Doc/lib/liblocale.tex

   1 \section{\module{locale} ---
   2          Internationalization services}
   3
   4 \declaremodule{standard}{locale}
   5 \modulesynopsis{Internationalization services.}
   6 \moduleauthor{Martin von Loewis}{loewis@informatik.hu-berlin.de}
   7 \sectionauthor{Martin von Loewis}{loewis@informatik.hu-berlin.de}
   8
   9
  10 The \module{locale} module opens access to the \POSIX{} locale database
  11 and functionality. The \POSIX{} locale mechanism allows programmers
  12 to deal with certain cultural issues in an application, without
  13 requiring the programmer to know all the specifics of each country
  14 where the software is executed.
  15
  16 The \module{locale} module is implemented on top of the
  17 \module{_locale}\refbimodindex{_locale} module, which in turn uses an
  18 ANSI C locale implementation if available.
  19
  20 The \module{locale} module defines the following exception and
  21 functions:
  22
  23
  24 \begin{funcdesc}{setlocale}{category\optional{, value}}
  25 If \var{value} is specified, modifies the locale setting for the
  26 \var{category}. The available categories are listed in the data
  27 description below. The value is the name of a locale. An empty string
  28 specifies the user's default settings. If the modification of the
  29 locale fails, the exception \exception{Error} is
  30 raised. If successful, the new locale setting is returned.
  31
  32 If no \var{value} is specified, the current setting for the
  33 \var{category} is returned.
  34
  35 \function{setlocale()} is not thread safe on most systems. Applications
  36 typically start with a call of
  37 \begin{verbatim}
  38 import locale
  39 locale.setlocale(locale.LC_ALL,"")
  40 \end{verbatim}
  41 This sets the locale for all categories to the user's default setting
  42 (typically specified in the \envvar{LANG} environment variable). If
  43 the locale is not changed thereafter, using multithreading should not
  44 cause problems.
  45 \end{funcdesc}
  46
  47 \begin{excdesc}{Error}
  48 Exception raised when \function{setlocale()} fails.
  49 \end{excdesc}
  50
  51 \begin{funcdesc}{localeconv}{}
  52 Returns the database of of the local conventions as a dictionary. This
  53 dictionary has the following strings as keys:
  54 \begin{itemize}
  55 \item \code{decimal_point} specifies the decimal point used in
  56 floating point number representations for the \constant{LC_NUMERIC}
  57 category.
  58 \item \code{grouping} is a sequence of numbers specifying at which
  59 relative positions the \code{thousands_sep} is expected. If the
  60 sequence is terminated with \constant{CHAR_MAX}, no further
  61 grouping is performed. If the sequence terminates with a \code{0}, the last
  62 group size is repeatedly used.
  63 \item \code{thousands_sep} is the character used between groups.
  64 \item \code{int_curr_symbol} specifies the international currency
  65 symbol from the \constant{LC_MONETARY} category.
  66 \item \code{currency_symbol} is the local currency symbol.
  67 \item \code{mon_decimal_point} is the decimal point used in monetary
  68 values.
  69 \item \code{mon_thousands_sep} is the separator for grouping of
  70 monetary values.
  71 \item \code{mon_grouping} has the same format as the \code{grouping}
  72 key; it is used for monetary values.
  73 \item \code{positive_sign} and \code{negative_sign} gives the sign
  74 used for positive and negative monetary quantities.
  75 \item \code{int_frac_digits} and \code{frac_digits} specify the number
  76 of fractional digits used in the international and local formatting
  77 of monetary values.
  78 \item \code{p_cs_precedes} and \code{n_cs_precedes} specifies whether
  79 the currency symbol precedes the value for positive or negative
  80 values.
  81 \item \code{p_sep_by_space} and \code{n_sep_by_space} specifies
  82 whether there is a space between the positive or negative value and
  83 the currency symbol.
  84 \item \code{p_sign_posn} and \code{n_sign_posn} indicate how the
  85 sign should be placed for positive and negative monetary values.
  86 \end{itemize}
  87
  88 The possible values for \code{p_sign_posn} and
  89 \code{n_sign_posn} are given below.
