tesseract  v4.0.0-17-g361f3264
Open Source OCR Engine
validate_indic.h
1 #ifndef TESSERACT_TRAINING_VALIDATE_INDIC_H_
2 #define TESSERACT_TRAINING_VALIDATE_INDIC_H_
3 
4 #include "validator.h"
5 
6 namespace tesseract {
7 
8 // Subclass of Validator that validates and segments Indic scripts in the
9 // unicode range 0x900-0xdff (Devanagari-Sinhala).
10 class ValidateIndic : public Validator {
11  public:
12  ValidateIndic(ViramaScript script, bool report_errors)
13  : Validator(script, report_errors) {}
15 
16  protected:
17  // Returns whether codes matches the pattern for an Indic Grapheme.
18  // Consumes the next Grapheme in codes_[codes_used_++...] and copies it to
19  // parts_ and output_. Returns true if a valid Grapheme was consumed,
20  // otherwise does not increment codes_used_.
21  bool ConsumeGraphemeIfValid() override;
22  // Returns the CharClass corresponding to the given Unicode ch.
24 
25  private:
26  // Helper consumes/copies a virama and any associated post-virama joiners.
27  bool ConsumeViramaIfValid(IndicPair joiner, bool post_matra);
28  // Helper consumes/copies a series of consonants separated by viramas while
29  // valid, but not any vowel or other modifiers.
31  // Helper consumes/copies a tail part of a consonant, comprising optional
32  // matra/piece, vowel modifier, vedic mark, terminating virama.
34  // Helper consumes/copies a vowel and optional modifiers.
35  bool ConsumeVowelIfValid();
36 
37  // Some special unicodes used only for Indic processing.
38  static const char32 kYayana = 0xdba; // Sinhala Ya
39  static const char32 kRayana = 0xdbb; // Sinhala Ra
40 };
41 
42 } // namespace tesseract
43 
44 #endif // TESSERACT_TRAINING_VALIDATE_INDIC_H_
bool ConsumeVowelIfValid()
Definition: validate_indic.cpp:261
Validator::CharClass UnicodeToCharClass(char32 ch) const override
Definition: validate_indic.cpp:44
ValidateIndic(ViramaScript script, bool report_errors)
Definition: validate_indic.h:12
signed int char32
Definition: unichar.h:52
static const char32 kRayana
Definition: validate_indic.h:39
Definition: baseapi.cpp:94
Definition: validate_indic.h:10
std::pair< CharClass, char32 > IndicPair
Definition: validator.h:134
ViramaScript
Definition: validator.h:53
bool ConsumeGraphemeIfValid() override
Definition: validate_indic.cpp:18
CharClass
Definition: validator.h:112
bool ConsumeViramaIfValid(IndicPair joiner, bool post_matra)
Definition: validate_indic.cpp:97
bool ConsumeConsonantHeadIfValid()
Definition: validate_indic.cpp:169
static const char32 kYayana
Definition: validate_indic.h:38
~ValidateIndic()
Definition: validate_indic.h:14
bool ConsumeConsonantTailIfValid()
Definition: validate_indic.cpp:231
Definition: validator.h:72