Home | History | Annotate | Download | only in common
      1 /*
      2 **********************************************************************
      3 *   Copyright (C) 1999-2002, International Business Machines
      4 *   Corporation and others.  All Rights Reserved.
      5 **********************************************************************
      6 *
      7 * File USC_IMPL.H
      8 *
      9 * Modification History:
     10 *
     11 *   Date        Name        Description
     12 *   07/08/2002  Eric Mader  Creation.
     13 ******************************************************************************
     14 */
     15 
     16 #ifndef USC_IMPL_H
     17 #define USC_IMPL_H
     18 #include "unicode/utypes.h"
     19 #include "unicode/uscript.h"
     20 
     21 /**
     22  * <code>UScriptRun</code> is used to find runs of characters in
     23  * the same script. It implements a simple iterator over an array
     24  * of characters. The iterator will resolve script-neutral characters
     25  * like punctuation into the script of the surrounding characters.
     26  *
     27  * The iterator will try to match paired punctuation. If it sees an
     28  * opening punctuation character, it will remember the script that
     29  * was assigned to that character, and assign the same script to the
     30  * matching closing punctuation.
     31  *
     32  * Scripts are chosen based on the <code>UScriptCode</code> enumeration.
     33  * No attempt is made to combine related scripts into a single run. In
     34  * particular, Hiragana, Katakana, and Han characters will appear in seperate
     35  * runs.
     36 
     37  * Here is an example of how to iterate over script runs:
     38  * <pre>
     39  * \code
     40  * void printScriptRuns(const UChar *text, int32_t length)
     41  * {
     42  *     UErrorCode error = U_ZERO_ERROR;
     43  *     UScriptRun *scriptRun = uscript_openRun(text, testLength, &error);
     44  *     int32_t start = 0, limit = 0;
     45  *     UScriptCode code = USCRIPT_INVALID_CODE;
     46  *
     47  *     while (uscript_nextRun(&start, &limit, &code)) {
     48  *         printf("Script '%s' from %d to %d.\n", uscript_getName(code), start, limit);
     49  *     }
     50  *
     51  *     uscript_closeRun(scriptRun);
     52  *  }
     53  * </pre>
     54  *
     55  * @draft ICU 2.2
     56  */
     57 struct UScriptRun;
     58 
     59 typedef struct UScriptRun UScriptRun;
     60 
     61 /**
     62  * Create a <code>UScriptRun</code> object for iterating over the given text. This object must
     63  * be freed using <code>uscript_closeRun()</code>. Note that this object does not copy the source text,
     64  * only the pointer to it. You must make sure that the pointer remains valid until you call
     65  * <code>uscript_closeRun()</code> or <code>uscript_setRunText()</code>.
     66  *
     67  * @param src is the address of the array of characters over which to iterate.
     68  *        if <code>src == NULL</code> and <code>length == 0</code>,
     69  *        an empty <code>UScriptRun</code> object will be returned.
     70  *
     71  * @param length is the number of characters over which to iterate.
     72  *
     73  * @param pErrorCode is a pointer to a valid <code>UErrorCode</code> value. If this value
     74  *        indicates a failure on entry, the function will immediately return.
     75  *        On exit the value will indicate the success of the operation.
     76  *
     77  * @return the address of <code>UScriptRun</code> object which will iterate over the text,
     78  *         or <code>NULL</code> if the operation failed.
     79  *
     80  * @draft ICU 2.2
     81  */
     82 U_CAPI UScriptRun * U_EXPORT2
     83 uscript_openRun(const UChar *src, int32_t length, UErrorCode *pErrorCode);
     84 
     85 /**
     86  * Frees the given <code>UScriptRun</code> object and any storage associated with it.
     87  * On return, scriptRun no longer points to a valid <code>UScriptRun</code> object.
     88  *
     89  * @param scriptRun is the <code>UScriptRun</code> object which will be freed.
     90  *
     91  * @draft ICU 2.2
     92  */
     93 U_CAPI void U_EXPORT2
     94 uscript_closeRun(UScriptRun *scriptRun);
     95 
     96 /**
     97  * Reset the <code>UScriptRun</code> object so that it will start iterating from
     98  * the beginning.
     99  *
    100  * @param scriptRun is the address of the <code>UScriptRun</code> object to be reset.
    101  *
    102  * @draft ICU 2.2
    103  */
    104 U_CAPI void U_EXPORT2
    105 uscript_resetRun(UScriptRun *scriptRun);
    106 
    107 /**
    108  * Change the text over which the given <code>UScriptRun</code> object iterates.
    109  *
    110  * @param scriptRun is the <code>UScriptRun</code> object which will be changed.
    111  *
    112  * @param src is the address of the new array of characters over which to iterate.
    113  *        If <code>src == NULL</code> and <code>length == 0</code>,
    114  *        the <code>UScriptRun</code> object will become empty.
    115  *
    116  * @param length is the new number of characters over which to iterate
    117  *
    118  * @param pErrorCode is a pointer to a valid <code>UErrorCode</code> value. If this value
    119  *        indicates a failure on entry, the function will immediately return.
    120  *        On exit the value will indicate the success of the operation.
    121  *
    122  * @draft ICU 2.2
    123  */
    124 U_CAPI void U_EXPORT2
    125 uscript_setRunText(UScriptRun *scriptRun, const UChar *src, int32_t length, UErrorCode *pErrorCode);
    126 
    127 /**
    128  * Advance the <code>UScriptRun</code> object to the next script run, return the start and limit
    129  * offsets, and the script of the run.
    130  *
    131  * @param scriptRun is the address of the <code>UScriptRun</code> object.
    132  *
    133  * @param pRunStart is a pointer to the variable to receive the starting offset of the next run.
    134  *        This pointer can be <code>NULL</code> if the value is not needed.
    135  *
    136  * @param pRunLimit is a pointer to the variable to receive the limit offset of the next run.
    137  *        This pointer can be <code>NULL</code> if the value is not needed.
    138  *
    139  * @param pRunScript is a pointer to the variable to receive the UScriptCode for the
    140  *        script of the current run. This pointer can be <code>NULL</code> if the value is not needed.
    141  *
    142  * @return true if there was another script run.
    143  *
    144  * @draft ICU 2.2
    145  */
    146 U_CAPI UBool U_EXPORT2
    147 uscript_nextRun(UScriptRun *scriptRun, int32_t *pRunStart, int32_t *pRunLimit, UScriptCode *pRunScript);
    148 
    149 #endif
    150