view: generate query in single-message and commit views

The dfblob: search prefix is probably under-utilized, but is
extremely powerful IMHO.  To make it easier-to-use, add a search
textarea with it prefilled with values for the existing patch
message.  This allows users to easily run a query for all
patches which alter or result in either pre or post-image
blobs in the current patch.

Behavior changes are as follows: "changed" in the diffstat
jumps to the bottom of the message.  For /T/ and /t/, it
goes to the "related" anchor which is just above the reply
instructions in the single-message view.  For the single
message view, it'll jump to the textarea search form.

I initially wanted to use a normal `<a href=' link, but
figured the textarea is advantageous for two reasons:

1) users should be able to edit the query before submitting
2) crawlers are less likely to waste CPU/disk on forms

It's probably too noisy to add this directly to the /T/ and /t/
views, but seems like a good place to put above the reply
instructions in the single message view.

Note that the queries used by the /$COMMIT_OID/s/ view is
subtly different than the /$MSGID/ view since git will lengthen
its abbreviations over time, while emails are immutable.

I tried adding dfn: (filename) and s: (subject) support, but
couldn't come up with cases where it really made sense for
/$MSGID/.  /$COMMIT_OID/s/ may benefit from it, since patchid:
could be flaky due to non-standard diff generation options.

3 files changed, 77 insertions, 10 deletions

diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index c28505f1..6bfaf1bb 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -7,6 +7,7 @@ package PublicInbox::View;
 use strict;
 use v5.10.1;
 use List::Util qw(max);
+use Text::Wrap qw(wrap); # stdlib, we need Perl 5.6+ for $huge
 use PublicInbox::MsgTime qw(msg_datestamp);
 use PublicInbox::Hval qw(ascii_html obfuscate_addrs prurl mid_href
                         ts2str fmt_ts);
@@ -246,6 +247,7 @@ sub eml_entry {
 
         # scan through all parts, looking for displayable text
         $ctx->{mhref} = $mhref;
+        $ctx->{end_id} = "e$id";
         $ctx->{obuf} = \$rv;
         $eml->each_part(\&add_text_body, $ctx, 1);
         delete $ctx->{obuf};
@@ -256,6 +258,9 @@ sub eml_entry {
                 " <a\nhref=\"${mhref}raw\">raw</a>" .
                 " <a\nhref=\"${mhref}#R\">reply</a>";
 
+        delete($ctx->{-qry}) and
+                $rv .= qq[ <a\nhref="${mhref}#related">related</a>];
+
         my $hr;
         if (defined(my $pct = $smsg->{pct})) { # used by SearchView.pm
                 $rv .= "\t[relevance $pct%]";
@@ -820,15 +825,38 @@ sub _parent_headers {
 # returns a string buffer
 sub html_footer {
         my ($ctx, $hdr) = @_;
-        my $ibx = $ctx->{ibx};
         my $upfx = '../';
         my $skel;
         my $rv = '<pre>';
-        if ($ibx->over) {
+        my $related;
+        my $qry = delete $ctx->{-qry};
+        if ($qry && $ctx->{ibx}->isrch) {
+                my $q = ''; # search for either ancestor or descendent patches
+                for (@{$qry->{dfpre}}, @{$qry->{dfpost}}) {
+                        chop if length > 7; # include 1 abbrev "older" patches
+                        $q .= "dfblob:$_ ";
+                }
+                chop $q; # omit trailing SP
+                local $Text::Wrap::columns = COLS;
+                local $Text::Wrap::huge = 'overflow';
+                $q = wrap('', '', $q);
+                my $rows = ($q =~ tr/\n/\n/) + 1;
+                $q = ascii_html($q);
+                $related = <<EOM;
+<form id=related
+action=$upfx
+><pre>find likely ancestor, descendant, or conflicting patches:
+<textarea name=q cols=${\COLS} rows=$rows>$q</textarea>
+<input type=submit value=search
+/>\t(<a href=${upfx}_/text/help/>help</a>)</pre></form>
+EOM
+        }
+        if ($ctx->{ibx}->over) {
                 my $t = ts2str($ctx->{-t_max});
                 my $t_fmt = fmt_ts($ctx->{-t_max});
-                $skel .= <<EOF;
-        other threads:[<a
+                my $fallback = $related ? "\t" : "<a id=related>\t</a>";
+                $skel = <<EOF;
+${fallback}other threads:[<a
 href="$upfx?t=$t">~$t_fmt UTC</a>|<a
 href="$upfx">newest</a>]
 EOF
@@ -869,6 +897,7 @@ EOF
         $rv .= qq(<a\nhref="#R">reply</a>);
         $rv .= $skel;
         $rv .= '</pre>';
+        $rv .= $related // '';
         $rv .= msg_reply($ctx, $hdr);
 }
 
diff --git a/lib/PublicInbox/ViewDiff.pm b/lib/PublicInbox/ViewDiff.pm
index fb394b7c..960f7e61 100644
--- a/lib/PublicInbox/ViewDiff.pm
+++ b/lib/PublicInbox/ViewDiff.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2019-2021 all contributors <meta@public-inbox.org>
+# Copyright (C) all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 #
 # used by PublicInbox::View
@@ -141,13 +141,16 @@ sub diff_header ($$$) {
 
