1 |
578 |
markom |
# word.tcl --
|
2 |
|
|
#
|
3 |
|
|
# This file defines various procedures for computing word boundaries
|
4 |
|
|
# in strings. This file is primarily needed so Tk text and entry
|
5 |
|
|
# widgets behave properly for different platforms.
|
6 |
|
|
#
|
7 |
|
|
# Copyright (c) 1996 by Sun Microsystems, Inc.
|
8 |
|
|
#
|
9 |
|
|
# See the file "license.terms" for information on usage and redistribution
|
10 |
|
|
# of this file, and for a DISCLAIMER OF ALL WARRANTIES.
|
11 |
|
|
#
|
12 |
|
|
# RCS: @(#) $Id: word.tcl,v 1.1.1.1 2002-01-16 10:25:30 markom Exp $
|
13 |
|
|
#
|
14 |
|
|
# See the file "license.terms" for information on usage and redistribution
|
15 |
|
|
# of this file, and for a DISCLAIMER OF ALL WARRANTIES.
|
16 |
|
|
#
|
17 |
|
|
|
18 |
|
|
# The following variables are used to determine which characters are
|
19 |
|
|
# interpreted as white space.
|
20 |
|
|
|
21 |
|
|
# CYGNUS local: Always use Motif-style selection
|
22 |
|
|
#if {$tcl_platform(platform) == "windows"} {
|
23 |
|
|
# Windows style - any but space, tab, or newline
|
24 |
|
|
# set tcl_wordchars "\[^ \t\n\]"
|
25 |
|
|
# set tcl_nonwordchars "\[ \t\n\]"
|
26 |
|
|
#} else {
|
27 |
|
|
# Motif style - any number, letter, or underscore
|
28 |
|
|
set tcl_wordchars {[a-zA-Z0-9_]}
|
29 |
|
|
set tcl_nonwordchars {[^a-zA-Z0-9_]}
|
30 |
|
|
#}
|
31 |
|
|
|
32 |
|
|
# tcl_wordBreakAfter --
|
33 |
|
|
#
|
34 |
|
|
# This procedure returns the index of the first word boundary
|
35 |
|
|
# after the starting point in the given string, or -1 if there
|
36 |
|
|
# are no more boundaries in the given string. The index returned refers
|
37 |
|
|
# to the first character of the pair that comprises a boundary.
|
38 |
|
|
#
|
39 |
|
|
# Arguments:
|
40 |
|
|
# str - String to search.
|
41 |
|
|
# start - Index into string specifying starting point.
|
42 |
|
|
|
43 |
|
|
proc tcl_wordBreakAfter {str start} {
|
44 |
|
|
global tcl_nonwordchars tcl_wordchars
|
45 |
|
|
set str [string range $str $start end]
|
46 |
|
|
if {[regexp -indices "$tcl_wordchars$tcl_nonwordchars|$tcl_nonwordchars$tcl_wordchars" $str result]} {
|
47 |
|
|
return [expr {[lindex $result 1] + $start}]
