# Copyright (C) all contributors <meta@public-inbox.org>
# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>

# show any VCS object, similar to "git show"
#
# This can use a "solver" to reconstruct blobs based on git
# patches (with abbreviated OIDs in the header).  However, the
# abbreviated OIDs must match exactly what's in the original
# email (unless a normal code repo already has the blob).
#
# In other words, we can only reliably reconstruct blobs based
# on links generated by ViewDiff (and only if the emailed
# patches apply 100% cleanly to published blobs).

package PublicInbox::ViewVCS;
use strict;
use v5.10.1;
use File::Temp 0.19 (); # newdir
use PublicInbox::SolverGit;
use PublicInbox::GitAsyncCat;
use PublicInbox::WwwStream qw(html_oneshot);
use PublicInbox::Linkify;
use PublicInbox::Tmpfile;
use PublicInbox::ViewDiff qw(flush_diff uri_escape_path);
use PublicInbox::View;
use PublicInbox::Eml;
use Text::Wrap qw(wrap);
use PublicInbox::Hval qw(ascii_html to_filename);
my $hl = eval {
	require PublicInbox::HlMod;
	PublicInbox::HlMod->new;
};

my %QP_MAP = ( A => 'oid_a', a => 'path_a', b => 'path_b' );
our $MAX_SIZE = 1024 * 1024; # TODO: configurable
my $BIN_DETECT = 8000; # same as git
my $SHOW_FMT = '--pretty=format:'.join('%n', '%P', '%p', '%H', '%T', '%s', '%f',
	'%an <%ae>  %ai', '%cn <%ce>  %ci', '%b%x00');

my %GIT_MODE = (
	'100644' => ' ', # blob
	'100755' => 'x', # executable blob
	'040000' => 'd', # tree
	'120000' => 'l', # symlink
	'160000' => 'g', # commit (gitlink)
);

sub html_page ($$;@) {
	my ($ctx, $code) = @_[0, 1];
	my $wcb = delete $ctx->{-wcb};
	$ctx->{-upfx} = '../../'; # from "/$INBOX/$OID/s/"
	my $res = html_oneshot($ctx, $code, @_[2..$#_]);
	$wcb ? $wcb->($res) : $res;
}

sub dbg_log ($) {
	my ($ctx) = @_;
	my $log = delete $ctx->{lh} // die 'BUG: already captured debug log';
	if (!seek($log, 0, 0)) {
		warn "seek(log): $!";
		return '<pre>debug log seek error</pre>';
	}
	$log = do { local $/; <$log> } // do {
		warn "readline(log): $!";
		return '<pre>debug log read error</pre>';
	};
	$ctx->{-linkify} //= PublicInbox::Linkify->new;
	"<hr><pre>debug log:\n\n".
		$ctx->{-linkify}->to_html($log).'</pre>';
}

sub stream_blob_parse_hdr { # {parse_hdr} for Qspawn
	my ($r, $bref, $ctx) = @_;
	my ($git, $oid, $type, $size, $di) = @{$ctx->{-res}};
	my @cl = ('Content-Length', $size);
	if (!defined $r) { # sysread error
		html_page($ctx, 500, dbg_log($ctx));
	} elsif (index($$bref, "\0") >= 0) {
		[200, [qw(Content-Type application/octet-stream), @cl] ];
	} else {
		my $n = length($$bref);
		if ($n >= $BIN_DETECT || $n == $size) {
			return [200, [ 'Content-Type',
				'text/plain; charset=UTF-8', @cl ] ];
		}
		if ($r == 0) {
			my $log = dbg_log($ctx);
			warn "premature EOF on $oid $log";
			return html_page($ctx, 500, $log);
		}
		undef; # bref keeps growing
	}
}

sub stream_large_blob ($$) {
	my ($ctx, $res) = @_;
	$ctx->{-res} = $res;
	my ($git, $oid, $type, $size, $di) = @$res;
	my $cmd = ['git', "--git-dir=$git->{git_dir}", 'cat-file', $type, $oid];
	my $qsp = PublicInbox::Qspawn->new($cmd);
	my $env = $ctx->{env};
	$env->{'qspawn.wcb'} = delete $ctx->{-wcb};
	$qsp->psgi_return($env, undef, \&stream_blob_parse_hdr, $ctx);
}