  90
  91 \begin{tableii}{c|l}{code}{Value}{Explanation}
  92 \lineii{0}{Currency and value are surrounded by parentheses.}
  93 \lineii{1}{The sign should precede the value and currency symbol.}
  94 \lineii{2}{The sign should follow the value and currency symbol.}
  95 \lineii{3}{The sign should immediately precede the value.}
  96 \lineii{4}{The sign should immediately follow the value.}
  97 \lineii{LC_MAX}{Nothing is specified in this locale.}
  98 \end{tableii}
  99 \end{funcdesc}
 100
 101 \begin{funcdesc}{strcoll}{string1,string2}
 102 Compares two strings according to the current \constant{LC_COLLATE}
 103 setting. As any other compare function, returns a negative, or a
 104 positive value, or \code{0}, depending on whether \var{string1}
 105 collates before or after \var{string2} or is equal to it.
 106 \end{funcdesc}
 107
 108 \begin{funcdesc}{strxfrm}{string}
 109 Transforms a string to one that can be used for the built-in function
 110 \function{cmp()}\bifuncindex{cmp}, and still returns locale-aware
 111 results.  This function can be used when the same string is compared
 112 repeatedly, e.g. when collating a sequence of strings.
 113 \end{funcdesc}
 114
 115 \begin{funcdesc}{format}{format, val, \optional{grouping\code{ = 0}}}
 116 Formats a number \var{val} according to the current
 117 \constant{LC_NUMERIC} setting.  The format follows the conventions of
 118 the \code{\%} operator.  For floating point values, the decimal point
 119 is modified if appropriate.  If \var{grouping} is true, also takes the
 120 grouping into account.
 121 \end{funcdesc}
 122
 123 \begin{funcdesc}{str}{float}
 124 Formats a floating point number using the same format as the built-in
 125 function \code{str(\var{float})}, but takes the decimal point into
 126 account.
 127 \end{funcdesc}
 128
 129 \begin{funcdesc}{atof}{string}
 130 Converts a string to a floating point number, following the
 131 \constant{LC_NUMERIC} settings.
 132 \end{funcdesc}
 133
 134 \begin{funcdesc}{atoi}{string}
 135 Converts a string to an integer, following the \constant{LC_NUMERIC}
 136 conventions.
 137 \end{funcdesc}
 138
 139 \begin{datadesc}{LC_CTYPE}
 140 \refstmodindex{string}
 141 Locale category for the character type functions. Depending on the
 142 settings of this category, the functions of module \refmodule{string}
 143 dealing with case change their behaviour.
 144 \end{datadesc}
 145
 146 \begin{datadesc}{LC_COLLATE}
 147 Locale category for sorting strings. The functions
 148 \function{strcoll()} and \function{strxfrm()} of the \module{locale}
 149 module are affected.
 150 \end{datadesc}
 151
 152 \begin{datadesc}{LC_TIME}
 153 Locale category for the formatting of time. The function
 154 \function{time.strftime()} follows these conventions.
 155 \end{datadesc}
 156
 157 \begin{datadesc}{LC_MONETARY}
 158 Locale category for formatting of monetary values. The available
 159 options are available from the \function{localeconv()} function.
 160 \end{datadesc}
 161
 162 \begin{datadesc}{LC_MESSAGES}
 163 Locale category for message display. Python currently does not support
 164 application specific locale-aware messages. Messages displayed by the
 165 operating system, like those returned by \function{os.strerror()}
 166 might be affected by this category.
 167 \end{datadesc}
 168
 169 \begin{datadesc}{LC_NUMERIC}
 170 Locale category for formatting numbers. The functions
 171 \function{format()}, \function{atoi()}, \function{atof()} and
 172 \function{str()} of the \module{locale} module are affected by that
 173 category. All other numeric formatting operations are not affected.
 174 \end{datadesc}
 175
 176 \begin{datadesc}{LC_ALL}
 177 Combination of all locale settings. If this flag is used when the
 178 locale is changed, setting the locale for all categories is
 179 attempted. If that fails for any category, no category is changed at
 180 all. When the locale is retrieved using this flag, a string indicating
 181 the setting for all categories is returned. This string can be later
 182 used to restore the settings.
 183 \end{datadesc}
 184
 185 \begin{datadesc}{CHAR_MAX}
 186 This is a symbolic constant used for different values returned by
 187 \function{localeconv()}.
