docs/latex/wx/regex.tex

   1 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
   2 %% Name:        regex.tex
   3 %% Purpose:     wxRegEx documentation
   4 %% Author:      Vadim Zeitlin
   5 %% Modified by:
   6 %% Created:     14.07.01
   7 %% RCS-ID:      $Id$
   8 %% Copyright:   (c) 2001 Vadim Zeitlin
   9 %% License:     wxWindows license
  10 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
  11
  12 \section{\class{wxRegEx}}\label{wxregex}
  13
  14 wxRegEx represents a regular expression.  This class provides support
  15 for regular expressions matching and also replacement.
  16
  17 It is built on top of either the system library (if it has support
  18 for POSIX regular expressions - which is the case of the most modern
  19 Unices) or uses the built in Henry Spencer's library.  Henry Spencer
  20 would appreciate being given credit in the documentation of software
  21 which uses his library, but that is not a requirement.
  22
  23 Regular expressions, as defined by POSIX, come in two flavours: {\it extended}
  24 and {\it basic}.  The builtin library also adds a third flavour
  25 of expression \helpref{advanced}{wxresyn}, which is not available
  26 when using the system library.
  27
  28 Unicode is fully supported only when using the builtin library.
  29 When using the system library in Unicode mode, the expressions and data
  30 are translated to the default 8-bit encoding before being passed to
  31 the library.
  32
  33 On platforms where a system library is available, the default is to use
  34 the builtin library for Unicode builds, and the system library otherwise.
  35 It is possible to use the other if preferred by selecting it when building
  36 the wxWidgets.
  37
  38 \wxheading{Derived from}
  39
  40 No base class
  41
  42 \wxheading{Data structures}
  43
  44 Flags for regex compilation to be used with \helpref{Compile()}{wxregexcompile}:
  45
  46 \begin{verbatim}
  47 enum
  48 {
  49     // use extended regex syntax
  50     wxRE_EXTENDED = 0,
  51
  52     // use advanced RE syntax (built-in regex only)
  53 #ifdef wxHAS_REGEX_ADVANCED
  54     wxRE_ADVANCED = 1,
  55 #endif
  56
  57     // use basic RE syntax
  58     wxRE_BASIC    = 2,
  59
  60     // ignore case in match
  61     wxRE_ICASE    = 4,
  62
  63     // only check match, don't set back references
  64     wxRE_NOSUB    = 8,
  65
  66     // if not set, treat '\n' as an ordinary character, otherwise it is
  67     // special: it is not matched by '.' and '^' and '$' always match
  68     // after/before it regardless of the setting of wxRE_NOT[BE]OL
  69     wxRE_NEWLINE  = 16,
  70
  71     // default flags
  72     wxRE_DEFAULT  = wxRE_EXTENDED
  73 }
  74 \end{verbatim}
  75
  76 Flags for regex matching to be used with \helpref{Matches()}{wxregexmatches}.
  77
  78 These flags are mainly useful when doing several matches in a long string
  79 to prevent erroneous matches for {\tt '\textasciicircum'} and {\tt '\$'}:
  80
  81 \begin{verbatim}
  82 enum
  83 {
  84     // '^' doesn't match at the start of line
  85     wxRE_NOTBOL = 32,
  86
  87     // '$' doesn't match at the end of line
  88     wxRE_NOTEOL = 64
  89 }
  90 \end{verbatim}
  91
  92 \wxheading{Examples}
  93
  94 A bad example of processing some text containing email addresses (the example
  95 is bad because the real email addresses can have more complicated form than
  96 {\tt user@host.net}):
  97
  98 \begin{verbatim}
  99 wxString text;
 100 ...
 101 wxRegEx reEmail = wxT("([^@]+)@([[:alnum:].-_].)+([[:alnum:]]+)");
 102 if ( reEmail.Matches(text) )
 103 {
 104     wxString text = reEmail.GetMatch(email);
 105     wxString username = reEmail.GetMatch(email, 1);
 106     if ( reEmail.GetMatch(email, 3) == wxT("com") ) // .com TLD?
 107     {
 108         ...
 109     }
 110 }
 111
 112 // or we could do this to hide the email address
 113 size_t count = reEmail.ReplaceAll(text, wxT("HIDDEN@\\2\\3"));
 114 printf("text now contains %u hidden addresses", count);
 115 \end{verbatim}
 116
 117 \wxheading{Include files}
 118
 119 <wx/regex.h>
 120
 121 \latexignore{\rtfignore{\wxheading{Members}}}
 122
 123 \membersection{wxRegEx::wxRegEx}\label{wxregexwxregex}
 124
 125 \func{}{wxRegEx}{\void}
 126
 127 Default ctor: use \helpref{Compile()}{wxregexcompile} later.
 128
 129 \func{}{wxRegEx}{\param{const wxString\& }{expr}, \param{int }{flags = wxRE\_DEFAULT}}
 130
 131 Create and compile the regular expression, use
 132 \helpref{IsValid}{wxregexisvalid} to test for compilation errors.
 133
 134 \membersection{wxRegEx::\destruct{wxRegEx}}\label{wxregexdtor}
 135
 136 \func{}{\destruct{wxRegEx}}{\void}
 137
 138 dtor not virtual, don't derive from this class
 139
 140 \membersection{wxRegEx::Compile}\label{wxregexcompile}
 141
 142 \func{bool}{Compile}{\param{const wxString\& }{pattern}, \param{int }{flags = wxRE\_DEFAULT}}
 143
 144 Compile the string into regular expression, return {\tt true} if ok or {\tt false}
 145 if string has a syntax error.
