ICU 65.1
65.1
|
C-compatible API for localized number formatting; not recommended for C++. More...
#include "unicode/utypes.h"
#include "unicode/parseerr.h"
#include "unicode/ufieldpositer.h"
#include "unicode/umisc.h"
#include "unicode/uformattedvalue.h"
Go to the source code of this file.
Namespaces | |
icu | |
File coll.h. | |
Typedefs | |
typedef enum UNumberUnitWidth | UNumberUnitWidth |
An enum declaring how to render units, including currencies. More... | |
typedef enum UNumberGroupingStrategy | UNumberGroupingStrategy |
An enum declaring the strategy for when and how to display grouping separators (i.e., the separator, often a comma or period, after every 2-3 powers of ten). More... | |
typedef enum UNumberSignDisplay | UNumberSignDisplay |
An enum declaring how to denote positive and negative numbers. More... | |
typedef enum UNumberDecimalSeparatorDisplay | UNumberDecimalSeparatorDisplay |
An enum declaring how to render the decimal separator. More... | |
typedef struct UNumberFormatter | UNumberFormatter |
C-compatible version of icu::number::LocalizedNumberFormatter. More... | |
typedef struct UFormattedNumber | UFormattedNumber |
C-compatible version of icu::number::FormattedNumber. More... | |
Functions | |
UNumberFormatter * | unumf_openForSkeletonAndLocale (const UChar *skeleton, int32_t skeletonLen, const char *locale, UErrorCode *ec) |
Creates a new UNumberFormatter for the given skeleton string and locale. More... | |
UNumberFormatter * | unumf_openForSkeletonAndLocaleWithError (const UChar *skeleton, int32_t skeletonLen, const char *locale, UParseError *perror, UErrorCode *ec) |
Like unumf_openForSkeletonAndLocale, but accepts a UParseError, which will be populated with the location of a skeleton syntax error if such a syntax error exists. More... | |
UFormattedNumber * | unumf_openResult (UErrorCode *ec) |
Creates an object to hold the result of a UNumberFormatter operation. More... | |
void | unumf_formatInt (const UNumberFormatter *uformatter, int64_t value, UFormattedNumber *uresult, UErrorCode *ec) |
Uses a UNumberFormatter to format an integer to a UFormattedNumber. More... | |
void | unumf_formatDouble (const UNumberFormatter *uformatter, double value, UFormattedNumber *uresult, UErrorCode *ec) |
Uses a UNumberFormatter to format a double to a UFormattedNumber. More... | |
void | unumf_formatDecimal (const UNumberFormatter *uformatter, const char *value, int32_t valueLen, UFormattedNumber *uresult, UErrorCode *ec) |
Uses a UNumberFormatter to format a decimal number to a UFormattedNumber. More... | |
const UFormattedValue * | unumf_resultAsValue (const UFormattedNumber *uresult, UErrorCode *ec) |
Returns a representation of a UFormattedNumber as a UFormattedValue, which can be subsequently passed to any API requiring that type. More... | |
int32_t | unumf_resultToString (const UFormattedNumber *uresult, UChar *buffer, int32_t bufferCapacity, UErrorCode *ec) |
Extracts the result number string out of a UFormattedNumber to a UChar buffer if possible. More... | |
UBool | unumf_resultNextFieldPosition (const UFormattedNumber *uresult, UFieldPosition *ufpos, UErrorCode *ec) |
Determines the start and end indices of the next occurrence of the given field in the output string. More... | |
void | unumf_resultGetAllFieldPositions (const UFormattedNumber *uresult, UFieldPositionIterator *ufpositer, UErrorCode *ec) |
Populates the given iterator with all fields in the formatted output string. More... | |
void | unumf_close (UNumberFormatter *uformatter) |
Releases the UNumberFormatter created by unumf_openForSkeletonAndLocale(). More... | |
void | unumf_closeResult (UFormattedNumber *uresult) |
Releases the UFormattedNumber created by unumf_openResult(). More... | |
C-compatible API for localized number formatting; not recommended for C++.
