENV{HOME}.$cred_link[1])) {
+ diag "symlink @cred_link: $! (non-fatal)";
+ $cred_set = undef;
+ }
+ $set_cred_helper->("$ENV{HOME}/.gitconfig", $cred_set) if $cred_set;
+
+ lei_ok qw(q f:qp@example.com -o), $$folder_uri;
+ $nwr->imap_each($$folder_uri, $imap_slurp_all, my $res = []);
+ is(scalar(@$res), 1, 'got one deduped result') or diag explain($res);
+ is_deeply($res->[0]->[1], $plack_qp_eml,
+ 'lei q wrote expected result');
+
+ lei_ok qw(q f:matz -a -o), $$folder_uri;
+ $nwr->imap_each($$folder_uri, $imap_slurp_all, my $aug = []);
+ is(scalar(@$aug), 2, '2 results after augment') or diag explain($aug);
+ my $exp = $res->[0]->[1]->as_string;
+ is(scalar(grep { $_->[1]->as_string eq $exp } @$aug), 1,
+ 'original remains after augment');
+ $exp = eml_load('t/iso-2202-jp.eml')->as_string;
+ is(scalar(grep { $_->[1]->as_string eq $exp } @$aug), 1,
+ 'new result shown after augment');
+
+ lei_ok qw(q s:thisbetternotgiveanyresult -o), $folder_uri->as_string;
+ $nwr->imap_each($$folder_uri, $imap_slurp_all, my $empty = []);
+ is(scalar(@$empty), 0, 'no results w/o augment');
+
+});
-undef $cleanup;
+undef $cleanup; # remove temporary folder
done_testing;
next prev parent reply other threads:[~2021-02-21 7:41 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-21 7:41 [PATCH 0/7] "lei q -o imaps://..." support Eric Wong
2021-02-21 7:41 ` [PATCH 1/7] inbox_writable: require PublicInbox::MdirReader Eric Wong
2021-02-21 7:41 ` Eric Wong [this message]
2021-02-21 7:41 ` [PATCH 3/7] ipc: add wq_broadcast Eric Wong
2021-02-21 7:41 ` [PATCH 4/7] lei q: move augment into lei2mail workers Eric Wong
2021-02-21 7:41 ` [PATCH 5/7] ipc: support setting a locked number of WQ workers Eric Wong
2021-02-21 7:41 ` [PATCH 6/7] net_reader: use and accept URIimap objects in more places Eric Wong
2021-02-21 7:41 ` [PATCH 7/7] lei2mail: parallel augment for lock-free stores Eric Wong
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://public-inbox.org/README
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210221074134.15084-3-e@80x24.org \
--to=e@80x24.org \
--cc=meta@public-inbox.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).