wxWidgets/interface/wx/intl.h

660 lines
26 KiB
C
Raw Normal View History

/////////////////////////////////////////////////////////////////////////////
// Name: intl.h
// Purpose: interface of wxLocale
// Author: wxWidgets team
// RCS-ID: $Id$
// Licence: wxWindows license
/////////////////////////////////////////////////////////////////////////////
/**
@class wxLocale
wxLocale class encapsulates all language-dependent settings and is a
generalization of the C locale concept.
In wxWidgets this class manages message catalogs which contain the translations
of the strings used to the current language.
@b wxPerl note: In wxPerl you can't use the '_' function name, so
the @c Wx::Locale module can export the @c gettext and
@c gettext_noop under any given name.
@code
# this imports gettext ( equivalent to Wx::GetTranslation
# and gettext_noop ( a noop )
# into your module
use Wx::Locale qw(:default);
# ....
# use the functions
print gettext( "Panic!" );
button = Wx::Button-new( window, -1, gettext( "Label" ) );
@endcode
If you need to translate a lot of strings, then adding gettext( ) around
each one is a long task ( that is why _( ) was introduced ), so just choose
a shorter name for gettext:
@code
#
use Wx::Locale 'gettext' = 't',
'gettext_noop' = 'gettext_noop';
# ...
# use the functions
print t( "Panic!!" );
# ...
@endcode
@library{wxbase}
@category{FIXME}
@see @ref overview_internationalization, @ref overview_sampleinternat "Internat
sample", wxXLocale
*/
class wxLocale
{
public:
//@{
/**
See Init() for parameters description.
The call of this function has several global side effects which you should
understand: first of all, the application locale is changed - note that this
will affect many of standard C library functions such as printf() or strftime().
Second, this wxLocale object becomes the new current global locale for the
application and so all subsequent calls to wxGetTranslation() will try to
translate the messages using the message catalogs for this locale.
*/
wxLocale();
wxLocale(int language,
int flags =
wxLOCALE_LOAD_DEFAULT | wxLOCALE_CONV_ENCODING);
wxLocale(const wxString& name,
const wxString& short = wxEmptyString,
const wxString& locale = wxEmptyString,
bool bLoadDefault = true,
bool bConvertEncoding = false);
//@}
/**
The destructor, like the constructor, also has global side effects: the
previously
set locale is restored and so the changes described in
Init() documentation are rolled back.
*/
~wxLocale();
//@{
/**
Add a catalog for use with the current locale: it is searched for in standard
places (current directory first, then the system one), but you may also prepend
additional directories to the search path with
AddCatalogLookupPathPrefix().
All loaded catalogs will be used for message lookup by
GetString() for the current locale.
Returns @true if catalog was successfully loaded, @false otherwise (which might
mean that the catalog is not found or that it isn't in the correct format).
The second form of this method takes two additional arguments,
@a msgIdLanguage and @e msgIdCharset.
@a msgIdLanguage specifies the language of "msgid" strings in source code
(i.e. arguments to GetString(),
wxGetTranslation() and the
_() macro). It is used if AddCatalog cannot find any
catalog for current language: if the language is same as source code language,
then strings from source code are used instead.
@a msgIdCharset lets you specify the charset used for msgids in sources
in case they use 8-bit characters (e.g. German or French strings). This
argument has no effect in Unicode build, because literals in sources are
Unicode strings; you have to use compiler-specific method of setting the right
charset when compiling with Unicode.
By default (i.e. when you use the first form), msgid strings are assumed
to be in English and written only using 7-bit ASCII characters.
If you have to deal with non-English strings or 8-bit characters in the source
code, see the instructions in
@ref overview_nonenglishoverview "Writing non-English applications".
*/
bool AddCatalog(const wxString& domain);
bool AddCatalog(const wxString& domain,
wxLanguage msgIdLanguage,
const wxString& msgIdCharset);
//@}
/**
Add a prefix to the catalog lookup path: the message catalog files will be
looked up under prefix/lang/LC_MESSAGES, prefix/lang and prefix
(in this order).
This only applies to subsequent invocations of AddCatalog().