This is the C-compatible version of the NumberFormatter API introduced in ICU 60. C++ users should include unicode/numberformatter.h and use the proper C++ APIs.
The C API accepts a number skeleton string for specifying the settings for formatting, which covers a very large subset of all possible number formatting features. For more information on number skeleton strings, see unicode/numberformatter.h.
When using UNumberFormatter, which is treated as immutable, the results are exported to a mutable UFormattedNumber object, which you subsequently use for populating your string buffer or iterating over the fields.
Example code:
// Setup: UErrorCode ec = U_ZERO_ERROR; UNumberFormatter* uformatter = unumf_openForSkeletonAndLocale(u"precision-integer", -1, "en", &ec); UFormattedNumber* uresult = unumf_openResult(&ec); if (U_FAILURE(ec)) { return; }
// Format a double: unumf_formatDouble(uformatter, 5142.3, uresult, &ec); if (U_FAILURE(ec)) { return; }
// Export the string to a malloc'd buffer: int32_t len = unumf_resultToString(uresult, NULL, 0, &ec); // at this point, ec == U_BUFFER_OVERFLOW_ERROR ec = U_ZERO_ERROR; UChar* buffer = (UChar*) malloc((len+1)*sizeof(UChar)); unumf_resultToString(uresult, buffer, len+1, &ec); if (U_FAILURE(ec)) { return; } // buffer should equal "5,142"
// Cleanup: unumf_close(uformatter); unumf_closeResult(uresult); free(buffer);
If you are a C++ user linking against the C libraries, you can use the LocalPointer versions of these APIs. The following example uses LocalPointer with the decimal number and field position APIs:
// Setup: LocalUNumberFormatterPointer uformatter(unumf_openForSkeletonAndLocale(u"percent", -1, "en", &ec)); LocalUFormattedNumberPointer uresult(unumf_openResult(&ec)); if (U_FAILURE(ec)) { return; }
// Format a decimal number: unumf_formatDecimal(uformatter.getAlias(), "9.87E-3", -1, uresult.getAlias(), &ec); if (U_FAILURE(ec)) { return; }
// Get the location of the percent sign: UFieldPosition ufpos = {UNUM_PERCENT_FIELD, 0, 0}; unumf_resultNextFieldPosition(uresult.getAlias(), &ufpos, &ec); // ufpos should contain beginIndex=7 and endIndex=8 since the string is "0.00987%"
// No need to do any cleanup since we are using LocalPointer.
Definition in file unumberformatter.h.
typedef struct UFormattedNumber UFormattedNumber |
C-compatible version of icu::number::FormattedNumber.
NOTE: This is a C-compatible API; C++ users should build against numberformatter.h instead.
Definition at line 416 of file unumberformatter.h.
An enum declaring how to render the decimal separator.
typedef struct UNumberFormatter UNumberFormatter |
C-compatible version of icu::number::LocalizedNumberFormatter.
NOTE: This is a C-compatible API; C++ users should build against numberformatter.h instead.
Definition at line 406 of file unumberformatter.h.
typedef enum UNumberGroupingStrategy UNumberGroupingStrategy |
An enum declaring the strategy for when and how to display grouping separators (i.e., the separator, often a comma or period, after every 2-3 powers of ten).
The choices are several pre-built strategies for different use cases that employ locale data whenever possible. Example outputs for 1234 and 1234567 in en-IN:
The default is AUTO, which displays grouping separators unless the locale data says that grouping is not customary. To force grouping for all numbers greater than 1000 consistently across locales, use ON_ALIGNED. On the other hand, to display grouping less frequently than the default, use MIN2 or OFF. See the docs of each option for details.
Note: This enum specifies the strategy for grouping sizes. To set which character to use as the grouping separator, use the "symbols" setter.
typedef enum UNumberSignDisplay UNumberSignDisplay |
An enum declaring how to denote positive and negative numbers.