sub show_other_result ($$) { # tag
	my ($bref, $ctx) = @_;
	if (my $qsp_err = delete $ctx->{-qsp_err}) {
		return html_page($ctx, 500, dbg_log($ctx) .
				"git show error:$qsp_err");
	}
	my $l = PublicInbox::Linkify->new;
	utf8::decode($$bref);
	html_page($ctx, 200, '<pre>', $l->to_html($$bref), '</pre><hr>',
		dbg_log($ctx));
}

sub cmt_title { # git->cat_async callback
	my ($bref, $oid, $type, $size, $ctx) = @_;
	utf8::decode($$bref);
	my $title = $$bref =~ /\r?\n\r?\n([^\r\n]+)\r?\n?/ ? $1 : '';
	push(@{$ctx->{-cmt_pt}} , ascii_html($title)) == @{$ctx->{-cmt_P}} and
		cmt_finalize($ctx);
}

sub show_commit_start { # ->psgi_qx callback
	my ($bref, $ctx) = @_;
	if (my $qsp_err = delete $ctx->{-qsp_err}) {
		return html_page($ctx, 500, dbg_log($ctx) .
				"git show/patch-id error:$qsp_err");
	}
	my $patchid = (split(/ /, $$bref))[0]; # ignore commit
	$ctx->{-q_value_html} = "patchid:$patchid" if defined $patchid;
	open my $fh, '<:utf8', "$ctx->{-tmp}/h" or
		die "open $ctx->{-tmp}/h: $!";
	chop(my $buf = do { local $/ = "\0"; <$fh> });
	chomp $buf;
	my ($P, $p);
	($P, $p, @{$ctx->{cmt_info}}) = split(/\n/, $buf, 9);
	return cmt_finalize($ctx) if !$P;
	@{$ctx->{-cmt_P}} = split(/ /, $P);
	@{$ctx->{-cmt_p}} = split(/ /, $p); # abbreviated
	if ($ctx->{env}->{'pi-httpd.async'}) {
		for (@{$ctx->{-cmt_P}}) {
			ibx_async_cat($ctx, $_, \&cmt_title, $ctx);
		}
	} else { # synchronous
		for (@{$ctx->{-cmt_P}}) {
			$ctx->{git}->cat_async($_, \&cmt_title, $ctx);
		}
		$ctx->{git}->cat_async_wait;
	}
}

sub cmt_finalize {
	my ($ctx) = @_;
	$ctx->{-linkify} //= PublicInbox::Linkify->new;
	my $upfx = $ctx->{-upfx} = '../../'; # from "/$INBOX/$OID/s/"
	my ($H, $T, $s, $f, $au, $co, $bdy) = @{delete $ctx->{cmt_info}};
	# try to keep author and committer dates lined up
	my $x = length($au) - length($co);
	if ($x > 0) {
		$x = ' ' x $x;
		$co =~ s/>/>$x/;
	} elsif ($x < 0) {
		$x = ' ' x (-$x);
		$au =~ s/>/>$x/;
	}
	$_ = ascii_html($_) for ($au, $co);
	$au =~ s!(&gt; +)([0-9]{4,}-\S+ \S+)!
		my ($gt, $t) = ($1, $2);
		$t =~ tr/ :-//d;
		qq($gt<a
href="$upfx?t=$t"
title="list contemporary emails">$2</a>)
		!e;
	$ctx->{-title_html} = $s = $ctx->{-linkify}->to_html($s);
	my ($P, $p, $pt) = delete @$ctx{qw(-cmt_P -cmt_p -cmt_pt)};
	$_ = qq(<a href="$upfx$_/s/">).shift(@$p).'</a> '.shift(@$pt) for @$P;
	if (@$P == 1) {
		$x = qq{ (<a
href="$f.patch">patch</a>)\n   <a href=#parent>parent</a> $P->[0]};
	} elsif (@$P > 1) {
		$x = qq(\n  <a href=#parents>parents</a> $P->[0]\n);
		shift @$P;
		$x .= qq(          $_\n) for @$P;
		chop $x;
	} else {
		$x = ' (<a href=#root_commit>root commit</a>)';
	}
	PublicInbox::WwwStream::html_init($ctx);
	my $zfh = $ctx->zfh;
	print $zfh <<EOM;
<pre>   <a href=#commit>commit</a> $H$x
     <a href=#tree>tree</a> <a href="$upfx$T/s/">$T</a>
   author $au
committer $co

