1Text::Shellwords::CursoUrs(e3r)Contributed Perl DocumentTaetxito:n:Shellwords::Cursor(3)
2
6 Text::Shellwords::Cursor - Parse a string into tokens
7
9 use Text::Shellwords::Cursor;
10 my $parser = Text::Shellwords::Cursor->new();
11 my $str = 'ab cdef "ghi" j"k\"l "';
12 my ($tok1) = $parser->parse_line($str);
13 $tok1 = ['ab', 'cdef', 'ghi', 'j', 'k"l ']
14 my ($tok2, $tokno, $tokoff) = $parser->parse_line($str, cursorpos => 6);
15 as above, but $tokno=1, $tokoff=3 (under the 'f')
16 DESCRIPTION
18 This module is very similar to Text::Shellwords and Text::ParseWords.
20 However, it has one very significant difference: it keeps track of a
21 character position in the line it's parsing. For instance, if you pass
22 it ("zq fmgb", cursorpos=>6), it would return (['zq', 'fmgb'], 1, 3).
23 The cursorpos parameter tells where in the input string the cursor
24 resides (just before the 'b'), and the result tells you that the cursor
25 was on token 1 ('fmgb'), character 3 ('b'). This is very useful when
26 computing command-line completions involving quoting, escaping, and
27 tokenizing characters (like '(' or '=').
28 A few helper utilities are included as well. You can escape a string
30 to ensure that parsing it will produce the original string
31 (parse_escape). You can also reassemble the tokens with a visually
32 pleasing amount of whitespace between them (join_line).
33 This module started out as an integral part of Term::GDBUI using code
35 loosely based on Text::ParseWords. However, it is now basically a
36 ground-up reimplementation. It was split out of Term::GDBUI for
37 version 0.8.
38
40 new
41 Creates a new parser. Takes named arguments on the command line.
42 keep_quotes
44 Normally all unescaped, unnecessary quote marks are stripped.
45 If you specify "keep_quotes=>1", however, they are preserved.
46 This is useful if you need to know whether the string was quoted
47 or not (string constants) or what type of quotes was around it
48 (affecting variable interpolation, for instance).
49 token_chars
51 This argument specifies the characters that should be considered
52 tokens all by themselves. For instance, if I pass
53 token_chars=>'=', then 'ab=123' would be parsed to ('ab', '=',
54 '123'). Without token_chars, 'ab=123' remains a single string.
55 NOTE: you cannot change token_chars after the constructor has
57 been called! The regexps that use it are compiled once (m//o).
58 Also, until the Gnu Readline library can accept "=[]," without
59 diving into an endless loop, we will not tell history expansion
60 to use token_chars (it uses " \t\n()<>;&|" by default).
61 debug
63 Turns on rather copious debugging to try to show what the parser
64 is thinking at every step.
65 space_none
67 space_before
68 space_after
69 These variables affect how whitespace in the line is normalized
70 and it is reassembled into a string. See the join_line routine.
71 error
73 This is a reference to a routine that should be called to
74 display a parse error. The routine takes two arguments: a
75 reference to the parser, and the error message to display as a
76 string.
77 parsebail(msg)
79 If the parsel routine or any of its subroutines runs into a
80 fatal error, they call parsebail to present a very descriptive
81 diagnostic.
82 parsel
84 This is the heinous routine that actually does the parsing. You
85 should never need to call it directly. Call parse_line instead.
86 parse_line(line, named args)
88 This is the entrypoint to this module's parsing functionality.
89 It converts a line into tokens, respecting quoted text, escaped
90 characters, etc. It also keeps track of a cursor position on
91 the input text, returning the token number and offset within the
92 token where that position can be found in the output.
93 This routine originally bore some resemblance to
95 Text::ParseWords. It has changed almost completely, however, to
96 support keeping track of the cursor position. It also has nicer
97 failure modes, modular quoting, token characters (see
98 token_chars in "new"), etc. This routine now does much more.
99 Arguments:
101 line
103 This is a string containing the command-line to parse.
104 This routine also accepts the following named parameters:
106 cursorpos
108 This is the character position in the line to keep track of.
109 Pass undef (by not specifying it) or the empty string to have
110 the line processed with cursorpos ignored.
111 Note that passing undef is not the same as passing some
113 random number and ignoring the result! For instance, if you
114 pass 0 and the line begins with whitespace, you'll get a
115 0-length token at the beginning of the line to represent the
116 cursor in the middle of the whitespace. This allows command
117 completion to work even when the cursor is not near any
118 tokens. If you pass undef, all whitespace at the beginning
119 and end of the line will be trimmed as you would expect.
120 If it is ambiguous whether the cursor should belong to the
122 previous token or to the following one (i.e. if it's between
123 two quoted strings, say "a""b" or a token_char), it always
124 gravitates to the previous token. This makes more sense when
125 completing.
126 fixclosequote
128 Sometimes you want to try to recover from a missing close
129 quote (for instance, when calculating completions), but
130 usually you want a missing close quote to be a fatal error.
131 fixclosequote=>1 will implicitly insert the correct quote if
132 it's missing. fixclosequote=>0 is the default.
133 messages
135 parse_line is capable of printing very informative error
136 messages. However, sometimes you don't care enough to print
137 a message (like when calculating completions). Messages are
138 printed by default, so pass messages=>0 to turn them off.
139 This function returns a reference to an array containing three
141 items:
142 tokens
144 A the tokens that the line was separated into (ref to an
145 array of strings).
146 tokno
148 The number of the token (index into the previous array) that
149 contains cursorpos.
150 tokoff
152 The character offet into tokno of cursorpos.
153 If the cursor is at the end of the token, tokoff will point to 1
155 character past the last character in tokno, a non-existant
156 character. If the cursor is between tokens (surrounded by
157 whitespace), a zero-length token will be created for it.
158 parse_escape(lines)
160 Escapes characters that would be otherwise interpreted by the
161 parser. Will accept either a single string or an arrayref of
162 strings (which will be modified in-place).
163 join_line(tokens)
165 This routine does a somewhat intelligent job of joining tokens
166 back into a command line. If token_chars (see "new") is empty
167 (the default), then it just escapes backslashes and quotes, and
168 joins the tokens with spaces.
169 However, if token_chars is nonempty, it tries to insert a
171 visually pleasing amount of space between the tokens. For
172 instance, rather than 'a ( b , c )', it tries to produce 'a (b,
173 c)'. It won't reformat any tokens that aren't found in
174 $self->{token_chars}, of course.
175 To change the formatting, you can redefine the variables
177 $self->{space_none}, $self->{space_before}, and
178 $self->{space_after}. Each variable is a string containing all
179 characters that should not be surrounded by whitespace, should
180 have whitespace before, and should have whitespace after,
181 respectively. Any character found in token_chars, but non in
182 any of these space_ variables, will have space placed both
183 before and after.
184
186 None known.
187
189 Copyright (c) 2003-2011 Scott Bronson, all rights reserved. This
190 program is covered by the MIT license.
191
193 Scott Bronson <bronson@rinspin.com>
194perl v5.38.0 2023-07-21 Text::Shellwords::Cursor(3)