1 | #! perl |
1 | #! perl |
|
|
2 | |
|
|
3 | #:META:RESOURCE:%.pattern-0:string:first selection pattern |
|
|
4 | |
|
|
5 | =head1 NAME |
|
|
6 | |
|
|
7 | selection - more intelligent selection |
|
|
8 | |
|
|
9 | =head1 DESCRIPTION |
|
|
10 | |
|
|
11 | This extension tries to be more intelligent when the user extends |
|
|
12 | selections (double-click and further clicks). Right now, it tries to |
|
|
13 | select words, urls and complete shell-quoted arguments, which is very |
|
|
14 | convenient, too, if your F<ls> supports C<--quoting-style=shell>. |
|
|
15 | |
|
|
16 | A double-click usually selects the word under the cursor, further clicks |
|
|
17 | will enlarge the selection. |
|
|
18 | |
|
|
19 | The selection works by trying to match a number of regexes and displaying |
|
|
20 | them in increasing order of length. You can add your own regexes by |
|
|
21 | specifying resources of the form: |
|
|
22 | |
|
|
23 | URxvt.selection.pattern-0: perl-regex |
|
|
24 | URxvt.selection.pattern-1: perl-regex |
|
|
25 | ... |
|
|
26 | |
|
|
27 | The index number (0, 1...) must not have any holes, and each regex must |
|
|
28 | contain at least one pair of capturing parentheses, which will be used for |
|
|
29 | the match. For example, the following adds a regex that matches everything |
|
|
30 | between two vertical bars: |
|
|
31 | |
|
|
32 | URxvt.selection.pattern-0: \\|([^|]+)\\| |
|
|
33 | |
|
|
34 | Another example: Programs I use often output "absolute path: " at the |
|
|
35 | beginning of a line when they process multiple files. The following |
|
|
36 | pattern matches the filename (note, there is a single space at the very |
|
|
37 | end): |
|
|
38 | |
|
|
39 | URxvt.selection.pattern-0: ^(/[^:]+):\ |
|
|
40 | |
|
|
41 | You can look at the source of the selection extension to see more |
|
|
42 | interesting uses, such as parsing a line from beginning to end. |
|
|
43 | |
|
|
44 | This extension also offers the following actions: |
|
|
45 | |
|
|
46 | =over 4 |
|
|
47 | |
|
|
48 | =item rot13 |
|
|
49 | |
|
|
50 | Rot-13 the selection when activated. |
|
|
51 | |
|
|
52 | Example: |
|
|
53 | |
|
|
54 | URxvt.keysym.C-M-r: selection:rot13 |
|
|
55 | |
|
|
56 | =back |
|
|
57 | |
|
|
58 | =cut |
2 | |
59 | |
3 | sub on_user_command { |
60 | sub on_user_command { |
4 | my ($self, $cmd) = @_; |
61 | my ($self, $cmd) = @_; |
5 | |
62 | |
6 | $cmd eq "selection:rot13" |
63 | $cmd eq "selection:rot13" |
|
|
64 | and $self->selection (map { y/A-Za-z/N-ZA-Mn-za-m/; $_ } $self->selection); |
|
|
65 | |
|
|
66 | () |
|
|
67 | } |
|
|
68 | |
|
|
69 | sub on_action { |
|
|
70 | my ($self, $action) = @_; |
|
|
71 | |
|
|
72 | $action eq "rot13" |