<b>$s</b>
EOM
	print $zfh "\n", $ctx->{-linkify}->to_html($bdy) if length($bdy);
	$bdy = '';
	open my $fh, '<:utf8', "$ctx->{-tmp}/p" or
		die "open $ctx->{-tmp}/p: $!";
	if (-s $fh > $MAX_SIZE) {
		print $zfh "---\n patch is too large to show\n";
	} else { # prepare flush_diff:
		read($fh, $x, -s _);
		$ctx->{-apfx} = $ctx->{-spfx} = $upfx;
		$x =~ s/\r?\n/\n/gs;
		$ctx->{-anchors} = {} if $x =~ /^diff --git /sm;
		flush_diff($ctx, \$x); # undefs $x
		# 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);
			print $zfh <<EOM;
<hr><form action=$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
		}
	}
	chop($x = <<EOM);
<hr><pre>glossary
--------
<dfn
id=commit>Commit</dfn> objects reference one tree, and zero or more parents.

Single <dfn
id=parent>parent</dfn> commits can typically generate a patch in
unified diff format via `git format-patch'.

Multiple <dfn id=parents>parents</dfn> means the commit is a merge.

<dfn id=root_commit>Root commits</dfn> have no ancestor.  Note that it is
possible to have multiple root commits when merging independent histories.

Every commit references one top-level <dfn id=tree>tree</dfn> object.</pre>
EOM
	delete($ctx->{env}->{'qspawn.wcb'})->($ctx->html_done($x));
}

sub stream_patch_parse_hdr { # {parse_hdr} for Qspawn
	my ($r, $bref, $ctx) = @_;
	if (!defined $r) { # sysread error
		html_page($ctx, 500, dbg_log($ctx));
	} elsif (index($$bref, "\n\n") >= 0) {
		my $eml = bless { hdr => $bref }, 'PublicInbox::Eml';
		my $fn = to_filename($eml->header('Subject') // '');
		$fn = substr($fn // 'PATCH-no-subject', 6); # drop "PATCH-"
		return [ 200, [ 'Content-Type', 'text/plain; charset=UTF-8',
				'Content-Disposition',
				qq(inline; filename=$fn.patch) ] ];
	} elsif ($r == 0) {
		my $log = dbg_log($ctx);
		warn "premature EOF on $ctx->{patch_oid} $log";
		return html_page($ctx, 500, $log);
	} else {
		undef; # bref keeps growing until "\n\n"
	}
}

sub show_patch ($$) {
	my ($ctx, $res) = @_;
	my ($git, $oid) = @$res;
	my @cmd = ('git', "--git-dir=$git->{git_dir}",
		qw(format-patch -1 --stdout -C),
		"--signature=git format-patch -1 --stdout -C $oid", $oid);
	my $qsp = PublicInbox::Qspawn->new(\@cmd);
	$ctx->{env}->{'qspawn.wcb'} = delete $ctx->{-wcb};
	$ctx->{patch_oid} = $oid;
	$qsp->psgi_return($ctx->{env}, undef, \&stream_patch_parse_hdr, $ctx);
}

sub show_commit ($$) {
	my ($ctx, $res) = @_;
	return show_patch($ctx, $res) if ($ctx->{fn} // '') =~ /\.patch\z/;
	my ($git, $oid) = @$res;
	# patch-id needs two passes, and we use the initial show to ensure
	# a patch embedded inside the commit message body doesn't get fed
	# to patch-id:
	my $cmd = [ '/bin/sh', '-c',
		"git show --encoding=UTF-8 '$SHOW_FMT'".
		" -z --no-notes --no-patch $oid >h && ".
		'git show --encoding=UTF-8 --pretty=format:%n -M'.
		" --stat -p $oid >p && ".
		"git patch-id --stable <p" ];
	my $e = { GIT_DIR => $git->{git_dir} };
	my $qsp = PublicInbox::Qspawn->new($cmd, $e, { -C => "$ctx->{-tmp}" });
	$qsp->{qsp_err} = \($ctx->{-qsp_err} = '');
	$ctx->{env}->{'qspawn.wcb'} = delete $ctx->{-wcb};
	$ctx->{git} = $git;
	$qsp->psgi_qx($ctx->{env}, undef, \&show_commit_start, $ctx);
}