*/
void AddCatalogLookupPathPrefix(const wxString& prefix);
/**
Adds custom, user-defined language to the database of known languages. This
database is used in conjunction with the first form of
Init().
wxLanguageInfo is defined as follows:
@e Language should be greater than wxLANGUAGE_USER_DEFINED.
Wx::LanguageInfo-new( language, canonicalName, WinLang, WinSubLang, Description
)
*/
static void AddLanguage(const wxLanguageInfo& info);
/**
This function may be used to find the language description structure for the
given locale, specified either as a two letter ISO language code (for example,
"pt"), a language code followed by the country code ("pt_BR") or a full, human
readable, language description ("Portuguese-Brazil").
Returns the information for the given language or @NULL if this language
is unknown. Note that even if the returned pointer is valid, the caller should
@e not delete it.
@see GetLanguageInfo()
*/
static wxLanguageInfo* FindLanguageInfo(const wxString& locale);
/**
Returns the canonical form of current locale name. Canonical form is the
one that is used on UNIX systems: it is a two- or five-letter string in xx or
xx_YY format, where xx is ISO 639 code of language and YY is ISO 3166 code of
the country. Examples are "en", "en_GB", "en_US" or "fr_FR".
This form is internally used when looking up message catalogs.
Compare GetSysName().
*/
wxString GetCanonicalName() const;
/**
Returns the header value for header @e header. The search for @a header is case
sensitive. If an @e domain
is passed, this domain is searched. Else all domains will be searched until a
header has been found.
The return value is the value of the header if found. Else this will be empty.
*/
wxString GetHeaderValue(const wxString& header,
const wxString& domain = wxEmptyString) const;
/**
Returns wxLanguage() constant of current language.
Note that you can call this function only if you used the form of
Init() that takes wxLanguage argument.
*/
int GetLanguage() const;
/**
Returns a pointer to wxLanguageInfo structure containing information about the
given language or @NULL if this language is unknown. Note that even if the
returned pointer is valid, the caller should @e not delete it.
See AddLanguage() for the wxLanguageInfo
description.
As with Init(), @c wxLANGUAGE_DEFAULT has the
special meaning if passed as an argument to this function and in this case the
result of GetSystemLanguage() is used.
*/
static wxLanguageInfo* GetLanguageInfo(int lang) const;
/**
Returns English name of the given language or empty string if this
language is unknown.
See GetLanguageInfo() for a remark about
special meaning of @c wxLANGUAGE_DEFAULT.
*/
static wxString GetLanguageName(int lang) const;
/**
Returns the locale name as passed to the constructor or
Init(). This is full, human-readable name,
e.g. "English" or "French".
*/
const wxString GetLocale() const;
/**
Returns the current short name for the locale (as given to the constructor or
the Init() function).
*/
const wxString GetName() const;
//@{
/**
Retrieves the translation for a string in all loaded domains unless the szDomain
parameter is specified (and then only this catalog/domain is searched).
Returns original string if translation is not available
(in this case an error message is generated the first time
a string is not found; use wxLogNull to suppress it).
The second form is used when retrieving translation of string that has
different singular and plural form in English or different plural forms in some
other language. It takes two extra arguments: @e origString
parameter must contain the singular form of the string to be converted.
It is also used as the key for the search in the catalog.
The @a origString2 parameter is the plural form (in English).
The parameter @a n is used to determine the plural form. If no
message catalog is found @a origString is returned if 'n == 1',
otherwise @e origString2.
See GNU gettext manual for additional information on plural forms handling.
This method is called by the wxGetTranslation()
function and _() macro.
@remarks Domains are searched in the last to first order, i.e. catalogs
added later override those added before.
*/
const wxString GetString(const wxString& origString,
const wxString& domain = wxEmptyString) const;
const const wxString& GetString(const wxString& origString,
const wxString& origString2,
size_t n,
const wxString& domain = NULL) const;
//@}
/**
Returns current platform-specific locale name as passed to setlocale().
Compare GetCanonicalName().
*/
wxString GetSysName() const;
/**
Tries to detect the user's default font encoding.