Example outputs when formatting 123, 0, and -123 in en-US:
The exact format, including the position and the code point of the sign, differ by locale.
typedef enum UNumberUnitWidth UNumberUnitWidth |
An enum declaring how to render units, including currencies.
Example outputs when formatting 123 USD and 123 meters in en-CA:
This enum is similar to UMeasureFormatWidth.
An enum declaring how to render the decimal separator.
Enumerator | |
---|---|
UNUM_DECIMAL_SEPARATOR_AUTO | Show the decimal separator when there are one or more digits to display after the separator, and do not show it otherwise. This is the default behavior.
|
UNUM_DECIMAL_SEPARATOR_ALWAYS | Always show the decimal separator, even if there are no digits to display after the separator.
|
UNUM_DECIMAL_SEPARATOR_COUNT | One more than the highest UNumberDecimalSeparatorDisplay value.
|
Definition at line 374 of file unumberformatter.h.
An enum declaring the strategy for when and how to display grouping separators (i.e., the separator, often a comma or period, after every 2-3 powers of ten).
The choices are several pre-built strategies for different use cases that employ locale data whenever possible. Example outputs for 1234 and 1234567 in en-IN:
The default is AUTO, which displays grouping separators unless the locale data says that grouping is not customary. To force grouping for all numbers greater than 1000 consistently across locales, use ON_ALIGNED. On the other hand, to display grouping less frequently than the default, use MIN2 or OFF. See the docs of each option for details.
Note: This enum specifies the strategy for grouping sizes. To set which character to use as the grouping separator, use the "symbols" setter.
Enumerator | |
---|---|
UNUM_GROUPING_OFF | Do not display grouping separators in any locale.
|
UNUM_GROUPING_MIN2 | Display grouping using locale defaults, except do not show grouping on values smaller than 10000 (such that there is a minimum of two digits before the first separator). Note that locales may restrict grouping separators to be displayed only on 1 million or greater (for example, ee and hu) or disable grouping altogether (for example, bg currency). Locale data is used to determine whether to separate larger numbers into groups of 2 (customary in South Asia) or groups of 3 (customary in Europe and the Americas).
|
UNUM_GROUPING_AUTO | Display grouping using the default strategy for all locales. This is the default behavior. Note that locales may restrict grouping separators to be displayed only on 1 million or greater (for example, ee and hu) or disable grouping altogether (for example, bg currency). Locale data is used to determine whether to separate larger numbers into groups of 2 (customary in South Asia) or groups of 3 (customary in Europe and the Americas).
|
UNUM_GROUPING_ON_ALIGNED | Always display the grouping separator on values of at least 1000. This option ignores the locale data that restricts or disables grouping, described in MIN2 and AUTO. This option may be useful to normalize the alignment of numbers, such as in a spreadsheet. Locale data is used to determine whether to separate larger numbers into groups of 2 (customary in South Asia) or groups of 3 (customary in Europe and the Americas).
|
UNUM_GROUPING_THOUSANDS | Use the Western defaults: groups of 3 and enabled for all numbers 1000 or greater. Do not use locale data for determining the grouping strategy.
|
UNUM_GROUPING_COUNT | One more than the highest UNumberGroupingStrategy value.
|
Definition at line 193 of file unumberformatter.h.
enum UNumberSignDisplay |
An enum declaring how to denote positive and negative numbers.
Example outputs when formatting 123, 0, and -123 in en-US:
The exact format, including the position and the code point of the sign, differ by locale.
