This function performs basic operations for "shaping" Arabic text. It is most sl@0: * useful for use with legacy data formats and legacy display technology sl@0: * (simple terminals). All operations are performed on Unicode characters.
sl@0: * sl@0: *Text-based shaping means that some character code points in the text are sl@0: * replaced by others depending on the context. It transforms one kind of text sl@0: * into another. In comparison, modern displays for Arabic text select sl@0: * appropriate, context-dependent font glyphs for each text element, which means sl@0: * that they transform text into a glyph vector.
sl@0: * sl@0: *Text transformations are necessary when modern display technology is not sl@0: * available or when text needs to be transformed to or from legacy formats that sl@0: * use "shaped" characters. Since the Arabic script is cursive, connecting sl@0: * adjacent letters to each other, computers select images for each letter based sl@0: * on the surrounding letters. This usually results in four images per Arabic sl@0: * letter: initial, middle, final, and isolated forms. In Unicode, on the other sl@0: * hand, letters are normally stored abstract, and a display system is expected sl@0: * to select the necessary glyphs. (This makes searching and other text sl@0: * processing easier because the same letter has only one code.) It is possible sl@0: * to mimic this with text transformations because there are characters in sl@0: * Unicode that are rendered as letters with a specific shape sl@0: * (or cursive connectivity). They were included for interoperability with sl@0: * legacy systems and codepages, and for unsophisticated display systems.
sl@0: * sl@0: *A second kind of text transformations is supported for Arabic digits: sl@0: * For compatibility with legacy codepages that only include European digits, sl@0: * it is possible to replace one set of digits by another, changing the sl@0: * character code points. These operations can be performed for either sl@0: * Arabic-Indic Digits (U+0660...U+0669) or Eastern (Extended) Arabic-Indic sl@0: * digits (U+06f0...U+06f9).
sl@0: * sl@0: *Some replacements may result in more or fewer characters (code points). sl@0: * By default, this means that the destination buffer may receive text with a sl@0: * length different from the source length. Some legacy systems rely on the sl@0: * length of the text to be constant. They expect extra spaces to be added sl@0: * or consumed either next to the affected character or at the end of the sl@0: * text.
sl@0: * sl@0: *For details about the available operations, see the description of the
sl@0: * U_SHAPE_... options.
source.
sl@0: *
sl@0: * @param dest The destination buffer that will receive the results of the
sl@0: * requested operations. It may be NULL only if
sl@0: * destSize is 0. The source and destination must not
sl@0: * overlap.
sl@0: *
sl@0: * @param destSize The size (capacity) of the destination buffer in UChars.
sl@0: * If destSize is 0, then no output is produced,
sl@0: * but the necessary buffer size is returned ("preflighting").
sl@0: *
sl@0: * @param options This is a 32-bit set of flags that specify the operations
sl@0: * that are performed on the input text. If no error occurs,
sl@0: * then the result will always be written to the destination
sl@0: * buffer.
sl@0: *
sl@0: * @param pErrorCode must be a valid pointer to an error code value,
sl@0: * which must not indicate a failure before the function call.
sl@0: *
sl@0: * @return The number of UChars written to the destination buffer.
sl@0: * If an error occured, then no output was written, or it may be
sl@0: * incomplete. If U_BUFFER_OVERFLOW_ERROR is set, then
sl@0: * the return value indicates the necessary destination buffer size.
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: U_STABLE int32_t U_EXPORT2
sl@0: u_shapeArabic(const UChar *source, int32_t sourceLength,
sl@0: UChar *dest, int32_t destSize,
sl@0: uint32_t options,
sl@0: UErrorCode *pErrorCode);
sl@0:
sl@0: /**
sl@0: * Memory option: allow the result to have a different length than the source.
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: #define U_SHAPE_LENGTH_GROW_SHRINK 0
sl@0:
sl@0: /**
sl@0: * Memory option: the result must have the same length as the source.
sl@0: * If more room is necessary, then try to consume spaces next to modified characters.
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: #define U_SHAPE_LENGTH_FIXED_SPACES_NEAR 1
sl@0:
sl@0: /**
sl@0: * Memory option: the result must have the same length as the source.
sl@0: * If more room is necessary, then try to consume spaces at the end of the text.
