… | |
… | |
12 | sub on_init { |
12 | sub on_init { |
13 | my ($self) = @_; |
13 | my ($self) = @_; |
14 | |
14 | |
15 | for (my $idx = 0; defined (my $res = $self->x_resource ("selection.pattern-$idx")); $idx++) { |
15 | for (my $idx = 0; defined (my $res = $self->x_resource ("selection.pattern-$idx")); $idx++) { |
16 | no re 'eval'; # just to be sure |
16 | no re 'eval'; # just to be sure |
|
|
17 | $res = $self->locale_decode ($res); |
|
|
18 | utf8::encode $res; |
17 | push @{ $self->{patterns} }, qr/$res/; |
19 | push @{ $self->{patterns} }, qr/$res/; |
18 | } |
20 | } |
19 | |
21 | |
20 | () |
22 | () |
21 | } |
23 | } |
… | |
… | |
65 | my $curlen = $line->offset_of ($self->selection_end) |
67 | my $curlen = $line->offset_of ($self->selection_end) |
66 | - $line->offset_of ($self->selection_beg); |
68 | - $line->offset_of ($self->selection_beg); |
67 | |
69 | |
68 | my @matches; |
70 | my @matches; |
69 | |
71 | |
70 | for my $regex (@mark_patterns, @{ $self->{patterns} }) { |
72 | if ($markofs < $line->l) { |
71 | while ($text =~ /$regex/g) { |
73 | # convert markofs form character to UTF-8 offset space |
72 | if ($-[1] <= $markofs and $markofs <= $+[1]) { |
74 | { |
73 | my $ofs = $-[1]; |
75 | my $prefix = substr $text, 0, $markofs; |
74 | my $match = $1; |
76 | utf8::encode $prefix; |
|
|
77 | $markofs = length $prefix; |
|
|
78 | } |
75 | |
79 | |
|
|
80 | # not doing matches in unicode mode helps speed |
|
|
81 | # enourmously here. working in utf-8 should be |
|
|
82 | # equivalent due to the magic of utf-8 encoding. |
|
|
83 | utf8::encode $text; |
|
|
84 | study $text; # _really_ helps, too :) |
|
|
85 | |
|
|
86 | for my $regex (@mark_patterns, @{ $self->{patterns} }) { |
|
|
87 | while ($text =~ /$regex/g) { |
|
|
88 | if ($-[1] <= $markofs and $markofs <= $+[1]) { |
|
|
89 | my $ofs = $-[1]; |
|
|
90 | my $match = $1; |
|
|
91 | |
76 | for my $regex (@simplify_patterns) { |
92 | for my $regex (@simplify_patterns) { |
77 | if ($match =~ $regex) { |
93 | if ($match =~ $regex) { |
78 | $match = $1; |
94 | $match = $1; |
79 | $ofs += $-[1]; |
95 | $ofs += $-[1]; |
|
|
96 | } |
80 | } |
97 | } |
|
|
98 | |
|
|
99 | push @matches, [$ofs, length $match]; |
81 | } |
100 | } |
82 | |
|
|
83 | push @matches, [$ofs, length $match]; |
|
|
84 | } |
101 | } |
85 | } |
102 | } |
86 | } |
103 | } |
87 | |
104 | |
88 | # whole line |
105 | # whole line |
… | |
… | |
91 | for (sort { $a->[1] <=> $b->[1] or $b->[0] <=> $a->[0] } @matches) { |
108 | for (sort { $a->[1] <=> $b->[1] or $b->[0] <=> $a->[0] } @matches) { |
92 | my ($ofs, $len) = @$_; |
109 | my ($ofs, $len) = @$_; |
93 | |
110 | |
94 | next if $len <= $curlen; |
111 | next if $len <= $curlen; |
95 | |
112 | |
|
|
113 | # convert back from UTF-8 offset space to character space |
|
|
114 | { |
|
|
115 | my $length = substr $text, $ofs, $len; |
|
|
116 | utf8::decode $length; |
|
|
117 | $len = length $length; |
|
|
118 | } |
|
|
119 | { |
|
|
120 | my $prefix = substr $text, 0, $ofs; |
|
|
121 | utf8::decode $prefix; |
|
|
122 | $ofs = length $prefix; |
|
|
123 | } |
|
|
124 | |
96 | $self->selection_beg ($line->coord_of ($ofs)); |
125 | $self->selection_beg ($line->coord_of ($ofs)); |
97 | $self->selection_end ($line->coord_of ($ofs + $len)); |
126 | $self->selection_end ($line->coord_of ($ofs + $len)); |
98 | return 1; |
127 | return 1; |
99 | } |
128 | } |
100 | |
129 | |