+// © 2016 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html
//
// file: rbbirb.cpp
//
-// Copyright (C) 2002-2005, International Business Machines Corporation and others.
+// Copyright (C) 2002-2011, International Business Machines Corporation and others.
// All Rights Reserved.
//
// This file contains the RBBIRuleBuilder class implementation. This is the main class for
#include "unicode/uchriter.h"
#include "unicode/parsepos.h"
#include "unicode/parseerr.h"
+
#include "cmemory.h"
#include "cstring.h"
-
#include "rbbirb.h"
#include "rbbinode.h"
-
#include "rbbiscan.h"
#include "rbbisetb.h"
#include "rbbitblb.h"
#include "rbbidata.h"
+#include "uassert.h"
U_NAMESPACE_BEGIN
//
//----------------------------------------------------------------------------------------
RBBIRuleBuilder::RBBIRuleBuilder(const UnicodeString &rules,
- UParseError &parseErr,
+ UParseError *parseErr,
UErrorCode &status)
- : fRules(rules)
+ : fRules(rules), fStrippedRules(rules)
{
fStatus = &status; // status is checked below
- fParseError = &parseErr;
+ fParseError = parseErr;
fDebugEnv = NULL;
#ifdef RBBI_DEBUG
fDebugEnv = getenv("U_RBBIDEBUG");
fSafeFwdTree = NULL;
fSafeRevTree = NULL;
fDefaultTree = &fForwardTree;
- fForwardTables = NULL;
- fReverseTables = NULL;
- fSafeFwdTables = NULL;
- fSafeRevTables = NULL;
+ fForwardTable = NULL;
fRuleStatusVals = NULL;
fChainRules = FALSE;
fLBCMNoChain = FALSE;
fRuleStatusVals = NULL;
fScanner = NULL;
fSetBuilder = NULL;
+ if (parseErr) {
+ uprv_memset(parseErr, 0, sizeof(UParseError));
+ }
if (U_FAILURE(status)) {
return;
delete fUSetNodes;
delete fSetBuilder;
- delete fForwardTables;
- delete fReverseTables;
- delete fSafeFwdTables;
- delete fSafeRevTables;
-
+ delete fForwardTable;
delete fForwardTree;
delete fReverseTree;
delete fSafeFwdTree;
return NULL;
}
- // Remove comments and whitespace from the rules to make it smaller.
- UnicodeString strippedRules((const UnicodeString&)RBBIRuleScanner::stripRules(fRules));
+ // Remove whitespace from the rules to make it smaller.
+ // The rule parser has already removed comments.
+ fStrippedRules = fScanner->stripRules(fStrippedRules);
// Calculate the size of each section in the data.
// Sizes here are padded up to a multiple of 8 for better memory alignment.
// without the padding.
//
int32_t headerSize = align8(sizeof(RBBIDataHeader));
- int32_t forwardTableSize = align8(fForwardTables->getTableSize());
- int32_t reverseTableSize = align8(fReverseTables->getTableSize());
- int32_t safeFwdTableSize = align8(fSafeFwdTables->getTableSize());
- int32_t safeRevTableSize = align8(fSafeRevTables->getTableSize());
+ int32_t forwardTableSize = align8(fForwardTable->getTableSize());
+ int32_t reverseTableSize = align8(fForwardTable->getSafeTableSize());
int32_t trieSize = align8(fSetBuilder->getTrieSize());
int32_t statusTableSize = align8(fRuleStatusVals->size() * sizeof(int32_t));
- int32_t rulesSize = align8((strippedRules.length()+1) * sizeof(UChar));
+ int32_t rulesSize = align8((fStrippedRules.length()+1) * sizeof(UChar));
- int32_t totalSize = headerSize + forwardTableSize + reverseTableSize
- + safeFwdTableSize + safeRevTableSize
+ int32_t totalSize = headerSize
+ + forwardTableSize
+ + reverseTableSize
+ statusTableSize + trieSize + rulesSize;
RBBIDataHeader *data = (RBBIDataHeader *)uprv_malloc(totalSize);