Returns wxFontEncoding() value or
@b wxFONTENCODING_SYSTEM if it couldn't be determined.
*/
static wxFontEncoding GetSystemEncoding() const;
/**
Tries to detect the name of the user's default font encoding. This string isn't
particularly useful for the application as its form is platform-dependent and
so you should probably use
GetSystemEncoding() instead.
Returns a user-readable string value or an empty string if it couldn't be
determined.
*/
static wxString GetSystemEncodingName() const;
/**
Tries to detect the user's default language setting.
Returns wxLanguage() value or
@b wxLANGUAGE_UNKNOWN if the language-guessing algorithm failed.
*/
static int GetSystemLanguage() const;
//@{
/**
The second form is deprecated, use the first one unless you know what you are
doing.
@param language
wxLanguage identifier of the locale.
wxLANGUAGE_DEFAULT has special meaning -- wxLocale will use system's
default
language (see GetSystemLanguage).
@param flags
Combination of the following:
wxLOCALE_LOAD_DEFAULT
Load the message catalog
for the given locale containing the translations of standard wxWidgets
messages
automatically.
wxLOCALE_CONV_ENCODING
Automatically convert message
catalogs to platform's default encoding. Note that it will do only basic
conversion between well-known pair like iso8859-1 and windows-1252 or
iso8859-2 and windows-1250. See Writing non-English applications for
detailed
description of this behaviour. Note that this flag is meaningless in
Unicode build.
@param name
The name of the locale. Only used in diagnostic messages.
@param short
The standard 2 letter locale abbreviation; it is used as the
directory prefix when looking for the message catalog files.
@param locale
The parameter for the call to setlocale(). Note that it is
platform-specific.
@param bLoadDefault
May be set to @false to prevent loading of the message catalog
for the given locale containing the translations of standard wxWidgets
messages.
This parameter would be rarely used in normal circumstances.
@param bConvertEncoding
May be set to @true to do automatic conversion of message
catalogs to platform's native encoding. Note that it will do only basic
conversion between well-known pair like iso8859-1 and windows-1252 or
iso8859-2 and windows-1250.
See Writing non-English applications for detailed
description of this behaviour.
*/
bool Init(int language = wxLANGUAGE_DEFAULT,
int flags =
wxLOCALE_LOAD_DEFAULT | wxLOCALE_CONV_ENCODING);
bool Init(const wxString& name,
const wxString& short = wxEmptyString,
const wxString& locale = wxEmptyString,
bool bLoadDefault = true,
bool bConvertEncoding = false);
//@}
/**
Check whether the operating system and/or C run time environment supports
this locale. For example in Windows 2000 and Windows XP, support for many
locales is not installed by default. Returns @true if the locale is
supported.
The argument @a lang is the wxLanguage identifier. To obtain this for a
given a two letter ISO language code, use
FindLanguageInfo() to obtain its
wxLanguageInfo structure. See AddLanguage() for
the wxLanguageInfo description.
@since 2.7.1.
*/
static bool IsAvailable(int lang);
/**
Check if the given catalog is loaded, and returns @true if it is.
According to GNU gettext tradition, each catalog
normally corresponds to 'domain' which is more or less the application name.
See also: AddCatalog()
*/
bool IsLoaded(const char* domain) const;
/**
Returns @true if the locale could be set successfully.
*/
bool IsOk() const;
/**
See @ref overview_languagecodes "list of recognized language constants".
These constants may be used to specify the language
in Init() and are returned by
GetSystemLanguage():
*/
};
/**
@class wxXLocale
wxXLocale::wxXLocale
wxXLocale::GetCLocale
wxXLocale::IsOk
Introduction
This class represents a locale object used by so-called xlocale API. Unlike
wxLocale it doesn't provide any non-trivial operations but
simply provides a portable wrapper for POSIX @c locale_t type. It exists
solely to be provided as an argument to various @c wxFoo_l() functions
which are the extensions of the standard locale-dependent functions (hence the
name xlocale). These functions do exactly the same thing as the corresponding
standard @c foo() except that instead of using the global program locale
they use the provided wxXLocale object. For example, if the user runs the
program in French locale, the standard @c printf() function will output
floating point numbers using decimal comma instead of decimal period. If the
program needs to format a floating-point number in a standard format it can
use @c wxPrintf_l(wxXLocale::GetCLocale(), "%g", number) to do it.