7 | and $self->selection (map { y/A-Za-z/N-ZA-Mn-za-m/; $_ } $self->selection); |
73 | and $self->selection (map { y/A-Za-z/N-ZA-Mn-za-m/; $_ } $self->selection); |
8 | |
74 | |
9 | () |
75 | () |
10 | } |
76 | } |
11 | |
77 | |
… | |
… | |
17 | push @{ $self->{patterns} }, qr{\G [\Q$res\E[:space:]]* ([^\Q$res\E[:space:]]+) }x; |
83 | push @{ $self->{patterns} }, qr{\G [\Q$res\E[:space:]]* ([^\Q$res\E[:space:]]+) }x; |
18 | } |
84 | } |
19 | |
85 | |
20 | for (my $idx = 0; defined (my $res = $self->x_resource ("selection.pattern-$idx")); $idx++) { |
86 | for (my $idx = 0; defined (my $res = $self->x_resource ("selection.pattern-$idx")); $idx++) { |
21 | $res = $self->locale_decode ($res); |
87 | $res = $self->locale_decode ($res); |
22 | utf8::encode $res; |
|
|
23 | push @{ $self->{patterns} }, qr/$res/; |
88 | push @{ $self->{patterns} }, qr/$res/; |
24 | } |
89 | } |
25 | |
90 | |
26 | $self->{enabled} = 1; |
91 | $self->{enabled} = 1; |
27 | |
92 | |
… | |
… | |
32 | () |
97 | () |
33 | } |
98 | } |
34 | |
99 | |
35 | # "find interesting things"-patterns |
100 | # "find interesting things"-patterns |
36 | my @mark_patterns = ( |
101 | my @mark_patterns = ( |
37 | qr{ ([[:word:]]+) }x, |
102 | # qr{ ([[:word:]]+) }x, |
38 | qr{ ([^[:space:]]+) }x, |
103 | qr{ ([^[:space:]]+) }x, |
39 | |
104 | |
40 | # common types of "parentheses" |
105 | # common types of "parentheses" |
41 | qr{ (?<![^[:space:]]) ` ([^‘’]+) ` (?![^[:space:]]) }x, |
106 | qr{ (?<![^[:space:]]) [`'] ([^`']+) [`'] (?![^[:space:]]) }x, |
42 | qr{ (?<![^[:space:]]) ‘ ([^‘’]+) ’ (?![^[:space:]]) }x, |
107 | qr{ (?<![^[:space:]]) ‘ ([^‘’]+) ’ (?![^[:space:]]) }x, |
43 | qr{ (?<![^[:space:]]) ` ([^`']+) ' (?![^[:space:]]) }x, |
108 | qr{ (?<![^[:space:]]) “ ([^“”]+) ” (?![^[:space:]]) }x, |
44 | |
109 | |
45 | qr{ (?<![^[:space:]]) ‘ ([^`']+) ’ (?![^[:space:]]) }x, |
|
|
46 | qr{ (?<![^[:space:]]) “ ([^`']+) ” (?![^[:space:]]) }x, |
|
|
47 | |
|
|
48 | qr{ (?<![^[:space:]]) (' [^[:space:]] [^']* ') }x, |
110 | qr{ (?<![^[:space:]]) (' [^[:space:]] [^']* ') }x, |
49 | qr{ (' [^']* [^[:space:]] ') (?![^[:space:]]) }x, |
111 | qr{ (' [^']* [^[:space:]] ') (?![^[:space:]]) }x, |
|
|
112 | qr{ (?<![^[:space:]]) (` [^[:space:]] [^']* ') }x, |
|
|
113 | qr{ (` [^']* [^[:space:]] ') (?![^[:space:]]) }x, |
50 | qr{ (?<![^[:space:]]) (" [^[:space:]] [^"]* ") }x, |
114 | qr{ (?<![^[:space:]]) (" [^[:space:]] [^"]* ") }x, |
51 | qr{ (" [^"]* [^[:space:]] ") (?![^[:space:]]) }x, |
115 | qr{ (" [^"]* [^[:space:]] ") (?![^[:space:]]) }x, |
52 | |
116 | |
53 | qr{ \< ([^>]+) \> }x, |
|
|
54 | qr{ \( ([^)]+) \) }x, |
|
|
55 | qr{ \{ ([^}]+) \} }x, |
117 | qr{ \{ ([^\{\}]+) \} }x, |
|
|
118 | qr{ \( ([^\(\)]+) \) }x, |