Enumerator | |
---|---|
UNUM_SIGN_AUTO | Show the minus sign on negative numbers, and do not show the sign on positive numbers. This is the default behavior.
|
UNUM_SIGN_ALWAYS | Show the minus sign on negative numbers and the plus sign on positive numbers, including zero. To hide the sign on zero, see UNUM_SIGN_EXCEPT_ZERO.
|
UNUM_SIGN_NEVER | Do not show the sign on positive or negative numbers.
|
UNUM_SIGN_ACCOUNTING | Use the locale-dependent accounting format on negative numbers, and do not show the sign on positive numbers. The accounting format is defined in CLDR and varies by locale; in many Western locales, the format is a pair of parentheses around the number. Note: Since CLDR defines the accounting format in the monetary context only, this option falls back to the AUTO sign display strategy when formatting without a currency unit. This limitation may be lifted in the future.
|
UNUM_SIGN_ACCOUNTING_ALWAYS | Use the locale-dependent accounting format on negative numbers, and show the plus sign on positive numbers, including zero. For more information on the accounting format, see the ACCOUNTING sign display strategy. To hide the sign on zero, see UNUM_SIGN_ACCOUNTING_EXCEPT_ZERO.
|
UNUM_SIGN_EXCEPT_ZERO | Show the minus sign on negative numbers and the plus sign on positive numbers. Do not show a sign on zero or NaN, unless the sign bit is set (-0.0 gets a sign).
|
UNUM_SIGN_ACCOUNTING_EXCEPT_ZERO | Use the locale-dependent accounting format on negative numbers, and show the plus sign on positive numbers. Do not show a sign on zero or NaN, unless the sign bit is set (-0.0 gets a sign). For more information on the accounting format, see the ACCOUNTING sign display strategy.
|
UNUM_SIGN_COUNT | One more than the highest UNumberSignDisplay value.
|
Definition at line 287 of file unumberformatter.h.
enum UNumberUnitWidth |
An enum declaring how to render units, including currencies.
Example outputs when formatting 123 USD and 123 meters in en-CA:
This enum is similar to UMeasureFormatWidth.
Enumerator | |
---|---|
UNUM_UNIT_WIDTH_NARROW | Print an abbreviated version of the unit name. Similar to SHORT, but always use the shortest available abbreviation or symbol. This option can be used when the context hints at the identity of the unit. For more information on the difference between NARROW and SHORT, see SHORT. In CLDR, this option corresponds to the "Narrow" format for measure units and the "¤¤¤¤¤" placeholder for currencies.
|
UNUM_UNIT_WIDTH_SHORT | Print an abbreviated version of the unit name. Similar to NARROW, but use a slightly wider abbreviation or symbol when there may be ambiguity. This is the default behavior. For example, in es-US, the SHORT form for Fahrenheit is "{0} °F", but the NARROW form is "{0}°", since Fahrenheit is the customary unit for temperature in that locale. In CLDR, this option corresponds to the "Short" format for measure units and the "¤" placeholder for currencies.
|
UNUM_UNIT_WIDTH_FULL_NAME | Print the full name of the unit, without any abbreviations. In CLDR, this option corresponds to the default format for measure units and the "¤¤¤" placeholder for currencies.
|
UNUM_UNIT_WIDTH_ISO_CODE | Use the three-digit ISO XXX code in place of the symbol for displaying currencies. The behavior of this option is currently undefined for use with measure units. In CLDR, this option corresponds to the "¤¤" placeholder for currencies.
|
UNUM_UNIT_WIDTH_HIDDEN | Format the number according to the specified unit, but do not display the unit. For currencies, apply monetary symbols and formats as with SHORT, but omit the currency symbol. For measure units, the behavior is equivalent to not specifying the unit at all.
|
UNUM_UNIT_WIDTH_COUNT | One more than the highest UNumberUnitWidth value.
|
Definition at line 98 of file unumberformatter.h.
void unumf_close | ( | UNumberFormatter * | uformatter | ) |
Releases the UNumberFormatter created by unumf_openForSkeletonAndLocale().
uformatter | An object created by unumf_openForSkeletonAndLocale(). |
void unumf_closeResult | ( | UFormattedNumber * | uresult | ) |
Releases the UFormattedNumber created by unumf_openResult().
uresult | An object created by unumf_openResult(). |
void unumf_formatDecimal | ( | const UNumberFormatter * | uformatter, |
const char * | value, | ||
int32_t | valueLen, | ||
UFormattedNumber * | uresult, | ||
UErrorCode * | ec | ||
) |
Uses a UNumberFormatter to format a decimal number to a UFormattedNumber.