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: #define U_SHAPE_LENGTH_FIXED_SPACES_AT_END 2
sl@0:
sl@0: /**
sl@0: * Memory option: the result must have the same length as the source.
sl@0: * If more room is necessary, then try to consume spaces at the beginning of the text.
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: #define U_SHAPE_LENGTH_FIXED_SPACES_AT_BEGINNING 3
sl@0:
sl@0: /** Bit mask for memory options. @stable ICU 2.0 */
sl@0: #define U_SHAPE_LENGTH_MASK 3
sl@0:
sl@0:
sl@0: /** Direction indicator: the source is in logical (keyboard) order. @stable ICU 2.0 */
sl@0: #define U_SHAPE_TEXT_DIRECTION_LOGICAL 0
sl@0:
sl@0: /**
sl@0: * Direction indicator:
sl@0: * the source is in visual LTR order,
sl@0: * the leftmost displayed character stored first.
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: #define U_SHAPE_TEXT_DIRECTION_VISUAL_LTR 4
sl@0:
sl@0: /** Bit mask for direction indicators. @stable ICU 2.0 */
sl@0: #define U_SHAPE_TEXT_DIRECTION_MASK 4
sl@0:
sl@0:
sl@0: /** Letter shaping option: do not perform letter shaping. @stable ICU 2.0 */
sl@0: #define U_SHAPE_LETTERS_NOOP 0
sl@0:
sl@0: /** Letter shaping option: replace abstract letter characters by "shaped" ones. @stable ICU 2.0 */
sl@0: #define U_SHAPE_LETTERS_SHAPE 8
sl@0:
sl@0: /** Letter shaping option: replace "shaped" letter characters by abstract ones. @stable ICU 2.0 */
sl@0: #define U_SHAPE_LETTERS_UNSHAPE 0x10
sl@0:
sl@0: /**
sl@0: * Letter shaping option: replace abstract letter characters by "shaped" ones.
sl@0: * The only difference with U_SHAPE_LETTERS_SHAPE is that Tashkeel letters
sl@0: * are always "shaped" into the isolated form instead of the medial form
sl@0: * (selecting code points from the Arabic Presentation Forms-B block).
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: #define U_SHAPE_LETTERS_SHAPE_TASHKEEL_ISOLATED 0x18
sl@0:
sl@0: /** Bit mask for letter shaping options. @stable ICU 2.0 */
sl@0: #define U_SHAPE_LETTERS_MASK 0x18
sl@0:
sl@0:
sl@0: /** Digit shaping option: do not perform digit shaping. @stable ICU 2.0 */
sl@0: #define U_SHAPE_DIGITS_NOOP 0
sl@0:
sl@0: /**
sl@0: * Digit shaping option:
sl@0: * Replace European digits (U+0030...) by Arabic-Indic digits.
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: #define U_SHAPE_DIGITS_EN2AN 0x20
sl@0:
sl@0: /**
sl@0: * Digit shaping option:
sl@0: * Replace Arabic-Indic digits by European digits (U+0030...).
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: #define U_SHAPE_DIGITS_AN2EN 0x40
sl@0:
sl@0: /**
sl@0: * Digit shaping option:
sl@0: * Replace European digits (U+0030...) by Arabic-Indic digits if the most recent
sl@0: * strongly directional character is an Arabic letter
sl@0: * (u_charDirection() result U_RIGHT_TO_LEFT_ARABIC [AL]).U_LEFT_TO_RIGHT [L] or U_RIGHT_TO_LEFT [R]).
sl@0: * @stable ICU 2.0
sl@0: */
sl@0: #define U_SHAPE_DIGITS_ALEN2AN_INIT_LR 0x60
sl@0:
sl@0: /**
sl@0: * Digit shaping option:
sl@0: * Replace European digits (U+0030...) by Arabic-Indic digits if the most recent
sl@0: * strongly directional character is an Arabic letter
sl@0: * (u_charDirection() result U_RIGHT_TO_LEFT_ARABIC [AL]).