ICU-1428 change to get scrptrun from extra/scrptrun instead of using a local ocopy

X-SVN-Rev: 7090
This commit is contained in:
Eric Mader 2001-11-27 00:52:51 +00:00
parent 10e4887cde
commit 9534b7c75b
3 changed files with 14 additions and 335 deletions

View File

@ -23,7 +23,7 @@ CLEANFILES = *~ $(DEPS)
TARGET = gnomelayout
DEFS = @DEFS@
CPPFLAGS = @CPPFLAGS@ `gnome-config --cflags gnomeui` -I$(top_builddir)/common -I$(top_srcdir)/common -I$(top_srcdir)/i18n -I$(top_srcdir)/layout -I$(top_srcdir)
CPPFLAGS = @CPPFLAGS@ `gnome-config --cflags gnomeui` -I$(top_builddir)/common -I$(top_srcdir)/common -I$(top_srcdir)/i18n -I$(top_srcdir)/layout -I$(top_srcdir)/extra/scrptrun -I$(top_srcdir)
CFLAGS = @CFLAGS@
CXXFLAGS = @CXXFLAGS@
ENABLE_RPATH = @ENABLE_RPATH@
@ -73,6 +73,19 @@ Makefile: $(srcdir)/Makefile.in $(top_builddir)/config.status
cd $(top_builddir) \
&& CONFIG_FILES=$(subdir)/$@ CONFIG_HEADERS= $(SHELL) ./config.status
# The following two rules make it possible to
# compile scrptrun.cpp from the extra/scrptrun directory.
# they were copied from the default rules in mh-linux which
# is probably OK because this sample will only run on Linux...
scrptrun.d: $(top_srcdir)/extra/scrptrun/scrptrun.cpp
@echo "Generating dependency information for $<"
@$(SHELL) -ec '$(GEN_DEPS.cc) $< \
| sed '\''s%\($*\)\.o[ :]*%\1.o $@ : %g'\'' > $@; \
[ -s $@ ] || rm -f $@'
scrptrun.o: $(top_srcdir)/extra/scrptrun/scrptrun.cpp
$(COMPILE.cc) $(DYNAMICCPPFLAGS) $(DYNAMICCXXFLAGS) -o $@ $<
$(TARGET) : $(OBJECTS)
$(LINK.cc) -o $@ $^ $(LIBS)

View File

@ -1,200 +0,0 @@
/*
*******************************************************************************
*
* Copyright (C) 1999-2001, International Business Machines
* Corporation and others. All Rights Reserved.
*
*******************************************************************************
* file name: scrptrun.cpp
*
* created on: 10/17/2001
* created by: Eric R. Mader
*/
#include "unicode/utypes.h"
#include "unicode/uscript.h"
#include "scrptrun.h"
#define ARRAY_SIZE(array) (sizeof array / sizeof array[0])
UChar32 ScriptRun::pairedChars[] = {
0x0028, 0x0029, // ascii paired punctuation
0x003c, 0x003e,
0x005b, 0x005d,
0x007b, 0x007d,
0x00ab, 0x00bb, // guillemets
0x2018, 0x2019, // general punctuation
0x201c, 0x201d,
0x2039, 0x203a,
0x3008, 0x3009, // chinese paired punctuation
0x300a, 0x300b,
0x300c, 0x300d,
0x300e, 0x300f,
0x3010, 0x3011,
0x3014, 0x3015,
0x3016, 0x3017,
0x3018, 0x3019,
0x301a, 0x301b
};
const int32_t ScriptRun::pairedCharCount = ARRAY_SIZE(pairedChars);
const int32_t ScriptRun::pairedCharPower = 1 << highBit(pairedCharCount);
const int32_t ScriptRun::pairedCharExtra = pairedCharCount - pairedCharPower;
int8_t ScriptRun::highBit(int32_t value)
{
if (value <= 0) {
return -32;
}
int8_t bit = 0;
if (value >= 1 << 16) {
value >>= 16;
bit += 16;
}
if (value >= 1 << 8) {
value >>= 8;
bit += 8;
}
if (value >= 1 << 4) {
value >>= 4;
bit += 4;
}
if (value >= 1 << 2) {
value >>= 2;
bit += 2;
}
if (value >= 1 << 1) {
value >>= 1;
bit += 1;
}
return bit;
}
int32_t ScriptRun::getPairIndex(UChar32 ch)
{
int32_t probe = pairedCharPower;
int32_t index = 0;
if (ch >= pairedChars[pairedCharExtra]) {
index = pairedCharExtra;
}
while (probe > (1 << 0)) {
probe >>= 1;
if (ch >= pairedChars[index + probe]) {
index += probe;
}
}
if (pairedChars[index] != ch) {
index = -1;
}
return index;
}
UBool ScriptRun::sameScript(int32_t scriptOne, int32_t scriptTwo)
{
return scriptOne <= USCRIPT_INHERITED || scriptTwo <= USCRIPT_INHERITED || scriptOne == scriptTwo;
}
UBool ScriptRun::next()
{
int32_t startSP = parenSP; // used to find the first new open character
UErrorCode error = U_ZERO_ERROR;
// if we've fallen off the end of the text, we're done
if (scriptEnd >= charLimit) {
return false;
}
scriptCode = USCRIPT_COMMON;
for (scriptStart = scriptEnd; scriptEnd < charLimit; scriptEnd += 1) {
UChar high = charArray[scriptEnd];
UChar32 ch = high;
// if the character is a high surrogate and it's not the last one
// in the text, see if it's followed by a low surrogate
if (high >= 0xD800 && high <= 0xDBFF && scriptEnd < charLimit - 1)
{
UChar low = charArray[scriptEnd + 1];
// if it is followed by a low surrogate,
// consume it and form the full character
if (low >= 0xDC00 && low <= 0xDFFF) {
ch = (high - 0xD800) * 0x0400 + low - 0xDC00 + 0x10000;
scriptEnd += 1;
}
}
UScriptCode sc = uscript_getScript(ch, &error);
int32_t pairIndex = getPairIndex(ch);
// Paired character handling:
//
// if it's an open character, push it onto the stack.
// if it's a close character, find the matching open on the
// stack, and use that script code. Any non-matching open
// characters above it on the stack will be poped.
if (pairIndex >= 0) {
if ((pairIndex & 1) == 0) {
parenStack[++parenSP].pairIndex = pairIndex;
parenStack[parenSP].scriptCode = scriptCode;
} else if (parenSP >= 0) {
int32_t pi = pairIndex & ~1;
while (parenSP >= 0 && parenStack[parenSP].pairIndex != pi) {
parenSP -= 1;
}
if (parenSP < startSP) {
startSP = parenSP;
}
if (parenSP >= 0) {
sc = parenStack[parenSP].scriptCode;
}
}
}
if (sameScript(scriptCode, sc)) {
if (scriptCode <= USCRIPT_INHERITED && sc > USCRIPT_INHERITED) {
scriptCode = sc;
// now that we have a final script code, fix any open
// characters we pushed before we knew the script code.
while (startSP < parenSP) {
parenStack[++startSP].scriptCode = scriptCode;
}
}
// if this character is a close paired character,
// pop it from the stack
if (pairIndex >= 0 && (pairIndex & 1) != 0 && parenSP >= 0) {
parenSP -= 1;
startSP -= 1;
}
} else {
// if the run broke on a surrogate pair,
// end it before the high surrogate
if (ch >= 0x10000) {
scriptEnd -= 1;
}
break;
}
}
return true;
}

