| … | |
… | |
| 7 | and $self->selection (map { y/A-Za-z/N-ZA-Mn-za-m/; $_ } $self->selection); |
7 | and $self->selection (map { y/A-Za-z/N-ZA-Mn-za-m/; $_ } $self->selection); |
| 8 | |
8 | |
| 9 | () |
9 | () |
| 10 | } |
10 | } |
| 11 | |
11 | |
|
|
12 | sub on_init { |
|
|
13 | my ($self) = @_; |
|
|
14 | |
|
|
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 = $self->locale_decode ($res); |
|
|
18 | utf8::encode $res; |
|
|
19 | push @{ $self->{patterns} }, qr/$res/; |
|
|
20 | } |
|
|
21 | |
|
|
22 | () |
|
|
23 | } |
|
|
24 | |
| 12 | # "find interetsing things"-patterns |
25 | # "find interetsing things"-patterns |
| 13 | my @mark_patterns = ( |
26 | my @mark_patterns = ( |
| 14 | qr{([[:alnum:]]+)}, |
|
|
| 15 | |
|
|
| 16 | # common "parentheses" |
27 | # common types of "parentheses" |
| 17 | qr{(?:^|\s) ‘ ([^‘’]+?) ’ (?:\s|\)|$)}x, |
28 | qr{ (?<![^[:space:]]) ‘ ([^‘’]+) ’ (?![^[:space]]) }x, |
| 18 | qr{(?:^|\s) ` ([^`']+?) ' (?:\s|\)|$)}x, |
29 | qr{ (?<![^[:space:]]) ` ([^`']+) ' (?![^[:space]]) }x, |
| 19 | qr{ \{ ([^{}]+?) \} }x, |
30 | qr{ (?<![^[:space:]]) (" [^[:space:]] [^"]* ") }x, |
| 20 | qr{ \[ ([^{}]+?) \] }x, |
31 | qr{ (" [^"]* [^[:space:]] ") (?![^[:space]]) }x, |
|
|
32 | qr{ \< ([^<>[:space:]]+) \> }x, |
|
|
33 | qr{ \{ ([^{}[:space:]]+) \} }x, |
|
|
34 | qr{ \[ ([^{}[:space:]]+) \] }x, |
|
|
35 | qr{ \( ([^()[:space:]]+) \) }x, |
| 21 | |
36 | |
| 22 | # urls, just a heuristic |
37 | # urls, just a heuristic |
| 23 | qr{( |
38 | qr{( |
| 24 | (?:https?|ftp|news|mailto|file)://[ab-zA-Z0-9\-\@;\/?:&=%\$_.+!*\x27(),~]+ |
39 | (?:https?|ftp|news|mailto|file)://[ab-zA-Z0-9\-\@;\/?:&=%\$_.+!*\x27(),~]+ |
| 25 | [ab-zA-Z0-9\-\@;\/?:&=%\$_+!*\x27()~] # exclude some trailing characters (heuristic) |
40 | [ab-zA-Z0-9\-\@;\/?:&=%\$_+!*\x27()~] # exclude some trailing characters (heuristic) |
| … | |
… | |
| 52 | my $curlen = $line->offset_of ($self->selection_end) |
67 | my $curlen = $line->offset_of ($self->selection_end) |
| 53 | - $line->offset_of ($self->selection_beg); |
68 | - $line->offset_of ($self->selection_beg); |
| 54 | |
69 | |
| 55 | my @matches; |
70 | my @matches; |
| 56 | |
71 | |
| 57 | for my $regex (@mark_patterns) { |
72 | # not doing matches in unicode mode helps speed |
|
|
73 | # enourmously here. working in utf-8 should be |
|
|
74 | # equivalent due to the magic of utf-8 encoding. |
|
|
75 | utf8::encode $text; |
|
|
76 | study $text; # _really_ helps, too :) |
|
|
77 | |
|
|
78 | for my $regex (@mark_patterns, @{ $self->{patterns} }) { |
| 58 | while ($text =~ /$regex/g) { |
79 | while ($text =~ /$regex/g) { |
| 59 | if ($-[1] <= $markofs and $markofs <= $+[1]) { |
80 | if ($-[1] <= $markofs and $markofs <= $+[1]) { |
| 60 | my $ofs = $-[1]; |
81 | my $ofs = $-[1]; |
| 61 | my $match = $1; |
82 | my $match = $1; |
| 62 | |
|
|
| 63 | push @matches, [$ofs, length $match]; |
|
|
| 64 | |
83 | |
| 65 | for my $regex (@simplify_patterns) { |
84 | for my $regex (@simplify_patterns) { |
| 66 | if ($match =~ $regex) { |
85 | if ($match =~ $regex) { |
| 67 | $match = $1; |
86 | $match = $1; |
| 68 | $ofs += $-[1]; |
87 | $ofs += $-[1]; |