56 | qr{ \[ ([^]]+) \] }x, |
119 | qr{ \[ ([^\[\]]+) \] }x, |
|
|
120 | qr{ \< ([^\<\>]+) \> }x, |
57 | |
121 | |
58 | # urls, just a heuristic |
122 | # urls, just a heuristic |
59 | qr{( |
123 | qr{( |
60 | (?:https?://|ftp://|news://|mailto:|file://|\bwww\.)[ab-zA-Z0-9\-\@;\/?:&=%\$_.+!*\x27(),~#]+ |
124 | (?:https?://|ftp://|news://|mailto:|file://|\bwww\.)[ab-zA-Z0-9\-\@;\/?:&=%\$_.+!*\x27(),~#]+ |
61 | [ab-zA-Z0-9\-\@;\/?:&=%\$_+*()~] # exclude some trailing characters (heuristic) |
125 | [ab-zA-Z0-9\-\@;\/?:&=%\$_+*()~] # exclude some trailing characters (heuristic) |
… | |
… | |
92 | - $line->offset_of ($self->selection_beg); |
156 | - $line->offset_of ($self->selection_beg); |
93 | |
157 | |
94 | my @matches; |
158 | my @matches; |
95 | |
159 | |
96 | if ($markofs < $line->l) { |
160 | if ($markofs < $line->l) { |
97 | # convert markofs from character to UTF-8 offset space |
|
|
98 | { |
|
|
99 | my $prefix = substr $text, 0, $markofs; |
|
|
100 | utf8::encode $prefix; |
|
|
101 | $markofs = length $prefix; |
|
|
102 | } |
|
|
103 | |
|
|
104 | # not doing matches in unicode mode helps speed |
|
|
105 | # enourmously here. working in utf-8 should be |
|
|
106 | # equivalent due to the magic of utf-8 encoding. |
|
|
107 | utf8::encode $text; |
|
|
108 | study $text; # _really_ helps, too :) |
161 | study $text; # _really_ helps, too :) |
109 | |
162 | |
110 | for my $regex (@mark_patterns, @{ $self->{patterns} }) { |
163 | for my $regex (@mark_patterns, @{ $self->{patterns} }) { |
111 | while ($text =~ /$regex/g) { |
164 | while ($text =~ /$regex/g) { |
112 | if ($-[1] <= $markofs and $markofs <= $+[1]) { |
165 | if ($-[1] <= $markofs and $markofs <= $+[1]) { |
… | |
… | |
132 | for (sort { $a->[1] <=> $b->[1] or $b->[0] <=> $a->[0] } @matches) { |
185 | for (sort { $a->[1] <=> $b->[1] or $b->[0] <=> $a->[0] } @matches) { |
133 | my ($ofs, $len) = @$_; |
186 | my ($ofs, $len) = @$_; |
134 | |
187 | |
135 | next if $len <= $curlen; |
188 | next if $len <= $curlen; |
136 | |
189 | |
137 | # convert back from UTF-8 offset space to character space |
|
|
138 | { |
|
|
139 | my $length = substr "$text ", $ofs, $len; |
|
|
140 | utf8::decode $length; |
|
|
141 | $len = length $length; |
|
|
142 | } |
|
|
143 | { |
|
|
144 | my $prefix = substr $text, 0, $ofs; |
|
|
145 | utf8::decode $prefix; |
|
|
146 | $ofs = length $prefix; |
|
|
147 | } |
|
|
148 | |
|
|
149 | $self->selection_beg ($line->coord_of ($ofs)); |
190 | $self->selection_beg ($line->coord_of ($ofs)); |
150 | $self->selection_end ($line->coord_of ($ofs + $len)); |
191 | $self->selection_end ($line->coord_of ($ofs + $len)); |
151 | return 1; |
192 | return 1; |
152 | } |
193 | } |
153 | |
194 | |