static xmlChar * xmlPatScanNCName(xmlPatParserContextPtr ctxt) { const xmlChar *q, *cur; xmlChar *ret = NULL; int val, len; SKIP_BLANKS; cur = q = CUR_PTR; val = xmlStringCurrentChar(NULL, cur, &len); if (!IS_LETTER(val) && (val != '_')) return(NULL); while ((IS_LETTER(val)) || (IS_DIGIT(val)) || (val == '.') || (val == '-') || (val == '_') || (IS_COMBINING(val)) || (IS_EXTENDER(val))) { cur += len; val = xmlStringCurrentChar(NULL, cur, &len); } ret = xmlStrndup(q, cur - q); CUR_PTR = cur; return(ret); }
void doinsert(ZLE_STRING_T zstr, int len) { ZLE_STRING_T s; ZLE_CHAR_T c1 = *zstr; /* first character */ int neg = zmult < 0; /* insert *after* the cursor? */ int m = neg ? -zmult : zmult; /* number of copies to insert */ int count; UNMETACHECK(); iremovesuffix(c1, 0); invalidatelist(); /* In overwrite mode, don't replace newlines. */ if (insmode || zleline[zlecs] == ZWC('\n')) spaceinline(m * len); else { int pos = zlecs, diff, i; #ifdef MULTIBYTE_SUPPORT /* * Calculate the number of character positions we are * going to be using. The algorithm is that * anything that shows up as a logical single character * (i.e. even if control, or double width, or with combining * characters) is treated as 1 for the purpose of replacing * what's there already. * * This can cause inserting of a combining character in * places where it should overwrite, such as the start * of a line. However, combining characters aren't * useful there anyway and this doesn't cause any * particular harm. */ for (i = 0, count = 0; i < len * m; i++) { if (!IS_COMBINING(zstr[i])) count++; } #else count = len * m; #endif /* * Ensure we replace a complete combining characterfor each * character we overwrite. Switch to inserting at first newline. */ for (i = count; pos < zlell && zleline[pos] != ZWC('\n') && i--; ) { INCPOS(pos); } /* * Calculate how many raw line places we need. * pos - zlecs is the raw line distance we're replacing, * m * len the number we're inserting. */ diff = pos - zlecs - m * len; if (diff < 0) { spaceinline(-diff); } else if (diff > 0) { /* * We use shiftchars() here because we don't * want combining char alignment fixed up: we * are going to write over any that remain. */ shiftchars(zlecs, diff); } } while (m--) for (s = zstr, count = len; count; s++, count--) zleline[zlecs++] = *s; if (neg) zlecs += zmult * len; /* if we ended up on a combining character, skip over it */ CCRIGHT(); }
static int32_t doWriteReverse(const UChar *src, int32_t srcLength, UChar *dest, int32_t destSize, uint16_t options, UErrorCode *pErrorCode) { /* * RTL run - * * RTL runs need to be copied to the destination in reverse order * of code points, not code units, to keep Unicode characters intact. * * The general strategy for this is to read the source text * in backward order, collect all code units for a code point * (and optionally following combining characters, see below), * and copy all these code units in ascending order * to the destination for this run. * * Several options request whether combining characters * should be kept after their base characters, * whether BiDi control characters should be removed, and * whether characters should be replaced by their mirror-image * equivalent Unicode characters. */ int32_t i, j; UChar32 c; /* optimize for several combinations of options */ switch(options&(UBIDI_REMOVE_BIDI_CONTROLS|UBIDI_DO_MIRRORING|UBIDI_KEEP_BASE_COMBINING)) { case 0: /* * With none of the "complicated" options set, the destination * run will have the same length as the source run, * and there is no mirroring and no keeping combining characters * with their base characters. */ if(destSize<srcLength) { *pErrorCode=U_BUFFER_OVERFLOW_ERROR; return srcLength; } destSize=srcLength; /* preserve character integrity */ do { /* i is always after the last code unit known to need to be kept in this segment */ i=srcLength; /* collect code units for one base character */ UTF_BACK_1(src, 0, srcLength); /* copy this base character */ j=srcLength; do { *dest++=src[j++]; } while(j<i); } while(srcLength>0); break; case UBIDI_KEEP_BASE_COMBINING: /* * Here, too, the destination * run will have the same length as the source run, * and there is no mirroring. * We do need to keep combining characters with their base characters. */ if(destSize<srcLength) { *pErrorCode=U_BUFFER_OVERFLOW_ERROR; return srcLength; } destSize=srcLength; /* preserve character integrity */ do { /* i is always after the last code unit known to need to be kept in this segment */ i=srcLength; /* collect code units and modifier letters for one base character */ do { UTF_PREV_CHAR(src, 0, srcLength, c); } while(srcLength>0 && IS_COMBINING(u_charType(c))); /* copy this "user character" */ j=srcLength; do { *dest++=src[j++]; } while(j<i); } while(srcLength>0); break; default: /* * With several "complicated" options set, this is the most * general and the slowest copying of an RTL run. * We will do mirroring, remove BiDi controls, and * keep combining characters with their base characters * as requested. */ if(!(options&UBIDI_REMOVE_BIDI_CONTROLS)) { i=srcLength; } else { /* we need to find out the destination length of the run, which will not include the BiDi control characters */ int32_t length=srcLength; UChar ch; i=0; do { ch=*src++; if(!IS_BIDI_CONTROL_CHAR(ch)) { ++i; } } while(--length>0); src-=srcLength; } if(destSize<i) { *pErrorCode=U_BUFFER_OVERFLOW_ERROR; return i; } destSize=i; /* preserve character integrity */ do { /* i is always after the last code unit known to need to be kept in this segment */ i=srcLength; /* collect code units for one base character */ UTF_PREV_CHAR(src, 0, srcLength, c); if(options&UBIDI_KEEP_BASE_COMBINING) { /* collect modifier letters for this base character */ while(srcLength>0 && IS_COMBINING(u_charType(c))) { UTF_PREV_CHAR(src, 0, srcLength, c); } } if(options&UBIDI_REMOVE_BIDI_CONTROLS && IS_BIDI_CONTROL_CHAR(c)) { /* do not copy this BiDi control character */ continue; } /* copy this "user character" */ j=srcLength; if(options&UBIDI_DO_MIRRORING) { /* mirror only the base character */ int32_t k=0; c=u_charMirror(c); UTF_APPEND_CHAR_UNSAFE(dest, k, c); dest+=k; j+=k; } while(j<i) { *dest++=src[j++]; } } while(srcLength>0); break; } /* end of switch */ return destSize; }