… | |
… | |
11 | |
11 | |
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 |
|
|
17 | $res = utf8::encode $self->locale_decode ($res); |
16 | $res = $self->locale_decode ($res); |
|
|
17 | utf8::encode $res; |
18 | push @{ $self->{patterns} }, qr/$res/; |
18 | push @{ $self->{patterns} }, qr/$res/; |
19 | } |
19 | } |
20 | |
20 | |
21 | () |
21 | () |
22 | } |
22 | } |
23 | |
23 | |
24 | # "find interetsing things"-patterns |
24 | # "find interesting things"-patterns |
25 | my @mark_patterns = ( |
25 | my @mark_patterns = ( |
26 | # common types of "parentheses" |
26 | # common types of "parentheses" |
27 | qr{ (?<![^[:space:]]) ‘ ([^‘’]+) ’ (?![^[:space]]) }x, |
27 | qr{ (?<![^[:space:]]) ‘ ([^‘’]+) ’ (?![^[:space]]) }x, |
28 | qr{ (?<![^[:space:]]) ` ([^`']+) ' (?![^[:space]]) }x, |
28 | qr{ (?<![^[:space:]]) ` ([^`']+) ' (?![^[:space]]) }x, |
29 | qr{ (?<![^[:space:]]) (" [^[:space:]] [^"]* ") }x, |
29 | qr{ (?<![^[:space:]]) (" [^[:space:]] [^"]* ") }x, |
… | |
… | |
38 | (?:https?|ftp|news|mailto|file)://[ab-zA-Z0-9\-\@;\/?:&=%\$_.+!*\x27(),~]+ |
38 | (?:https?|ftp|news|mailto|file)://[ab-zA-Z0-9\-\@;\/?:&=%\$_.+!*\x27(),~]+ |
39 | [ab-zA-Z0-9\-\@;\/?:&=%\$_+!*\x27()~] # exclude some trailing characters (heuristic) |
39 | [ab-zA-Z0-9\-\@;\/?:&=%\$_+!*\x27()~] # exclude some trailing characters (heuristic) |
40 | )}x, |
40 | )}x, |
41 | |
41 | |
42 | # shell-like argument quoting, basically always matches |
42 | # shell-like argument quoting, basically always matches |
43 | qr{\G [\ \t|&;<>()] *( |
43 | qr{\G [\ \t|&;<>()]* ( |
44 | (?: |
44 | (?: |
45 | [^\\"'\ \t|&;<>()]+ |
45 | [^\\"'\ \t|&;<>()]+ |
46 | | \\. |
46 | | \\. |
47 | | " (?: [^\\"]+ | \\. )* " |
47 | | " (?: [^\\"]+ | \\. )* " |
48 | | ' [^']* ' |
48 | | ' [^']* ' |
… | |
… | |
66 | my $curlen = $line->offset_of ($self->selection_end) |
66 | my $curlen = $line->offset_of ($self->selection_end) |
67 | - $line->offset_of ($self->selection_beg); |
67 | - $line->offset_of ($self->selection_beg); |
68 | |
68 | |
69 | my @matches; |
69 | my @matches; |
70 | |
70 | |
71 | # not doing matches in unicode mode helps speed |
71 | if ($markofs < $line->l) { |
72 | # enourmously here. working in utf-8 should be |
72 | # convert markofs form character to UTF-8 offset space |
73 | # equivalent due to the magic of utf-8 encoding. |
73 | { |
|
|
74 | my $prefix = substr $text, 0, $markofs; |
74 | utf8::encode $text; |
75 | utf8::encode $prefix; |
75 | study $text; # _really_ helps, too :) |
76 | $markofs = length $prefix; |
|
|
77 | } |
76 | |
78 | |
77 | for my $regex (@mark_patterns, @{ $self->{patterns} }) { |
79 | # not doing matches in unicode mode helps speed |
78 | while ($text =~ /$regex/g) { |
80 | # enourmously here. working in utf-8 should be |
79 | if ($-[1] <= $markofs and $markofs <= $+[1]) { |
81 | # equivalent due to the magic of utf-8 encoding. |
80 | my $ofs = $-[1]; |
82 | utf8::encode $text; |
81 | my $match = $1; |
83 | study $text; # _really_ helps, too :) |
82 | |
84 | |
|
|
85 | for my $regex (@mark_patterns, @{ $self->{patterns} }) { |
|
|
86 | while ($text =~ /$regex/g) { |
|
|
87 | if ($-[1] <= $markofs and $markofs <= $+[1]) { |
|
|
88 | my $ofs = $-[1]; |
|
|
89 | my $match = $1; |
|
|
90 | |
83 | for my $regex (@simplify_patterns) { |
91 | for my $regex (@simplify_patterns) { |
84 | if ($match =~ $regex) { |
92 | if ($match =~ $regex) { |
85 | $match = $1; |
93 | $match = $1; |
86 | $ofs += $-[1]; |
94 | $ofs += $-[1]; |
|
|
95 | } |
87 | } |
96 | } |
|
|
97 | |
|
|
98 | push @matches, [$ofs, length $match]; |
88 | } |
99 | } |
89 | |
|
|
90 | push @matches, [$ofs, length $match]; |
|
|
91 | } |
100 | } |
92 | } |
101 | } |
93 | } |
102 | } |
94 | |
103 | |
95 | # whole line |
104 | # whole line |
… | |
… | |
98 | for (sort { $a->[1] <=> $b->[1] or $b->[0] <=> $a->[0] } @matches) { |
107 | for (sort { $a->[1] <=> $b->[1] or $b->[0] <=> $a->[0] } @matches) { |
99 | my ($ofs, $len) = @$_; |
108 | my ($ofs, $len) = @$_; |
100 | |
109 | |
101 | next if $len <= $curlen; |
110 | next if $len <= $curlen; |
102 | |
111 | |
|
|
112 | # convert back from UTF-8 offset space to character space |
|
|
113 | { |
|
|
114 | my $length = substr "$text ", $ofs, $len; |
|
|
115 | utf8::decode $length; |
|
|
116 | $len = length $length; |
|
|
117 | } |
|
|
118 | { |
|
|
119 | my $prefix = substr $text, 0, $ofs; |
|
|
120 | utf8::decode $prefix; |
|
|
121 | $ofs = length $prefix; |
|
|
122 | } |
|
|
123 | |
103 | $self->selection_beg ($line->coord_of ($ofs)); |
124 | $self->selection_beg ($line->coord_of ($ofs)); |
104 | $self->selection_end ($line->coord_of ($ofs + $len)); |
125 | $self->selection_end ($line->coord_of ($ofs + $len)); |
105 | return 1; |
126 | return 1; |
106 | } |
127 | } |
107 | |
128 | |