data->fMagic = 0xb1a0;
- data->fFormatVersion[0] = 3;
- data->fFormatVersion[1] = 1;
- data->fFormatVersion[2] = 0;
- data->fFormatVersion[3] = 0;
+ data->fFormatVersion[0] = RBBI_DATA_FORMAT_VERSION[0];
+ data->fFormatVersion[1] = RBBI_DATA_FORMAT_VERSION[1];
+ data->fFormatVersion[2] = RBBI_DATA_FORMAT_VERSION[2];
+ data->fFormatVersion[3] = RBBI_DATA_FORMAT_VERSION[3];
data->fLength = totalSize;
data->fCatCount = fSetBuilder->getNumCharCategories();
data->fFTable = headerSize;
data->fFTableLen = forwardTableSize;
- data->fRTable = data->fFTable + forwardTableSize;
+
+ data->fRTable = data->fFTable + data->fFTableLen;
data->fRTableLen = reverseTableSize;
- data->fSFTable = data->fRTable + reverseTableSize;
- data->fSFTableLen = safeFwdTableSize;
- data->fSRTable = data->fSFTable + safeFwdTableSize;
- data->fSRTableLen = safeRevTableSize;
- data->fTrie = data->fSRTable + safeRevTableSize;
+ data->fTrie = data->fRTable + data->fRTableLen;
data->fTrieLen = fSetBuilder->getTrieSize();
data->fStatusTable = data->fTrie + trieSize;
data->fStatusTableLen= statusTableSize;
data->fRuleSource = data->fStatusTable + statusTableSize;
- data->fRuleSourceLen = strippedRules.length() * sizeof(UChar);
+ data->fRuleSourceLen = fStrippedRules.length() * sizeof(UChar);
uprv_memset(data->fReserved, 0, sizeof(data->fReserved));
- fForwardTables->exportTable((uint8_t *)data + data->fFTable);
- fReverseTables->exportTable((uint8_t *)data + data->fRTable);
- fSafeFwdTables->exportTable((uint8_t *)data + data->fSFTable);
- fSafeRevTables->exportTable((uint8_t *)data + data->fSRTable);
+ fForwardTable->exportTable((uint8_t *)data + data->fFTable);
+ fForwardTable->exportSafeTable((uint8_t *)data + data->fRTable);
fSetBuilder->serializeTrie ((uint8_t *)data + data->fTrie);
int32_t *ruleStatusTable = (int32_t *)((uint8_t *)data + data->fStatusTable);
ruleStatusTable[i] = fRuleStatusVals->elementAti(i);
}
- strippedRules.extract((UChar *)((uint8_t *)data+data->fRuleSource), rulesSize/2+1, *fStatus);
+ fStrippedRules.extract((UChar *)((uint8_t *)data+data->fRuleSource), rulesSize/2+1, *fStatus);
return data;
}
-
-
-
-
//----------------------------------------------------------------------------------------
//
// createRuleBasedBreakIterator construct from source rules that are passed in
//----------------------------------------------------------------------------------------
BreakIterator *
RBBIRuleBuilder::createRuleBasedBreakIterator( const UnicodeString &rules,
- UParseError &parseError,
+ UParseError *parseError,
UErrorCode &status)
{
- // status checked below
-
//
// Read the input rules, generate a parse tree, symbol table,
// and list of all Unicode Sets referenced by the rules.
//
RBBIRuleBuilder builder(rules, parseError, status);
- builder.fScanner->parse();
if (U_FAILURE(status)) { // status checked here bcos build below doesn't
return NULL;
}
- //
- // UnicodeSet processing.
- // Munge the Unicode Sets to create a set of character categories.
- // Generate the mapping tables (TRIE) from input 32-bit characters to
- // the character categories.
- //
- builder.fSetBuilder->build();
+ RBBIDataHeader *data = builder.build(status);
+ if (U_FAILURE(status)) {
+ return nullptr;
+ }
//
- // Generate the DFA state transition table.
+ // Create a break iterator from the compiled rules.