sub show_other ($$) {
	my ($ctx, $res) = @_;
	my ($git, $oid, $type, $size) = @$res;
	$size > $MAX_SIZE and return html_page($ctx, 200,
		ascii_html($type)." $oid is too big to show\n". dbg_log($ctx));
	my $cmd = ['git', "--git-dir=$git->{git_dir}",
		qw(show --encoding=UTF-8 --no-color --no-abbrev), $oid ];
	my $qsp = PublicInbox::Qspawn->new($cmd);
	$qsp->{qsp_err} = \($ctx->{-qsp_err} = '');
	$qsp->psgi_qx($ctx->{env}, undef, \&show_other_result, $ctx);
}

sub show_tree_result ($$) {
	my ($bref, $ctx) = @_;
	if (my $qsp_err = delete $ctx->{-qsp_err}) {
		return html_page($ctx, 500, dbg_log($ctx) .
				"git ls-tree -z error:$qsp_err");
	}
	my @ent = split(/\0/, $$bref);
	my $qp = delete $ctx->{qp};
	my $l = $ctx->{-linkify} //= PublicInbox::Linkify->new;
	my $pfx = $qp->{b};
	$$bref = "<pre><a href=#tree>tree</a> $ctx->{tree_oid}";
	if (defined $pfx) {
		my $x = ascii_html($pfx);
		$pfx .= '/';
		$$bref .= qq(  <a href=#path>path</a>: $x</a>\n);
	} else {
		$pfx = '';
		$$bref .= qq[  (<a href=#path>path</a> unknown)\n];
	}
	my ($x, $m, $t, $oid, $sz, $f, $n);
	$$bref .= "\n	size	name";
	for (@ent) {
		($x, $f) = split(/\t/, $_, 2);
		undef $_;
		($m, $t, $oid, $sz) = split(/ +/, $x, 4);
		$m = $GIT_MODE{$m} // '?';
		utf8::decode($f);
		$n = ascii_html($f);
		if ($m eq 'g') { # gitlink submodule commit
			$$bref .= "\ng\t\t$n @ <a\nhref=#g>commit</a>$oid";
			next;
		}
		my $q = 'b='.ascii_html(uri_escape_path($pfx.$f));
		if ($m eq 'd') { $n .= '/' }
		elsif ($m eq 'x') { $n = "<b>$n</b>" }
		elsif ($m eq 'l') { $n = "<i>$n</i>" }
		$$bref .= qq(\n$m\t$sz\t<a\nhref="../../$oid/s/?$q">$n</a>);
	}
	$$bref .= dbg_log($ctx);
	$$bref .= <<EOM;
<pre>glossary
--------
<dfn
id=tree>Tree</dfn> objects belong to commits or other tree objects.  Trees may
reference blobs, sub-trees, or commits of submodules.

<dfn
id=path>Path</dfn> names are stored in tree objects, but trees do not know
their own path name.  A tree's path name comes from their parent tree,
or it is the root tree referenced by a commit object.  Thus, this web UI
relies on the `b=' URI parameter as a hint to display the path name.

<dfn title="submodule commit"
id=g>Commit</dfn> objects may be stored in trees to reference submodules.</pre>
EOM
	chop $$bref;
	html_page($ctx, 200, $$bref);
}

sub show_tree ($$) {
	my ($ctx, $res) = @_;
	my ($git, $oid, undef, $size) = @$res;
	$size > $MAX_SIZE and return html_page($ctx, 200,
			"tree $oid is too big to show\n". dbg_log($ctx));
	my $cmd = [ 'git', "--git-dir=$git->{git_dir}",
		qw(ls-tree -z -l --no-abbrev), $oid ];
	my $qsp = PublicInbox::Qspawn->new($cmd);
	$ctx->{tree_oid} = $oid;
	$qsp->{qsp_err} = \($ctx->{-qsp_err} = '');
	$qsp->psgi_qx($ctx->{env}, undef, \&show_tree_result, $ctx);
}