         # no need to capture oid_a and oid_b on add/delete,
         # we just linkify OIDs directly via s///e in conditional
-        if (($$x =~ s/$NULL_TO_BLOB/$1 . oid($dctx, $spfx, $2)/e) ||
-                ($$x =~ s/$BLOB_TO_NULL/
-                        'index ' . oid($dctx, $spfx, $1) . $2/e)) {
+        if ($$x =~ s/$NULL_TO_BLOB/$1 . oid($dctx, $spfx, $2)/e) {
+                push @{$ctx->{-qry}->{dfpost}}, $2;
+        } elsif ($$x =~ s/$BLOB_TO_NULL/'index '.oid($dctx, $spfx, $1).$2/e) {
+                push @{$ctx->{-qry}->{dfpre}}, $1;
         } elsif ($$x =~ $BLOB_TO_BLOB) {
                 # modification-only, not add/delete:
                 # linkify hunk headers later using oid_a and oid_b
                 @$dctx{qw(oid_a oid_b)} = ($1, $2);
+                push @{$ctx->{-qry}->{dfpre}}, $1;
+                push @{$ctx->{-qry}->{dfpost}}, $2;
         } else {
                 warn "BUG? <$$x> had no ^index line";
         }
@@ -172,9 +175,16 @@ sub diff_before_or_after ($$) {
                         # ok, looks like a diffstat, go line-by-line:
                         for my $l (split(/^/m, $y)) {
                                 if ($l =~ /^ (.+)( +\| .*\z)/s) {
-                                        anchor0($dst, $ctx, $1, $2) and next;
+                                        anchor0($dst, $ctx, $1, $2) or
+                                                $$dst .= $linkify->to_html($l);
+                                } elsif ($l =~ s/^( [0-9]+ files? )changed,//) {
+                                        $$dst .= $1;
+                                        my $end = $ctx->{end_id} // 'related';
+                                        $$dst .= "<a href=#$end>changed</a>,";
+                                        $$dst .= ascii_html($l);
+                                } else {
+                                        $$dst .= $linkify->to_html($l);
                                 }
-                                $$dst .= $linkify->to_html($l);
                         }
                 } else { # commit message, notes, etc
                         $$dst .= $linkify->to_html($y);
diff --git a/lib/PublicInbox/ViewVCS.pm b/lib/PublicInbox/ViewVCS.pm
index 19d34092..dcb5c67f 100644
--- a/lib/PublicInbox/ViewVCS.pm
+++ b/lib/PublicInbox/ViewVCS.pm
@@ -21,6 +21,8 @@ use PublicInbox::WwwStream qw(html_oneshot);
 use PublicInbox::Linkify;
 use PublicInbox::Tmpfile;
 use PublicInbox::ViewDiff qw(flush_diff);
+use PublicInbox::View;
+use Text::Wrap qw(wrap);
 use PublicInbox::Hval qw(ascii_html to_filename);
 my $hl = eval {
         require PublicInbox::HlMod;
@@ -157,6 +159,32 @@ EOM
                 $bdy =~ s/\r?\n/\n/gs;
                 flush_diff($ctx, \$bdy);
                 $ctx->zmore($buf);
+                undef $buf;
+                # TODO: should there be another textarea which attempts to
+                # search for the exact email which was applied to make this
+                # commit?
+                if (my $qry = delete $ctx->{-qry}) {
+                        my $q = '';
+                        for (@{$qry->{dfpost}}, @{$qry->{dfpre}}) {
+                                # keep blobs as short as reasonable, emails
+                                # are going to be older than what's in git
+                                substr($_, 7, 64, '');
+                                $q .= "dfblob:$_ ";
+                        }
+                        chop $q; # no trailing SP
+                        local $Text::Wrap::columns = PublicInbox::View::COLS;
+                        local $Text::Wrap::huge = 'overflow';
+                        $q = wrap('', '', $q);
+                        my $rows = ($q =~ tr/\n/\n/) + 1;
+                        $q = ascii_html($q);
+                        $ctx->zmore(<<EOM);
+<hr><form action=$upfx
+id=related><pre>find related emails, including ancestors/descendants/conflicts
+<textarea name=q cols=${\PublicInbox::View::COLS} rows=$rows>$q</textarea>
+<input type=submit value=search
+/>\t(<a href=${upfx}_/text/help/>help</a>)</pre></form>
+EOM
+                }
         }
         $x = $ctx->zflush($ctx->_html_end);
         my $res_hdr = delete $ctx->{-res_hdr};