+ // (Identical to creation from stored pre-compiled rules)
//
- builder.fForwardTables = new RBBITableBuilder(&builder, &builder.fForwardTree);
- builder.fReverseTables = new RBBITableBuilder(&builder, &builder.fReverseTree);
- builder.fSafeFwdTables = new RBBITableBuilder(&builder, &builder.fSafeFwdTree);
- builder.fSafeRevTables = new RBBITableBuilder(&builder, &builder.fSafeRevTree);
- if (U_SUCCESS(status)
- && (builder.fForwardTables == NULL || builder.fReverseTables == NULL ||
- builder.fSafeFwdTables == NULL || builder.fSafeRevTables == NULL))
- {
+ // status is checked after init in construction.
+ RuleBasedBreakIterator *This = new RuleBasedBreakIterator(data, status);
+ if (U_FAILURE(status)) {
+ delete This;
+ This = NULL;
+ }
+ else if(This == NULL) { // test for NULL
status = U_MEMORY_ALLOCATION_ERROR;
- return NULL;
}
+ return This;
+}
- builder.fForwardTables->build();
- builder.fReverseTables->build();
- builder.fSafeFwdTables->build();
- builder.fSafeRevTables->build();
+RBBIDataHeader *RBBIRuleBuilder::build(UErrorCode &status) {
if (U_FAILURE(status)) {
- return NULL;
+ return nullptr;
}
-#ifdef RBBI_DEBUG
- if (builder.fDebugEnv && uprv_strstr(builder.fDebugEnv, "states")) {
- builder.fForwardTables->printRuleStatusTable();
+ fScanner->parse();
+ if (U_FAILURE(status)) {
+ return nullptr;
}
-#endif
//
- // Package up the compiled data into a memory image
- // in the run-time format.
+ // UnicodeSet processing.
+ // Munge the Unicode Sets to create a set of character categories.
+ // Generate the mapping tables (TRIE) from input code points to
+ // the character categories.
//
- RBBIDataHeader *data = builder.flattenData(); // returns NULL if error
-
+ fSetBuilder->buildRanges();
//
- // Clean up the compiler related stuff
+ // Generate the DFA state transition table.
//
+ fForwardTable = new RBBITableBuilder(this, &fForwardTree, status);
+ if (fForwardTable == nullptr) {
+ status = U_MEMORY_ALLOCATION_ERROR;
+ return nullptr;
+ }
+
+ fForwardTable->buildForwardTable();
+ optimizeTables();
+ fForwardTable->buildSafeReverseTable(status);
+
+
+#ifdef RBBI_DEBUG
+ if (fDebugEnv && uprv_strstr(fDebugEnv, "states")) {
+ fForwardTable->printStates();
+ fForwardTable->printRuleStatusTable();
+ fForwardTable->printReverseTable();
+ }
+#endif
+ fSetBuilder->buildTrie();
//
- // Create a break iterator from the compiled rules.
- // (Identical to creation from stored pre-compiled rules)
+ // Package up the compiled data into a memory image
+ // in the run-time format.
//
- // status is checked after init in construction.
- RuleBasedBreakIterator *This = new RuleBasedBreakIterator(data, status);
+ RBBIDataHeader *data = flattenData(); // returns NULL if error
if (U_FAILURE(status)) {
- delete This;
- This = NULL;
- }
- else if(This == NULL) { // test for NULL
- status = U_MEMORY_ALLOCATION_ERROR;
+ return nullptr;
}
- return This;
+ return data;
+}
+
+void RBBIRuleBuilder::optimizeTables() {
+ bool didSomething;
+ do {
+ didSomething = false;
+
+ // Begin looking for duplicates with char class 3.
+ // Classes 0, 1 and 2 are special; they are unused, {bof} and {eof} respectively,
+ // and should not have other categories merged into them.
+ IntPair duplPair = {3, 0};
+ while (fForwardTable->findDuplCharClassFrom(&duplPair)) {
+ fSetBuilder->mergeCategories(duplPair);
+ fForwardTable->removeColumn(duplPair.second);
+ didSomething = true;
+ }
+
+ while (fForwardTable->removeDuplicateStates() > 0) {
+ didSomething = true;
+ }
+ } while (didSomething);
}
U_NAMESPACE_END
projects / apple / icu.git / blobdiff