* [PATCH 2/5] lei_mirror: do not re-fetch inbox.config.example
2023-03-13 12:00 [PATCH 0/5] clone improvements Eric Wong
2023-03-13 12:00 ` [PATCH 1/5] lei_mirror: describe why the {ibx} field is used Eric Wong
@ 2023-03-13 12:00 ` Eric Wong
2023-03-13 12:00 ` [PATCH 3/5] lei_mirror: do not fetch to read-only directories Eric Wong
` (2 subsequent siblings)
4 siblings, 0 replies; 8+ messages in thread
From: Eric Wong @ 2023-03-13 12:00 UTC (permalink / raw)
To: meta
It's a significant source of latency for incremental updates at
the moment, and not really needed since it's just an example.
---
lib/PublicInbox/LeiMirror.pm | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/lib/PublicInbox/LeiMirror.pm b/lib/PublicInbox/LeiMirror.pm
index d878f1e4..967a6422 100644
--- a/lib/PublicInbox/LeiMirror.pm
+++ b/lib/PublicInbox/LeiMirror.pm
@@ -620,7 +620,8 @@ sub clone_v1 {
\&run_puh, $self, $fini));
}
if (!$self->{-is_epoch} && $lei->{opt}->{'inbox-config'} =~
- /\A(?:always|v1)\z/s) {
+ /\A(?:always|v1)\z/s &&
+ !-f "$dst/inbox.config.example") {
_get_txt_start($self, '_/text/config/raw', $fini);
}
@@ -923,8 +924,10 @@ failed to extract epoch number from $src
$self->{dry_run} or File::Path::mkpath($dst);
- $lei->{opt}->{'inbox-config'} =~ /\A(?:always|v2)\z/s and
+ if ($lei->{opt}->{'inbox-config'} =~ /\A(?:always|v2)\z/s &&
+ !-f "$dst/inbox.config.example") {
_get_txt_start($task, '_/text/config/raw', $fini);
+ }
defined($desc) ? ($task->{'txt.description'} = $desc) :
_get_txt_start($task, 'description', $fini);
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH 3/5] lei_mirror: do not fetch to read-only directories
2023-03-13 12:00 [PATCH 0/5] clone improvements Eric Wong
2023-03-13 12:00 ` [PATCH 1/5] lei_mirror: describe why the {ibx} field is used Eric Wong
2023-03-13 12:00 ` [PATCH 2/5] lei_mirror: do not re-fetch inbox.config.example Eric Wong
@ 2023-03-13 12:00 ` Eric Wong
2023-03-13 12:00 ` [PATCH 4/5] lei_mirror: handle UTF-8 from manifest.js.gz properly Eric Wong
2023-03-13 12:00 ` [PATCH 5/5] doc: clone: document --remote-manifest= option Eric Wong
4 siblings, 0 replies; 8+ messages in thread
From: Eric Wong @ 2023-03-13 12:00 UTC (permalink / raw)
To: meta
As with public-inbox-fetch, we shouldn't waste time fetching
into read-only directories, since --epoch= will make unwanted
epoch directories read-only placeholders.
---
lib/PublicInbox/LeiMirror.pm | 9 ++++++++-
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/lib/PublicInbox/LeiMirror.pm b/lib/PublicInbox/LeiMirror.pm
index 967a6422..3ec8170f 100644
--- a/lib/PublicInbox/LeiMirror.pm
+++ b/lib/PublicInbox/LeiMirror.pm
@@ -593,8 +593,15 @@ sub clone_v1 {
die "$uri is a v1 inbox, --epoch is not supported\n";
$self->{-torsocks} //= $curl->torsocks($lei, $uri) or return;
my $dst = $self->{cur_dst} // $self->{dst};
- my $fini = PublicInbox::OnDestroy->new($$, \&v1_done, $self);
my $resume = -d $dst;
+ if ($resume) { # respect read-only cloned w/ --epoch=
+ my @st = stat(_); # for root
+ if (!-w _ || !($st[2] & 0222)) {
+ warn "# skipping $dst, not writable\n";
+ return;
+ }
+ }
+ my $fini = PublicInbox::OnDestroy->new($$, \&v1_done, $self);
if (my $fgrp = forkgroup_prep($self, $uri)) {
$fgrp->{-fini} = $fini;
if ($resume) {
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH 4/5] lei_mirror: handle UTF-8 from manifest.js.gz properly
2023-03-13 12:00 [PATCH 0/5] clone improvements Eric Wong
` (2 preceding siblings ...)
