tesseract  4.0.0-1-g2a2b
tesseract::ValidateMyanmar Class Reference

#include <validate_myanmar.h>

Inheritance diagram for tesseract::ValidateMyanmar:
tesseract::Validator

Public Member Functions

 ValidateMyanmar (ViramaScript script, bool report_errors)
 
 ~ValidateMyanmar ()
 
- Public Member Functions inherited from tesseract::Validator
virtual ~Validator ()
 

Protected Member Functions

bool ConsumeGraphemeIfValid () override
 
Validator::CharClass UnicodeToCharClass (char32 ch) const override
 
- Protected Member Functions inherited from tesseract::Validator
 Validator (ViramaScript script, bool report_errors)
 
bool ValidateCleanAndSegmentInternal (GraphemeNormMode g_mode, const std::vector< char32 > &src, std::vector< std::vector< char32 >> *dest)
 
void MoveResultsToDest (GraphemeNormMode g_mode, std::vector< std::vector< char32 >> *dest)
 
bool IsSubscriptScript () const
 
bool CodeOnlyToOutput ()
 
void MultiCodePart (int length)
 
bool UseMultiCode (int length)
 
void ComputeClassCodes (const std::vector< char32 > &text)
 
void Clear ()
 

Additional Inherited Members

- Static Public Member Functions inherited from tesseract::Validator
static bool ValidateCleanAndSegment (GraphemeNormMode g_mode, bool report_errors, const std::vector< char32 > &src, std::vector< std::vector< char32 >> *dest)
 
static bool IsZeroWidthMark (char32 ch)
 
- Static Public Attributes inherited from tesseract::Validator
static const char32 kZeroWidthSpace = 0x200B
 
static const char32 kZeroWidthNonJoiner = 0x200C
 
static const char32 kZeroWidthJoiner = 0x200D
 
static const char32 kLeftToRightMark = 0x200E
 
static const char32 kRightToLeftMark = 0x200F
 
static const char32 kInvalid = 0xfffd
 
- Protected Types inherited from tesseract::Validator
enum  CharClass {
  CharClass::kConsonant = 'C', CharClass::kVowel = 'V', CharClass::kVirama = 'H', CharClass::kMatra = 'M',
  CharClass::kMatraPiece = 'P', CharClass::kVowelModifier = 'D', CharClass::kZeroWidthNonJoiner = 'z', CharClass::kZeroWidthJoiner = 'Z',
  CharClass::kVedicMark = 'v', CharClass::kNukta = 'N', CharClass::kRobat = 'R', CharClass::kOther = 'O',
  CharClass::kWhitespace = ' ', CharClass::kCombiner = 'c'
}
 
using IndicPair = std::pair< CharClass, char32 >
 
- Static Protected Member Functions inherited from tesseract::Validator
static std::unique_ptr< ValidatorScriptValidator (ViramaScript script, bool report_errors)
 
static ViramaScript MostFrequentViramaScript (const std::vector< char32 > &utf32)
 
static bool IsVirama (char32 unicode)
 
static bool IsVedicAccent (char32 unicode)
 
- Protected Attributes inherited from tesseract::Validator
ViramaScript script_
 
std::vector< IndicPaircodes_
 
std::vector< std::vector< char32 > > parts_
 
std::vector< char32output_
 
int codes_used_
 
int output_used_
 
bool report_errors_
 
- Static Protected Attributes inherited from tesseract::Validator
static const int kIndicCodePageSize = 128
 
static const char32 kMinIndicUnicode = 0x900
 
static const char32 kMaxSinhalaUnicode = 0xdff
 
static const char32 kMaxViramaScriptUnicode = 0x17ff
 
static const char32 kSinhalaVirama = 0xdca
 
static const char32 kMyanmarVirama = 0x1039
 
static const char32 kKhmerVirama = 0x17d2
 
static const char32 kJavaneseVirama = 0xa9c0
 
static const char32 kMaxJavaneseUnicode = 0xa9df
 

Detailed Description

Definition at line 9 of file validate_myanmar.h.

Constructor & Destructor Documentation

◆ ValidateMyanmar()

tesseract::ValidateMyanmar::ValidateMyanmar ( ViramaScript  script,
bool  report_errors 
)
inline

Definition at line 11 of file validate_myanmar.h.

12  : Validator(script, report_errors) {}
Validator(ViramaScript script, bool report_errors)
Definition: validator.h:136

◆ ~ValidateMyanmar()

tesseract::ValidateMyanmar::~ValidateMyanmar ( )
inline

Definition at line 13 of file validate_myanmar.h.

13 {}

Member Function Documentation

◆ ConsumeGraphemeIfValid()

bool tesseract::ValidateMyanmar::ConsumeGraphemeIfValid ( )
overrideprotectedvirtual

Implements tesseract::Validator.

Definition at line 13 of file validate_myanmar.cpp.

13  {
14  int num_codes = codes_.size();
15  if (codes_used_ == num_codes) return true;
16  // Other.
17  if (IsMyanmarOther(codes_[codes_used_].second)) {
18  UseMultiCode(1);
19  return true;
20  }
21  // Kinzi.
22  if (codes_used_ + 2 < num_codes && codes_[codes_used_].second == 0x1004 &&
23  codes_[codes_used_ + 1].second == kMyanmarAsat &&
24  codes_[codes_used_ + 2].second == kMyanmarVirama) {
27  if (UseMultiCode(3)) return true;
28  }
29  // Base consonant/vowel. NOTE that since everything in Myanmar appears to be
30  // optional, except the base, this is the only place where invalid input can
31  // be detected and false returned.
32  if (IsMyanmarLetter(codes_[codes_used_].second)) {
33  if (UseMultiCode(1)) return true;
34  } else {
35  if (report_errors_) {
36  tprintf("Invalid start of Myanmar syllable:0x%x\n",
37  codes_[codes_used_].second);
38  }
39  return false; // One of these is required.
40  }
41  if (ConsumeSubscriptIfPresent()) return true;
42  ConsumeOptionalSignsIfPresent();
43  // What we have consumed so far is a valid syllable.
44  return true;
45 }
std::vector< IndicPair > codes_
Definition: validator.h:232
static const char32 kMyanmarVirama
Definition: validator.h:223
DLLSYM void tprintf(const char *format,...)
Definition: tprintf.cpp:37
bool UseMultiCode(int length)
Definition: validator.h:196
#define ASSERT_HOST(x)
Definition: errcode.h:84

◆ UnicodeToCharClass()

Validator::CharClass tesseract::ValidateMyanmar::UnicodeToCharClass ( char32  ch) const
overrideprotectedvirtual

Implements tesseract::Validator.

Definition at line 55 of file validate_myanmar.cpp.


The documentation for this class was generated from the following files: