From: Eric Wong <e@80x24.org>
To: meta@public-inbox.org
Subject: [PATCH 2/6] switch read-only uses of walk_parts to msg_iter
Date: Thu, 19 May 2016 21:28:43 +0000 [thread overview]
Message-ID: <20160519212847.4822-2-e@80x24.org> (raw)
In-Reply-To: <20160519212847.4822-1-e@80x24.org>
msg_iter lets us know the index of the attachment,
allow us to make more sensible labels and in a future
commit, hyperlinks to download attachments.
---
lib/PublicInbox/SearchIdx.pm | 6 ++--
lib/PublicInbox/View.pm | 77 ++++++++++++++++++++------------------------
t/view.t | 4 +--
3 files changed, 40 insertions(+), 47 deletions(-)
diff --git a/lib/PublicInbox/SearchIdx.pm b/lib/PublicInbox/SearchIdx.pm
index 63be681..9192bb0 100644
--- a/lib/PublicInbox/SearchIdx.pm
+++ b/lib/PublicInbox/SearchIdx.pm
@@ -11,6 +11,7 @@ use strict;
use warnings;
use base qw(PublicInbox::Search);
use PublicInbox::MID qw/mid_clean id_compress mid_mime/;
+use PublicInbox::MsgIter;
require PublicInbox::Git;
*xpfx = *PublicInbox::Search::xpfx;
@@ -114,9 +115,8 @@ sub add_message {
$tg->index_text($smsg->from);
$tg->increase_termpos;
- $mime->walk_parts(sub {
- my ($part) = @_;
- return if $part->subparts; # walk_parts already recurses
+ msg_iter($mime, sub {
+ my ($part, $depth, @idx) = @{$_[0]};
my $ct = $part->content_type || $ct_msg;
# account for filter bugs...
diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index 6c283ab..4260167 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -14,6 +14,7 @@ use Email::MIME::ContentType qw/parse_content_type/;
use PublicInbox::Hval qw/ascii_html/;
use PublicInbox::Linkify;
use PublicInbox::MID qw/mid_clean id_compress mid2path mid_mime/;
+use PublicInbox::MsgIter;
require POSIX;
use constant INDENT => ' ';
@@ -90,7 +91,6 @@ sub index_entry {
my $midx = $state->{anchor_idx}++;
my $ctx = $state->{ctx};
my $srch = $ctx->{srch};
- my $part_nr = 0;
my $hdr = $mime->header_obj;
my $subj = $hdr->header('Subject');
@@ -125,10 +125,7 @@ sub index_entry {
my $mhref = "${path}$href/";
# scan through all parts, looking for displayable text
- $mime->walk_parts(sub {
- index_walk($fh, $_[0], \$part_nr);
- });
- $mime->body_set('');
+ msg_iter($mime, sub { index_walk($fh, $_[0]) });
$rv = "\n" . html_footer($hdr, 0, $ctx, "$path$href/R/");
if (defined $irt) {
@@ -214,8 +211,8 @@ sub emit_thread_html {
}
sub index_walk {
- my ($fh, $part, $part_nr) = @_;
- my $s = add_text_body($part, $part_nr);
+ my ($fh, $p) = @_;
+ my $s = add_text_body($p);
return if $s eq '';
@@ -227,30 +224,19 @@ sub index_walk {
sub multipart_text_as_html {
my ($mime) = @_;
my $rv = "";
- my $part_nr = 0;
# scan through all parts, looking for displayable text
- $mime->walk_parts(sub {
- my ($part) = @_;
- $part = add_text_body($part, \$part_nr);
- $rv .= $part;
- $rv .= "\n" if $part ne '';
+ msg_iter($mime, sub {
+ my ($p) = @_;
+ $p = add_text_body($p);
+ $rv .= $p;
+ $rv .= "\n" if $p ne '';
});
- $mime->body_set('');
$rv;
}
-sub add_filename_line {
- my ($fn) = @_;
- my $len = 72;
- my $pad = "-";
- $len -= length($fn);
- $pad x= ($len/2) if ($len > 0);
- "$pad " . ascii_html($fn) . " $pad\n";
-}
-
sub flush_quote {
- my ($s, $l, $quot, $part_nr) = @_;
+ my ($s, $l, $quot) = @_;
# show everything in the full version with anchor from
# short version (see above)
@@ -263,41 +249,50 @@ sub flush_quote {
$$s .= qq(<span\nclass="q">) . $rv . '</span>'
}
-sub attach ($$) {
- my ($ct, $n) = @_;
- my $nl = $n ? "\n" : '';
- "$nl<b>[-- Attachment #$n: " . ascii_html($ct) . " --]\n".