|
48 |
|
|
}
|
49 |
|
|
return -1
|
50 |
|
|
}
|
51 |
|
|
|
52 |
|
|
# tcl_wordBreakBefore --
|
53 |
|
|
#
|
54 |
|
|
# This procedure returns the index of the first word boundary
|
55 |
|
|
# before the starting point in the given string, or -1 if there
|
56 |
|
|
# are no more boundaries in the given string. The index returned
|
57 |
|
|
# refers to the second character of the pair that comprises a boundary.
|
58 |
|
|
#
|
59 |
|
|
# Arguments:
|
60 |
|
|
# str - String to search.
|
61 |
|
|
# start - Index into string specifying starting point.
|
62 |
|
|
|
63 |
|
|
proc tcl_wordBreakBefore {str start} {
|
64 |
|
|
global tcl_nonwordchars tcl_wordchars
|
65 |
|
|
if {[string compare $start end] == 0} {
|
66 |
|
|
set start [string length $str]
|
67 |
|
|
}
|
68 |
|
|
if {[regexp -indices "^.*($tcl_wordchars$tcl_nonwordchars|$tcl_nonwordchars$tcl_wordchars)" [string range $str 0 $start] result]} {
|
69 |
|
|
return [lindex $result 1]
|
70 |
|
|
}
|
71 |
|
|
return -1
|
72 |
|
|
}
|
73 |
|
|
|
74 |
|
|
# tcl_endOfWord --
|
75 |
|
|
#
|
76 |
|
|
# This procedure returns the index of the first end-of-word location
|
77 |
|
|
# after a starting index in the given string. An end-of-word location
|
78 |
|
|
# is defined to be the first whitespace character following the first
|
79 |
|
|
# non-whitespace character after the starting point. Returns -1 if
|
80 |
|
|
# there are no more words after the starting point.
|
81 |
|
|
#
|
82 |
|
|
# Arguments:
|
83 |
|
|
# str - String to search.
|
84 |
|
|
# start - Index into string specifying starting point.
|
85 |
|
|
|
86 |
|
|
proc tcl_endOfWord {str start} {
|
87 |
|
|
global tcl_nonwordchars tcl_wordchars
|
88 |
|
|
if {[regexp -indices "$tcl_nonwordchars*$tcl_wordchars+$tcl_nonwordchars" \
|
89 |
|
|
[string range $str $start end] result]} {
|
90 |
|
|
return [expr {[lindex $result 1] + $start}]
|
91 |
|
|
}
|
92 |
|
|
return -1
|
93 |
|
|
}
|
94 |
|
|
|
95 |
|
|
# tcl_startOfNextWord --
|
96 |
|
|
#
|
97 |
|
|
# This procedure returns the index of the first start-of-word location
|
98 |
|
|
# after a starting index in the given string. A start-of-word
|
99 |
|
|
# location is defined to be a non-whitespace character following a
|
100 |
|
|
# whitespace character. Returns -1 if there are no more start-of-word
|
101 |
|
|
# locations after the starting point.
|
102 |
|
|
#
|
103 |
|
|
# Arguments:
|
104 |
|
|
# str - String to search.
|
105 |
|
|
# start - Index into string specifying starting point.
|
106 |
|
|
|
107 |
|
|
proc tcl_startOfNextWord {str start} {
|
108 |
|
|
global tcl_nonwordchars tcl_wordchars
|
109 |
|
|
if {[regexp -indices "$tcl_wordchars*$tcl_nonwordchars+$tcl_wordchars" \
|
110 |
|
|
[string range $str $start end] result]} {
|
111 |
|
|
return [expr {[lindex $result 1] + $start}]
|
112 |
|
|
}
|
113 |
|
|
return -1
|
114 |
|
|
}
|
115 |
|
|
|
116 |
|
|
# tcl_startOfPreviousWord --
|
117 |
|
|
#
|
118 |
|
|
# This procedure returns the index of the first start-of-word location
|
119 |
|
|
# before a starting index in the given string.
|
120 |
|
|
#
|
121 |
|
|
# Arguments:
|
122 |
|
|
# str - String to search.
|
123 |
|
|
# start - Index into string specifying starting point.
|
124 |
|
|
|
125 |
|
|
proc tcl_startOfPreviousWord {str start} {
|
126 |
|
|
global tcl_nonwordchars tcl_wordchars
|
127 |
|
|
if {[string compare $start end] == 0} {
|
128 |
|
|
set start [string length $str]
|
129 |
|
|
}
|
130 |
|
|
if {[regexp -indices \
|
131 |
|
|
"$tcl_nonwordchars*($tcl_wordchars+)$tcl_nonwordchars*\$" \
|
132 |
|
|
[string range $str 0 [expr {$start - 1}]] result word]} {
|
133 |
|
|
return [lindex $word 0]
|
134 |
|
|
}
|
135 |
|
|
return -1
|
136 |
|
|
}
|