Conversely, if a program wanted to output the number in French locale, even if
the current locale is different, it could use wxXLocale(wxLANGUAGE_FRENCH).
Availability
This class is fully implemented only under the platforms where xlocale POSIX
API or equivalent is available. Currently the xlocale API is available under
most of the recent Unix systems (including Linux, various BSD and Mac OS X) and
Microsoft Visual C++ standard library provides a similar API starting from
version 8 (Visual Studio 2005).
If neither POSIX API nor Microsoft proprietary equivalent are available, this
class is still available but works in degraded mode: the only supported locale
is the C one and attempts to create wxXLocale object for any other locale will
fail. You can use the preprocessor macro @c wxHAS_XLOCALE_SUPPORT to
test if full xlocale API is available or only skeleton C locale support is
present.
Notice that wxXLocale is new in wxWidgets 2.9.0 and is not compiled in if
@c wxUSE_XLOCALE was set to 0 during the library compilation.
Locale-dependent functions
Currently the following @c _l-functions are available:
Character classification functions: @c wxIsxxx_l(), e.g.
@c wxIsalpha_l(), @c wxIslower_l() and all the others.
Character transformation functions: @c wxTolower_l() and
@c wxToupper_l()
We hope to provide many more functions (covering numbers, time and formatted
IO) in the near future.
@library{wxbase}
@category{FIXME}
@see wxLocale
*/
class wxXLocale
{
public:
//@{
/**
Creates the locale object corresponding to the specified locale string. The
locale string is system-dependent, use constructor taking wxLanguage for better
portability.
*/
wxLocale();
wxLocale(wxLanguage lang);
wxLocale(const char* loc);
//@}
/**
This class is fully implemented only under the platforms where xlocale POSIX
API or equivalent is available. Currently the xlocale API is available under
most of the recent Unix systems (including Linux, various BSD and Mac OS X) and
Microsoft Visual C++ standard library provides a similar API starting from
version 8 (Visual Studio 2005).
If neither POSIX API nor Microsoft proprietary equivalent are available, this
class is still available but works in degraded mode: the only supported locale
is the C one and attempts to create wxXLocale object for any other locale will
fail. You can use the preprocessor macro @c wxHAS_XLOCALE_SUPPORT to
test if full xlocale API is available or only skeleton C locale support is
present.
Notice that wxXLocale is new in wxWidgets 2.9.0 and is not compiled in if
@c wxUSE_XLOCALE was set to 0 during the library compilation.
*/
/**
Returns the global object representing the "C" locale. For an even shorter
access to this object a global @c wxCLocale variable (implemented as a
macro) is provided and can be used instead of calling this method.
*/
static wxXLocale GetCLocale();
/**
This class represents a locale object used by so-called xlocale API. Unlike
wxLocale it doesn't provide any non-trivial operations but
simply provides a portable wrapper for POSIX @c locale_t type. It exists
solely to be provided as an argument to various @c wxFoo_l() functions
which are the extensions of the standard locale-dependent functions (hence the
name xlocale). These functions do exactly the same thing as the corresponding
standard @c foo() except that instead of using the global program locale
they use the provided wxXLocale object. For example, if the user runs the
program in French locale, the standard @c printf() function will output
floating point numbers using decimal comma instead of decimal period. If the
program needs to format a floating-point number in a standard format it can
use @c wxPrintf_l(wxXLocale::GetCLocale(), "%g", number) to do it.
Conversely, if a program wanted to output the number in French locale, even if
the current locale is different, it could use wxXLocale(wxLANGUAGE_FRENCH).
*/
/**
Returns @true if this object is initialized, i.e. represents a valid locale
or
@false otherwise.
*/
bool IsOk() const;
/**
Currently the following @c _l-functions are available:
Character classification functions: @c wxIsxxx_l(), e.g.
