… | |
… | |
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 |
16 | $res = $self->locale_decode ($res); |
|
|
17 | utf8::encode $res; |
17 | push @{ $self->{patterns} }, qr/$res/; |
18 | push @{ $self->{patterns} }, qr/$res/; |
18 | } |
19 | } |
19 | |
20 | |
20 | () |
21 | () |
21 | } |
22 | } |
22 | |
23 | |
23 | # "find interetsing things"-patterns |
24 | # "find interesting things"-patterns |
24 | my @mark_patterns = ( |
25 | my @mark_patterns = ( |
25 | # common types of "parentheses" |
26 | # common types of "parentheses" |
26 | qr{(?:^|\s) ‘ ([^‘’]+?) ’ (?:\s|\)|$)}x, |
27 | qr{ (?<![^[:space:]]) ‘ ([^‘’]+) ’ (?![^[:space]]) }x, |
27 | qr{(?:^|\s) ` ([^`']+?) ' (?:\s|\)|$)}x, |
28 | qr{ (?<![^[:space:]]) ` ([^`']+) ' (?![^[:space]]) }x, |
28 | qr{ \{ ([^{}]+?) \} }x, |
29 | qr{ (?<![^[:space:]]) (" [^[:space:]] [^"]* ") }x, |
29 | qr{ \[ ([^{}]+?) \] }x, |
30 | qr{ (" [^"]* [^[:space:]] ") (?![^[:space]]) }x, |
30 | qr{ \( ([^()]+?) \) }x, |
31 | qr{ \< ([^<>[:space:]]+) \> }x, |
|
|
32 | qr{ \{ ([^{}[:space:]]+) \} }x, |
|
|
33 | qr{ \[ ([^{}[:space:]]+) \] }x, |
|
|
34 | qr{ \( ([^()[:space:]]+) \) }x, |
31 | |
35 | |
32 | # urls, just a heuristic |
36 | # urls, just a heuristic |
33 | qr{( |
37 | qr{( |
34 | (?:https?|ftp|news|mailto|file)://[ab-zA-Z0-9\-\@;\/?:&=%\$_.+!*\x27(),~]+ |
38 | (?:https?|ftp|news|mailto|file)://[ab-zA-Z0-9\-\@;\/?:&=%\$_.+!*\x27(),~]+ |
35 | [ab-zA-Z0-9\-\@;\/?:&=%\$_+!*\x27()~] # exclude some trailing characters (heuristic) |
39 | [ab-zA-Z0-9\-\@;\/?:&=%\$_+!*\x27()~] # exclude some trailing characters (heuristic) |
36 | )}x, |
40 | )}x, |
37 | |
41 | |
38 | # shell-like argument quoting, basically always matches |
42 | # shell-like argument quoting, basically always matches |
39 | qr{\G [\ \t|&;<>()] *( |
43 | qr{\G [\ \t|&;<>()]* ( |
40 | (?: |
44 | (?: |
41 | [^\\"'\ \t|&;<>()]+ |
45 | [^\\"'\ \t|&;<>()]+ |
42 | | \\. |
46 | | \\. |
43 | | " (?: [^\\"]+ | \\. )* " |
47 | | " (?: [^\\"]+ | \\. )* " |
44 | | ' [^']* ' |
48 | | ' [^']* ' |
… | |
… | |
62 | my $curlen = $line->offset_of ($self->selection_end) |
66 | my $curlen = $line->offset_of ($self->selection_end) |
63 | - $line->offset_of ($self->selection_beg); |
67 | - $line->offset_of ($self->selection_beg); |
64 | |
68 | |
65 | my @matches; |
69 | my @matches; |
66 | |
70 | |
67 | for my $regex (@mark_patterns, @{ $self->{patterns} }) { |
71 | if ($markofs < $line->l) { |
68 | while ($text =~ /$regex/g) { |
72 | # convert markofs form character to UTF-8 offset space |
69 | if ($-[1] <= $markofs and $markofs <= $+[1]) { |
73 | { |
70 | my $ofs = $-[1]; |
74 | my $prefix = substr $text, 0, $markofs; |
71 | my $match = $1; |
75 | utf8::encode $prefix; |
|
|
76 | $markofs = length $prefix; |
|
|
77 | } |
72 | |
78 | |
73 | push @matches, [$ofs, length $match]; |
79 | # not doing matches in unicode mode helps speed |
|
|
80 | # enourmously here. working in utf-8 should be |
|
|
81 | # equivalent due to the magic of utf-8 encoding. |
|
|
82 | utf8::encode $text; |
|
|
83 | study $text; # _really_ helps, too :) |
74 | |
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 | |
75 | for my $regex (@simplify_patterns) { |
91 | for my $regex (@simplify_patterns) { |
76 | if ($match =~ $regex) { |
92 | if ($match =~ $regex) { |
77 | $match = $1; |
93 | $match = $1; |
78 | $ofs += $-[1]; |
94 | $ofs += $-[1]; |
|
|
95 | } |
79 | } |
96 | } |
|
|
97 | |
|
|
98 | push @matches, [$ofs, length $match]; |
80 | } |
99 | } |
81 | |
|
|
82 | push @matches, [$ofs, length $match]; |
|
|
83 | } |
100 | } |
84 | } |
101 | } |
85 | } |
102 | } |
86 | |
103 | |
87 | # whole line |
104 | # whole line |
… | |
… | |
90 | for (sort { $a->[1] <=> $b->[1] or $b->[0] <=> $a->[0] } @matches) { |
107 | for (sort { $a->[1] <=> $b->[1] or $b->[0] <=> $a->[0] } @matches) { |
91 | my ($ofs, $len) = @$_; |
108 | my ($ofs, $len) = @$_; |
92 | |
109 | |
93 | next if $len <= $curlen; |
110 | next if $len <= $curlen; |
94 | |
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 | |
95 | $self->selection_beg ($line->coord_of ($ofs)); |
124 | $self->selection_beg ($line->coord_of ($ofs)); |
96 | $self->selection_end ($line->coord_of ($ofs + $len)); |
125 | $self->selection_end ($line->coord_of ($ofs + $len)); |
97 | return 1; |
126 | return 1; |
98 | } |
127 | } |
99 | |
128 | |