 146
 147 \membersection{wxRegEx::IsValid}\label{wxregexisvalid}
 148
 149 \constfunc{bool}{IsValid}{\void}
 150
 151 Return {\tt true} if this is a valid compiled regular expression, {\tt false}
 152 otherwise.
 153
 154 \membersection{wxRegEx::GetMatch}\label{wxregexgetmatch}
 155
 156 \constfunc{bool}{GetMatch}{\param{size\_t* }{start}, \param{size\_t* }{len}, \param{size\_t }{index = 0}}
 157
 158 Get the start index and the length of the match of the expression
 159 (if {\it index} is $0$) or a bracketed subexpression ({\it index} different
 160 from $0$).
 161
 162 May only be called after successful call to \helpref{Matches()}{wxregexmatches}
 163 and only if {\tt wxRE\_NOSUB} was {\bf not} used in
 164 \helpref{Compile()}{wxregexcompile}.
 165
 166 Returns {\tt false} if no match or if an error occurred.
 167
 168 \constfunc{wxString}{GetMatch}{\param{const wxString\& }{text}, \param{size\_t }{index = 0}}
 169
 170 Returns the part of string corresponding to the match where {\it index} is
 171 interpreted as above. Empty string is returned if match failed
 172
 173 May only be called after successful call to \helpref{Matches()}{wxregexmatches}
 174 and only if {\tt wxRE\_NOSUB} was {\bf not} used in
 175 \helpref{Compile()}{wxregexcompile}.
 176
 177 \membersection{wxRegEx::GetMatchCount}\label{wxregexgetmatchcount}
 178
 179 \constfunc{size\_t}{GetMatchCount}{\void}
 180
 181 Returns the size of the array of matches, i.e. the number of bracketed
 182 subexpressions plus one for the expression itself, or $0$ on error.
 183
 184 May only be called after successful call to \helpref{Compile()}{wxregexcompile}.
 185 and only if {\tt wxRE\_NOSUB} was {\bf not} used.
 186
 187 \membersection{wxRegEx::Matches}\label{wxregexmatches}
 188
 189 \constfunc{bool}{Matches}{\param{const wxChar* }{text}, \param{int }{flags = 0}}
 190
 191 \constfunc{bool}{Matches}{\param{const wxChar* }{text}, \param{int }{flags}, \param{size\_t }{len}}
 192
 193 \constfunc{bool}{Matches}{\param{const wxString\& }{text}, \param{int }{flags = 0}}
 194
 195 Matches the precompiled regular expression against the string {\it text},
 196 returns {\tt true} if matches and {\tt false} otherwise.
 197
 198 {\it Flags} may be combination of {\tt wxRE\_NOTBOL} and {\tt wxRE\_NOTEOL}.
 199
 200 Some regex libraries assume that the text given is null terminated, while
 201 others require the length be given as a separate parameter. Therefore for
 202 maximum portability assume that {\it text} cannot contain embedded nulls.
 203
 204 When the {\it Matches(const wxChar *text, int flags = 0)} form is used,
 205 a {\it wxStrlen()} will be done internally if the regex library requires the
 206 length. When using {\it Matches()} in a loop
 207 the {\it Matches(text, flags, len)} form can be used instead, making it
 208 possible to avoid a {\it wxStrlen()} inside the loop.
 209
 210 May only be called after successful call to \helpref{Compile()}{wxregexcompile}.
 211
 212 \membersection{wxRegEx::Replace}\label{wxregexreplace}
 213
 214 \constfunc{int}{Replace}{\param{wxString* }{text}, \param{const wxString\& }{replacement}, \param{size\_t }{maxMatches = 0}}
 215
 216 Replaces the current regular expression in the string pointed to by
 217 {\it text}, with the text in {\it replacement} and return number of matches
 218 replaced (maybe $0$ if none found) or $-1$ on error.
 219
 220 The replacement text may contain back references {\tt $\backslash$number} which will be
 221 replaced with the value of the corresponding subexpression in the
 222 pattern match. {\tt $\backslash$0} corresponds to the entire match and {\tt \&} is a
 223 synonym for it. Backslash may be used to quote itself or {\tt \&} character.
 224
 225 {\it maxMatches} may be used to limit the number of replacements made, setting
 226 it to $1$, for example, will only replace first occurrence (if any) of the
 227 pattern in the text while default value of $0$ means replace all.
 228
 229 \membersection{wxRegEx::ReplaceAll}\label{wxregexreplaceall}
 230
 231 \constfunc{int}{ReplaceAll}{\param{wxString* }{text}, \param{const wxString\& }{replacement}}
 232
 233 Replace all occurrences: this is actually a synonym for
 234 \helpref{Replace()}{wxregexreplace}.
 235
 236 \wxheading{See also}
 237
 238 \helpref{ReplaceFirst}{wxregexreplacefirst}
 239
 240 \membersection{wxRegEx::ReplaceFirst}\label{wxregexreplacefirst}
 241
 242 \constfunc{int}{ReplaceFirst}{\param{wxString* }{text}, \param{const wxString\& }{replacement}}
 243
 244 Replace the first occurrence.
 245
 246 \wxheading{See also}
 247
 248 \helpref{Replace}{wxregexreplace}
 249