2023-03-13 12:00 ` [PATCH 3/5] lei_mirror: do not fetch to read-only directories Eric Wong
@ 2023-03-13 12:00 ` Eric Wong
2023-03-13 12:00 ` [PATCH 5/5] doc: clone: document --remote-manifest= option Eric Wong
4 siblings, 0 replies; 8+ messages in thread
From: Eric Wong @ 2023-03-13 12:00 UTC (permalink / raw)
To: meta
This should ensure we display the "git config gitweb.owner
$OWNER" command invocation properly and also ensures we set the
description properly without triggering wide character warnings.
Also tested with a smallish iproute2 repo
(/pub/scm/linux/kernel/git/toke/iproute2.git) using my mirror:
public-inbox-clone --remote-manifest=pub/manifest.js.gz \
--include='*/toke/iproute2.git' --inbox-config=never \
https://80x24.org/lore $DST
Anyways, I'm fairly certain this change and its tests are
correct; but I still struggle to understand Perl's approach to
Unicode and it's interactions with various JSON implementations.
Fixes: 0830817c132cb105 ("lei_mirror: show non-ASCII owner properly w/ --verbose")
---
lib/PublicInbox/LeiMirror.pm | 6 +++---
t/clone-coderepo.t | 8 ++++++--
2 files changed, 9 insertions(+), 5 deletions(-)
diff --git a/lib/PublicInbox/LeiMirror.pm b/lib/PublicInbox/LeiMirror.pm
index 3ec8170f..18932cf4 100644
--- a/lib/PublicInbox/LeiMirror.pm
+++ b/lib/PublicInbox/LeiMirror.pm
@@ -259,8 +259,7 @@ sub run_reap {
sub start_cmd {
my ($self, $cmd, $opt, $fini) = @_;
do_reap($self);
- utf8::decode(my $msg = "# @$cmd");
- $self->{lei}->qerr($msg);
+ $self->{lei}->qerr("# @$cmd");
return if $self->{dry_run};
$LIVE->{spawn($cmd, undef, $opt)} = [ \&reap_cmd, $self, $cmd, $fini ]
}
@@ -633,7 +632,7 @@ sub clone_v1 {
}
my $d = $self->{-ent} ? $self->{-ent}->{description} : undef;
- $self->{'txt.description'} = $d if defined $d;
+ utf8::encode($self->{'txt.description'} = $d) if defined $d;
(!defined($d) && !$end) and
_get_txt_start($self, 'description', $fini);
@@ -823,6 +822,7 @@ sub update_ent {
$new = $self->{-ent}->{owner} // return;
$cur = $self->{-local_manifest}->{$key}->{owner} // "\0";
return if $cur eq $new;
+ utf8::encode($new); # to octets
my $cmd = [ qw(git config -f), "$dst/config", 'gitweb.owner', $new ];
start_cmd($self, $cmd, { 2 => $self->{lei}->{2} });
}
diff --git a/t/clone-coderepo.t b/t/clone-coderepo.t
index 1f33a6d7..3a5997c9 100644
--- a/t/clone-coderepo.t
+++ b/t/clone-coderepo.t
@@ -63,11 +63,13 @@ EOM
my $env = { TEST_DOCROOT => "$tmpdir/src", PI_CONFIG => $pi_config };
$td = start_script($cmd, $env, { 3 => $tcp });
my $fp = sha1_hex(my $refs = xqx([@git, 'show-ref']));
+ my $alice = "\x{100}lice";
$m = {
'/a.git' => {
fingerprint => $fp,
modified => 1,
- owner => 'Alice',
+ owner => $alice,
+ description => "${alice}'s repo",
},
'/b.git' => {
fingerprint => $fp,
@@ -89,9 +91,11 @@ my $cmd = [qw(-clone --inbox-config=never --manifest= --project-list=
--objstore= -p -q), $url, "$tmpdir/dst", '--exit-code'];
ok(run_script($cmd), 'clone');
is(xqx([qw(git config gitweb.owner)], { GIT_DIR => "$tmpdir/dst/a.git" }),
- "Alice\n", 'a.git gitweb.owner set');
+ "\xc4\x80lice\n", 'a.git gitweb.owner set');
is(xqx([qw(git config gitweb.owner)], { GIT_DIR => "$tmpdir/dst/b.git" }),
"Bob\n", 'b.git gitweb.owner set');
+my $desc = PublicInbox::Git::try_cat("$tmpdir/dst/a.git/description");
+is($desc, "\xc4\x80lice's repo\n", 'description set');
my $dst_pl = "$tmpdir/dst/projects.list";
my $dst_mf = "$tmpdir/dst/manifest.js.gz";
^ permalink raw reply related [flat|nested] 8+ messages in thread