A string, field position, and other information can be retrieved from the UFormattedNumber.
The UNumberFormatter can be shared between threads. Each thread should have its own local UFormattedNumber, however, for storing the result of the formatting operation.
The syntax of the unformatted number is a "numeric string" as defined in the Decimal Arithmetic Specification, available at http://speleotrove.com/decimal
NOTE: This is a C-compatible API; C++ users should build against numberformatter.h instead.
uformatter | A formatter object created by unumf_openForSkeletonAndLocale or similar. |
value | The numeric string to be formatted. |
valueLen | The length of the numeric string, or -1 if it is NUL-terminated. |
uresult | The object that will be mutated to store the result; see unumf_openResult. |
ec | Set if an error occurs. |
void unumf_formatDouble | ( | const UNumberFormatter * | uformatter, |
double | value, | ||
UFormattedNumber * | uresult, | ||
UErrorCode * | ec | ||
) |
Uses a UNumberFormatter to format a double to a UFormattedNumber.
A string, field position, and other information can be retrieved from the UFormattedNumber.
The UNumberFormatter can be shared between threads. Each thread should have its own local UFormattedNumber, however, for storing the result of the formatting operation.
NOTE: This is a C-compatible API; C++ users should build against numberformatter.h instead.
uformatter | A formatter object created by unumf_openForSkeletonAndLocale or similar. |
value | The number to be formatted. |
uresult | The object that will be mutated to store the result; see unumf_openResult. |
ec | Set if an error occurs. |
void unumf_formatInt | ( | const UNumberFormatter * | uformatter, |
int64_t | value, | ||
UFormattedNumber * | uresult, | ||
UErrorCode * | ec | ||
) |
Uses a UNumberFormatter to format an integer to a UFormattedNumber.
A string, field position, and other information can be retrieved from the UFormattedNumber.
The UNumberFormatter can be shared between threads. Each thread should have its own local UFormattedNumber, however, for storing the result of the formatting operation.
NOTE: This is a C-compatible API; C++ users should build against numberformatter.h instead.
uformatter | A formatter object created by unumf_openForSkeletonAndLocale or similar. |
value | The number to be formatted. |
uresult | The object that will be mutated to store the result; see unumf_openResult. |
ec | Set if an error occurs. |
UNumberFormatter* unumf_openForSkeletonAndLocale | ( | const UChar * | skeleton, |
int32_t | skeletonLen, | ||
const char * | locale, | ||
UErrorCode * | ec | ||
) |
Creates a new UNumberFormatter for the given skeleton string and locale.
This is currently the only method for creating a new UNumberFormatter.
Objects of type UNumberFormatter returned by this method are threadsafe.
For more details on skeleton strings, see the documentation in numberformatter.h. For more details on the usage of this API, see the documentation at the top of unumberformatter.h.
NOTE: This is a C-compatible API; C++ users should build against numberformatter.h instead.
skeleton | The skeleton string, like u"percent precision-integer" |
skeletonLen | The number of UChars in the skeleton string, or -1 it it is NUL-terminated. |
locale | The NUL-terminated locale ID. |
ec | Set if an error occurs. |
UNumberFormatter* unumf_openForSkeletonAndLocaleWithError | ( | const UChar * | skeleton, |
int32_t | skeletonLen, | ||
const char * | locale, | ||
UParseError * | perror, | ||
UErrorCode * | ec | ||
) |
Like unumf_openForSkeletonAndLocale, but accepts a UParseError, which will be populated with the location of a skeleton syntax error if such a syntax error exists.
skeleton | The skeleton string, like u"percent precision-integer" |
skeletonLen | The number of UChars in the skeleton string, or -1 it it is NUL-terminated. |
locale | The NUL-terminated locale ID. |
perror | A parse error struct populated if an error occurs when parsing. Can be NULL. If no error occurs, perror->offset will be set to -1. |
ec | Set if an error occurs. |
UFormattedNumber* unumf_openResult | ( | UErrorCode * | ec | ) |
Creates an object to hold the result of a UNumberFormatter operation.