View File

@ -1,134 +0,0 @@
/*
*******************************************************************************
*
* Copyright (C) 1999-2001, International Business Machines
* Corporation and others. All Rights Reserved.
*
*******************************************************************************
* file name: scrptrun.h
*
* created on: 10/17/2001
* created by: Eric R. Mader
*/
#ifndef __SCRPTRUN_H
#define __SCRPTRUN_H
#include "unicode/utypes.h"
#include "unicode/uscript.h"
struct ScriptRecord
{
UChar32 startChar;
UChar32 endChar;
UScriptCode scriptCode;
};
struct ParenStackEntry
{
int32_t pairIndex;
UScriptCode scriptCode;
};
class ScriptRun
{
public:
ScriptRun();
ScriptRun(const UChar chars[], int32_t length);
ScriptRun(const UChar chars[], int32_t start, int32_t length);
void reset();
void reset(int32_t start, int32_t count);
void reset(const UChar chars[], int32_t start, int32_t length);
int32_t getScriptStart();
int32_t getScriptEnd();
UScriptCode getScriptCode();
UBool next();
private:
static UBool sameScript(int32_t scriptOne, int32_t scriptTwo);
int32_t charStart;
int32_t charLimit;
const UChar *charArray;
int32_t scriptStart;
int32_t scriptEnd;
UScriptCode scriptCode;
ParenStackEntry parenStack[128];
int32_t parenSP;
static int8_t highBit(int32_t value);
static int32_t getPairIndex(UChar32 ch);
static UChar32 pairedChars[];
static const int32_t pairedCharCount;
static const int32_t pairedCharPower;
static const int32_t pairedCharExtra;
};
inline ScriptRun::ScriptRun()
{
reset(NULL, 0, 0);
}
inline ScriptRun::ScriptRun(const UChar chars[], int32_t length)
{
reset(chars, 0, length);
}
inline ScriptRun::ScriptRun(const UChar chars[], int32_t start, int32_t length)
{
reset(chars, start, length);
}
inline int32_t ScriptRun::getScriptStart()
{
return scriptStart;
}
inline int32_t ScriptRun::getScriptEnd()
{
return scriptEnd;
}
inline UScriptCode ScriptRun::getScriptCode()
{
return scriptCode;
}
inline void ScriptRun::reset()
{
scriptStart = charStart;
scriptEnd = charStart;
scriptCode = USCRIPT_INVALID_CODE;
parenSP = -1;
}
inline void ScriptRun::reset(int32_t start, int32_t length)
{
charStart = start;
charLimit = start + length;
reset();
}
inline void ScriptRun::reset(const UChar chars[], int32_t start, int32_t length)
{
charArray = chars;
reset(start, length);
}
#endif