- "[-- TODO not shown --]</b>";
+sub attach_link ($$$) {
+ my ($ct, $p, $fn) = @_;
+ my ($part, $depth, @idx) = @$p;
+ my $nl = $idx[-1] > 1 ? "\n" : '';
+ my $idx = join('.', @idx);
+ my $size = bytes::length($part->body);
+ $ct ||= 'text/plain';
+ $ct =~ s/;.*//; # no attributes
+ $ct = ascii_html($ct);
+ my $desc = $part->header('Content-Description');
+ $desc = $fn unless defined $desc;
+ $desc = '' unless defined $desc;
+ $desc = ': '.$desc if $desc;
+ "$nl<b>[-- Attachment #$idx$desc --]\n" .
+ "[-- Type: $ct, Size: $size bytes --]</b>"
}
sub add_text_body {
- my ($part, $part_nr) = @_;
- return '' if $part->subparts;
+ my ($p) = @_; # from msg_iter: [ Email::MIME, depth, @idx ]
+ my ($part, $depth, @idx) = @$p;
my $ct = $part->content_type;
+ my $fn = $part->filename;
if (defined $ct && $ct =~ m!\btext/x?html\b!i) {
- return attach($ct, $$part_nr);
+ return attach_link($ct, $p, $fn);
}
my $s = eval { $part->body_str };
# badly-encoded message? tell the world about it!
- return attach($ct, $$part_nr) if $@;
+ return attach_link($ct, $p, $fn) if $@;
my @lines = split(/^/m, $s);
$s = '';
- if ($$part_nr > 0) {
- my $fn = $part->filename;
- defined($fn) or $fn = "part #" . ($$part_nr + 1);
- $s .= add_filename_line($fn);
+ if (defined($fn) || $depth > 1 || $idx[0] > 1) {
+ $s .= attach_link($ct, $p, $fn);
+ $s .= "\n\n";
}
-
my @quot;
my $l = PublicInbox::Linkify->new;
while (defined(my $cur = shift @lines)) {
if ($cur !~ /^>/) {
# show the previously buffered quote inline
- flush_quote(\$s, $l, \@quot, $$part_nr) if @quot;
+ flush_quote(\$s, $l, \@quot) if @quot;
# regular line, OK
$cur = $l->linkify_1($cur);
@@ -308,9 +303,7 @@ sub add_text_body {
}
}
- flush_quote(\$s, $l, \@quot, $$part_nr) if @quot;
- ++$$part_nr;
-
+ flush_quote(\$s, $l, \@quot) if @quot;
$s =~ s/[ \t]+$//sgm; # kill per-line trailing whitespace
$s =~ s/\A\n+//s; # kill leading blank lines
$s =~ s/\s+\z//s; # kill all trailing spaces (final "\n" added if ne '')
diff --git a/t/view.t b/t/view.t
index c6b1e7d..a70c52d 100644
--- a/t/view.t
+++ b/t/view.t
@@ -72,7 +72,7 @@ EOF
);
my $html = PublicInbox::View::msg_html(undef, $mime);
- like($html, qr/hi\n-+ part #2 -+\nbye\n/, "multipart split");
+ like($html, qr/hi\n.*-- Attachment #2.*\nbye\n/s, "multipart split");
}
# multipart email with attached patch
@@ -101,7 +101,7 @@ EOF
);
my $html = PublicInbox::View::msg_html(undef, $mime);
- like($html, qr!see attached patch\n-+ foo\.patch -+\n--- a/file\n!,
+ like($html, qr!.*Attachment #2: foo\.patch --!,
"parts split with filename");
}
--
EW
next prev parent reply other threads:[~2016-05-19 21:28 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-19 21:25 [PATCH 1/7] view: rely on Email::MIME::body_str for decoding Eric Wong
2016-05-19 21:28 ` [PATCH 1/6] msg_iter: new internal API for iterating through MIME Eric Wong
2016-05-19 21:28 ` Eric Wong [this message]
2016-05-19 21:28 ` [PATCH 3/6] www: support downloading attachments Eric Wong
2016-05-19 21:28 ` [PATCH 4/6] msg_iter: workaround broken Email::MIME versions Eric Wong
2016-05-19 21:28 ` [PATCH 5/6] www: validate and check filenames in URLs Eric Wong
2016-05-19 21:28 ` [PATCH 6/6] view: reduce clutter for attachments w/o description Eric Wong
2016-05-19 22:06 ` [PATCH 8/7] www: tighten up allowable filenames for attachments 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=20160519212847.4822-2-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).