The object can be used repeatedly; it is cleared whenever passed to a format function.
ec | Set if an error occurs. |
const UFormattedValue* unumf_resultAsValue | ( | const UFormattedNumber * | uresult, |
UErrorCode * | ec | ||
) |
Returns a representation of a UFormattedNumber as a UFormattedValue, which can be subsequently passed to any API requiring that type.
The returned object is owned by the UFormattedNumber and is valid only as long as the UFormattedNumber is present and unchanged in memory.
You can think of this method as a cast between types.
uresult | The object containing the formatted string. |
ec | Set if an error occurs. |
void unumf_resultGetAllFieldPositions | ( | const UFormattedNumber * | uresult, |
UFieldPositionIterator * | ufpositer, | ||
UErrorCode * | ec | ||
) |
Populates the given iterator with all fields in the formatted output string.
This allows you to determine the locations of the integer part, fraction part, and sign.
This is an alternative to the more powerful ufmtval_nextPosition API.
If you need information on only one field, use ufmtval_nextPosition or unumf_resultNextFieldPosition.
uresult | The object containing the formatted number. |
ufpositer | A pointer to a UFieldPositionIterator created by ufieldpositer_open. Iteration information already present in the UFieldPositionIterator is deleted, and the iterator is reset to apply to the fields in the formatted string created by this function call. The field values and indexes returned by ufieldpositer_next represent fields denoted by the UNumberFormatFields enum. Fields are not returned in a guaranteed order. Fields cannot overlap, but they may nest. For example, 1234 could format as "1,234" which might consist of a grouping separator field for ',' and an integer field encompassing the entire string. |
ec | Set if an error occurs. |
UBool unumf_resultNextFieldPosition | ( | const UFormattedNumber * | uresult, |
UFieldPosition * | ufpos, | ||
UErrorCode * | ec | ||
) |
Determines the start and end indices of the next occurrence of the given field in the output string.
This allows you to determine the locations of, for example, the integer part, fraction part, or symbols.
This is a simpler but less powerful alternative to ufmtval_nextPosition.
If a field occurs just once, calling this method will find that occurrence and return it. If a field occurs multiple times, this method may be called repeatedly with the following pattern:
UFieldPosition ufpos = {UNUM_GROUPING_SEPARATOR_FIELD, 0, 0}; while (unumf_resultNextFieldPosition(uresult, ufpos, &ec)) { // do something with ufpos. }
This method is useful if you know which field to query. If you want all available field position information, use unumf_resultGetAllFieldPositions().
NOTE: All fields of the UFieldPosition must be initialized before calling this method.
uresult | The object containing the formatted number. |
ufpos | Input+output variable. On input, the "field" property determines which field to look up, and the "endIndex" property determines where to begin the search. On output, the "beginIndex" field is set to the beginning of the first occurrence of the field after the input "endIndex", and "endIndex" is set to the end of that occurrence of the field (exclusive index). If a field position is not found, the FieldPosition is not changed and the method returns FALSE. |
ec | Set if an error occurs. |
int32_t unumf_resultToString | ( | const UFormattedNumber * | uresult, |
UChar * | buffer, | ||
int32_t | bufferCapacity, | ||
UErrorCode * | ec | ||
) |
Extracts the result number string out of a UFormattedNumber to a UChar buffer if possible.
If bufferCapacity is greater than the required length, a terminating NUL is written. If bufferCapacity is less than the required length, an error code is set.
Also see ufmtval_getString, which returns a NUL-terminated string:
int32_t len; const UChar* str = ufmtval_getString(unumf_resultAsValue(uresult, &ec), &len, &ec);
NOTE: This is a C-compatible API; C++ users should build against numberformatter.h instead.
uresult | The object containing the formatted number. |
buffer | Where to save the string output. |
bufferCapacity | The number of UChars available in the buffer. |
ec | Set if an error occurs. |