 188 \end{datadesc}
 189
 190 Example:
 191
 192 \begin{verbatim}
 193 >>> import locale
 194 >>> loc = locale.setlocale(locale.LC_ALL) # get current locale
 195 >>> locale.setlocale(locale.LC_ALL, "de") # use German locale
 196 >>> locale.strcoll("f\344n", "foo") # compare a string containing an umlaut
 197 >>> locale.setlocale(locale.LC_ALL, "") # use user's preferred locale
 198 >>> locale.setlocale(locale.LC_ALL, "C") # use default (C) locale
 199 >>> locale.setlocale(locale.LC_ALL, loc) # restore saved locale
 200 \end{verbatim}
 201
 202 \subsection{Background, details, hints, tips and caveats}
 203
 204 The C standard defines the locale as a program-wide property that may
 205 be relatively expensive to change.  On top of that, some
 206 implementation are broken in such a way that frequent locale changes
 207 may cause core dumps.  This makes the locale somewhat painful to use
 208 correctly.
 209
 210 Initially, when a program is started, the locale is the \samp{C} locale, no
 211 matter what the user's preferred locale is.  The program must
 212 explicitly say that it wants the user's preferred locale settings by
 213 calling \code{setlocale(LC_ALL, "")}.
 214
 215 It is generally a bad idea to call \function{setlocale()} in some library
 216 routine, since as a side effect it affects the entire program.  Saving
 217 and restoring it is almost as bad: it is expensive and affects other
 218 threads that happen to run before the settings have been restored.
 219
 220 If, when coding a module for general use, you need a locale
 221 independent version of an operation that is affected by the locale
 222 (e.g. \function{string.lower()}, or certain formats used with
 223 \function{time.strftime()})), you will have to find a way to do it
 224 without using the standard library routine.  Even better is convincing
 225 yourself that using locale settings is okay.  Only as a last resort
 226 should you document that your module is not compatible with
 227 non-\samp{C} locale settings.
 228
 229 The case conversion functions in the
 230 \refmodule{string}\refstmodindex{string} and
 231 \module{strop}\refbimodindex{strop} modules are affected by the locale
 232 settings.  When a call to the \function{setlocale()} function changes
 233 the \constant{LC_CTYPE} settings, the variables
 234 \code{string.lowercase}, \code{string.uppercase} and
 235 \code{string.letters} (and their counterparts in \module{strop}) are
 236 recalculated.  Note that this code that uses these variable through
 237 `\keyword{from} ... \keyword{import} ...', e.g. \code{from string
 238 import letters}, is not affected by subsequent \function{setlocale()}
 239 calls.
 240
 241 The only way to perform numeric operations according to the locale
 242 is to use the special functions defined by this module:
 243 \function{atof()}, \function{atoi()}, \function{format()},
 244 \function{str()}.
 245
 246 \subsection{For extension writers and programs that embed Python}
 247 \label{embedding-locale}
 248
 249 Extension modules should never call \function{setlocale()}, except to
 250 find out what the current locale is.  But since the return value can
 251 only be used portably to restore it, that is not very useful (except
 252 perhaps to find out whether or not the locale is \samp{C}).
 253
 254 When Python is embedded in an application, if the application sets the
 255 locale to something specific before initializing Python, that is
 256 generally okay, and Python will use whatever locale is set,
 257 \emph{except} that the \constant{LC_NUMERIC} locale should always be
 258 \samp{C}.
 259
 260 The \function{setlocale()} function in the \module{locale} module
 261 gives the Python progammer the impression that you can manipulate the
 262 \constant{LC_NUMERIC} locale setting, but this not the case at the C
 263 level: C code will always find that the \constant{LC_NUMERIC} locale
 264 setting is \samp{C}.  This is because too much would break when the
 265 decimal point character is set to something else than a period
 266 (e.g. the Python parser would break).  Caveat: threads that run
 267 without holding Python's global interpreter lock may occasionally find
 268 that the numeric locale setting differs; this is because the only
 269 portable way to implement this feature is to set the numeric locale
 270 settings to what the user requests, extract the relevant
 271 characteristics, and then restore the \samp{C} numeric locale.
 272
 273 When Python code uses the \module{locale} module to change the locale,
 274 this also affects the embedding application.  If the embedding
 275 application doesn't want this to happen, it should remove the
 276 \module{_locale} extension module (which does all the work) from the
 277 table of built-in modules in the \file{config.c} file, and make sure
 278 that the \module{_locale} module is not accessible as a shared library.