# user_cb for SolverGit, called as: user_cb->($result_or_error, $uarg)
sub solve_result {
	my ($res, $ctx) = @_;
	my $hints = delete $ctx->{hints};
	$res or return html_page($ctx, 404, dbg_log($ctx));
	ref($res) eq 'ARRAY' or return html_page($ctx, 500, dbg_log($ctx));

	my ($git, $oid, $type, $size, $di) = @$res;
	return show_commit($ctx, $res) if $type eq 'commit';
	return show_tree($ctx, $res) if $type eq 'tree';
	return show_other($ctx, $res) if $type ne 'blob';
	my $path = to_filename($di->{path_b} // $hints->{path_b} // 'blob');
	my $raw_link = "(<a\nhref=$path>raw</a>)";
	if ($size > $MAX_SIZE) {
		return stream_large_blob($ctx, $res) if defined $ctx->{fn};
		return html_page($ctx, 200, <<EOM . dbg_log($ctx));
<pre><b>Too big to show, download available</b>
blob $oid $size bytes $raw_link</pre>
EOM
	}
	@{$ctx->{-paths}} = ($path, $raw_link);
	$ctx->{git} = $git;
	if ($ctx->{env}->{'pi-httpd.async'}) {
		ibx_async_cat($ctx, $oid, \&show_blob, $ctx);
	} else { # synchronous
		$git->cat_async($oid, \&show_blob, $ctx);
		$git->cat_async_wait;
	}
}

sub show_blob { # git->cat_async callback
	my ($blob, $oid, $type, $size, $ctx) = @_;
	if (!$blob) {
		my $e = "Failed to retrieve generated blob ($oid)";
		warn "$e ($ctx->{git}->{git_dir}) type=$type";
		return html_page($ctx, 500, "<pre><b>$e</b></pre>".dbg_log($ctx))
	}

	my $bin = index(substr($$blob, 0, $BIN_DETECT), "\0") >= 0;
	if (defined $ctx->{fn}) {
		my $h = [ 'Content-Length', $size, 'Content-Type' ];
		push(@$h, ($bin ? 'application/octet-stream' : 'text/plain'));
		return delete($ctx->{-wcb})->([200, $h, [ $$blob ]]);
	}

	my ($path, $raw_link) = @{delete $ctx->{-paths}};
	$bin and return html_page($ctx, 200,
				"<pre>blob $oid $size bytes (binary)" .
				" $raw_link</pre>".dbg_log($ctx));

	# TODO: detect + convert to ensure validity
	utf8::decode($$blob);
	my $nl = ($$blob =~ s/\r?\n/\n/sg);
	my $pad = length($nl);

	($ctx->{-linkify} //= PublicInbox::Linkify->new)->linkify_1($$blob);
	my $ok = $hl->do_hl($blob, $path) if $hl;
	if ($ok) {
		$blob = $ok;
	} else {
		$$blob = ascii_html($$blob);
	}

	# using some of the same CSS class names and ids as cgit
	my $x = "<pre>blob $oid $size bytes $raw_link</pre>" .
		"<hr /><table\nclass=blob>".
		"<tr><td\nclass=linenumbers><pre>";
	# scratchpad in this loop is faster here than `printf $zfh':
	$x .= sprintf("<a id=n$_ href=#n$_>% ${pad}u</a>\n", $_) for (1..$nl);
	$x .= '</pre></td><td><pre> </pre></td>'. # pad for non-CSS users
		"<td\nclass=lines><pre\nstyle='white-space:pre'><code>";
	html_page($ctx, 200, $x, $ctx->{-linkify}->linkify_2($$blob),
		'</code></pre></td></tr></table>'.dbg_log($ctx));
}

# GET /$INBOX/$GIT_OBJECT_ID/s/
# GET /$INBOX/$GIT_OBJECT_ID/s/$FILENAME
sub show ($$;$) {
	my ($ctx, $oid_b, $fn) = @_;
	my $qp = $ctx->{qp};
	my $hints = $ctx->{hints} = {};
	while (my ($from, $to) = each %QP_MAP) {
		defined(my $v = $qp->{$from}) or next;
		$hints->{$to} = $v if $v ne '';
	}
	$ctx->{fn} = $fn;
	$ctx->{-tmp} = File::Temp->newdir("solver.$oid_b-XXXX", TMPDIR => 1);
	open $ctx->{lh}, '+>>', "$ctx->{-tmp}/solve.log" or die "open: $!";
	my $solver = PublicInbox::SolverGit->new($ctx->{ibx},
						\&solve_result, $ctx);
	$solver->{tmp} = $ctx->{-tmp}; # share tmpdir
	# PSGI server will call this immediately and give us a callback (-wcb)
	sub {
		$ctx->{-wcb} = $_[0]; # HTTP write callback
		$solver->solve($ctx->{env}, $ctx->{lh}, $oid_b, $hints);
	};
}

1;