@c wxIsalpha_l(), @c wxIslower_l() and all the others.
Character transformation functions: @c wxTolower_l() and
@c wxToupper_l()
We hope to provide many more functions (covering numbers, time and formatted
IO) in the near future.
@see wxLocale
*/
};
// ============================================================================
// Global functions/macros
// ============================================================================
/** @ingroup group_funcmacro_string */
//@{
/**
This macro is identical to _() but for the plural variant of
wxGetTranslation().
@return A const wxString.
@header{wx/intl.h}
*/
#define wxPLURAL(string, plural, n)
/**
This macro doesn't do anything in the program code -- it simply expands to
the value of its argument.
However it does have a purpose which is to mark the literal strings for the
extraction into the message catalog created by @c xgettext program. Usually
this is achieved using _() but that macro not only marks the string for
extraction but also expands into a wxGetTranslation() call which means that
it cannot be used in some situations, notably for static array
initialization.
Here is an example which should make it more clear: suppose that you have a
static array of strings containing the weekday names and which have to be
translated (note that it is a bad example, really, as wxDateTime already
can be used to get the localized week day names already). If you write:
@code
static const char * const weekdays[] = { _("Mon"), ..., _("Sun") };
...
// use weekdays[n] as usual
@endcode
The code wouldn't compile because the function calls are forbidden in the
array initializer. So instead you should do this:
@code
static const char * const weekdays[] = { wxTRANSLATE("Mon"), ...,
wxTRANSLATE("Sun") };
...
// use wxGetTranslation(weekdays[n])
@endcode
Note that although the code @b would compile if you simply omit
wxTRANSLATE() in the above, it wouldn't work as expected because there
would be no translations for the weekday names in the program message
catalog and wxGetTranslation() wouldn't find them.
@return A const wxChar*.
@header{wx/intl.h}
*/
#define wxTRANSLATE(string)
/**
This function returns the translation of @a string in the current
@c locale(). If the string is not found in any of the loaded message
catalogs (see @ref overview_i18n), the original string is returned. In
debug build, an error message is logged -- this should help to find the
strings which were not yet translated. If @a domain is specified then only
that domain/catalog is searched for a matching string. As this function is
used very often, an alternative (and also common in Unix world) syntax is
provided: the _() macro is defined to do the same thing as
wxGetTranslation().
This function calls wxLocale::GetString().
@note This function is not suitable for literal strings in Unicode builds
since the literal strings must be enclosed into _T() or wxT() macro
which makes them unrecognised by @c xgettext, and so they are not
extracted to the message catalog. Instead, use the _() and wxPLURAL()
macro for all literal strings.
@see wxGetTranslation(const wxString&, const wxString&, size_t, const wxString&)
@header{wx/intl.h}
*/
const wxString wxGetTranslation(const wxString& string,
const wxString& domain = wxEmptyString);
/**
This is an overloaded version of
wxGetTranslation(const wxString&, const wxString&), please see its
documentation for general information.
This version is used when retrieving translation of string that has
different singular and plural forms in English or different plural forms in
some other language. Like wxGetTranslation(const wxString&,const wxString&),
the @a string parameter must contain the singular form of the string to be
converted and is used as the key for the search in the catalog. The
@a plural parameter is the plural form (in English). The parameter @a n is
used to determine the plural form. If no message catalog is found,
@a string is returned if "n == 1", otherwise @a plural is returned.
See GNU gettext Manual for additional information on plural forms handling:
<http://www.gnu.org/software/gettext/manual/gettext.html#Plural-forms>
For a shorter alternative see the wxPLURAL() macro.
This function calls wxLocale::GetString().
@header{wx/intl.h}
*/
const wxString wxGetTranslation(const wxString& string,
const wxString& plural, size_t n,
const wxString& domain = wxEmptyString);
/**
This macro expands into a call to wxGetTranslation(), so it marks the
message for the extraction by @c xgettext just as wxTRANSLATE() does, but
also returns the translation of the string for the current locale during
execution.
Don't confuse this with _T()!
@header{wx/intl.h}
*/
const wxString _(const wxString& string);
//@}