1 |
root |
1.4 |
#! perl |
2 |
|
|
|
3 |
elmex |
1.1 |
sub on_keyboard_command { |
4 |
root |
1.7 |
my ($self, $cmd) = @_; |
5 |
elmex |
1.2 |
|
6 |
|
|
$cmd eq "selection:rot13" |
7 |
root |
1.7 |
and $self->selection (map { y/A-Za-z/N-ZA-Mn-za-m/; $_ } $self->selection); |
8 |
elmex |
1.3 |
|
9 |
|
|
() |
10 |
elmex |
1.1 |
} |
11 |
root |
1.5 |
|
12 |
root |
1.23 |
sub on_init { |
13 |
|
|
my ($self) = @_; |
14 |
|
|
|
15 |
|
|
for (my $idx = 0; defined (my $res = $self->x_resource ("selection.pattern-$idx")); $idx++) { |
16 |
root |
1.31 |
$res = $self->locale_decode ($res); |
17 |
|
|
utf8::encode $res; |
18 |
root |
1.23 |
push @{ $self->{patterns} }, qr/$res/; |
19 |
|
|
} |
20 |
|
|
|
21 |
|
|
() |
22 |
|
|
} |
23 |
|
|
|
24 |
root |
1.34 |
# "find interesting things"-patterns |
25 |
root |
1.15 |
my @mark_patterns = ( |
26 |
root |
1.24 |
# common types of "parentheses" |
27 |
root |
1.25 |
qr{ (?<![^[:space:]]) ‘ ([^‘’]+) ’ (?![^[:space]]) }x, |
28 |
|
|
qr{ (?<![^[:space:]]) ` ([^`']+) ' (?![^[:space]]) }x, |
29 |
|
|
qr{ (?<![^[:space:]]) (" [^[:space:]] [^"]* ") }x, |
30 |
|
|
qr{ (" [^"]* [^[:space:]] ") (?![^[:space]]) }x, |
31 |
|
|
qr{ \< ([^<>[:space:]]+) \> }x, |
32 |
|
|
qr{ \{ ([^{}[:space:]]+) \} }x, |
33 |
|
|
qr{ \[ ([^{}[:space:]]+) \] }x, |
34 |
|
|
qr{ \( ([^()[:space:]]+) \) }x, |
35 |
root |
1.14 |
|
36 |
root |
1.20 |
# urls, just a heuristic |
37 |
|
|
qr{( |
38 |
|
|
(?:https?|ftp|news|mailto|file)://[ab-zA-Z0-9\-\@;\/?:&=%\$_.+!*\x27(),~]+ |
39 |
|
|
[ab-zA-Z0-9\-\@;\/?:&=%\$_+!*\x27()~] # exclude some trailing characters (heuristic) |
40 |
|
|
)}x, |
41 |
|
|
|
42 |
root |
1.17 |
# shell-like argument quoting, basically always matches |
43 |
root |
1.34 |
qr{\G [\ \t|&;<>()]* ( |
44 |
root |
1.9 |
(?: |
45 |
root |
1.11 |
[^\\"'\ \t|&;<>()]+ |
46 |
root |
1.9 |
| \\. |
47 |
root |
1.13 |
| " (?: [^\\"]+ | \\. )* " |
48 |
root |
1.9 |
| ' [^']* ' |
49 |
|
|
)+ |
50 |
root |
1.14 |
)}x, |
51 |
root |
1.8 |
); |
52 |
|
|
|
53 |
root |
1.16 |
# "correct obvious? crap"-patterns |
54 |
root |
1.15 |
my @simplify_patterns = ( |
55 |
|
|
qr{^"([^\\"'\ \t|&;<>()*?]+)"$}, # "simple" => simple |
56 |
root |
1.16 |
qr{^(.*)[,\-]$}, # strip off trailing , and - |
57 |
root |
1.15 |
); |
58 |
root |
1.14 |
|
59 |
root |
1.6 |
sub on_sel_extend { |
60 |
root |
1.21 |
my ($self, $time) = @_; |
61 |
root |
1.8 |
|
62 |
|
|
my ($row, $col) = $self->selection_mark; |
63 |
|
|
my $line = $self->line ($row); |
64 |
|
|
my $text = $line->t; |
65 |
root |
1.20 |
my $markofs = $line->offset_of ($row, $col); |
66 |
|
|
my $curlen = $line->offset_of ($self->selection_end) |
67 |
|
|
- $line->offset_of ($self->selection_beg); |
68 |
|
|
|
69 |
|
|
my @matches; |
70 |
root |
1.8 |
|
71 |
root |
1.32 |
if ($markofs < $line->l) { |
72 |
|
|
# convert markofs form character to UTF-8 offset space |
73 |
|
|
{ |
74 |
|
|
my $prefix = substr $text, 0, $markofs; |
75 |
|
|
utf8::encode $prefix; |
76 |
|
|
$markofs = length $prefix; |
77 |
|
|
} |
78 |
|
|
|
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 :) |
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 |
|
|
|
91 |
|
|
for my $regex (@simplify_patterns) { |
92 |
|
|
if ($match =~ $regex) { |
93 |
|
|
$match = $1; |
94 |
|
|
$ofs += $-[1]; |
95 |
|
|
} |
96 |
root |
1.15 |
} |
97 |
root |
1.32 |
|
98 |
|
|
push @matches, [$ofs, length $match]; |
99 |
root |
1.15 |
} |
100 |
root |
1.8 |
} |
101 |
|
|
} |
102 |
|
|
} |
103 |
|
|
|
104 |
root |
1.21 |
# whole line |
105 |
|
|
push @matches, [0, ($line->end - $line->beg + 1) * $self->ncol]; |
106 |
|
|
|
107 |
root |
1.20 |
for (sort { $a->[1] <=> $b->[1] or $b->[0] <=> $a->[0] } @matches) { |
108 |
|
|
my ($ofs, $len) = @$_; |
109 |
|
|
|
110 |
|
|
next if $len <= $curlen; |
111 |
|
|
|
112 |
root |
1.32 |
# convert back from UTF-8 offset space to character space |
113 |
|
|
{ |
114 |
root |
1.33 |
my $length = substr "$text ", $ofs, $len; |
115 |
root |
1.32 |
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 |
|
|
|
124 |
root |
1.20 |
$self->selection_beg ($line->coord_of ($ofs)); |
125 |
|
|
$self->selection_end ($line->coord_of ($ofs + $len)); |
126 |
|
|
return 1; |
127 |
|
|
} |
128 |
|
|
|
129 |
root |
1.21 |
() |
130 |
root |
1.5 |
} |