tesseract  4.0.0-1-g2a2b
ligature_table.h
Go to the documentation of this file.
1 /**********************************************************************
2  * File: ligature_table.h
3  * Description: Class for adding and removing optional latin ligatures,
4  * conditional on codepoint support by a specified font
5  * (if specified).
6  * Author: Ranjith Unnikrishnan
7  * Created: Mon Nov 18 2013
8  *
9  * (C) Copyright 2013, Google Inc.
10  * Licensed under the Apache License, Version 2.0 (the "License");
11  * you may not use this file except in compliance with the License.
12  * You may obtain a copy of the License at
13  * http://www.apache.org/licenses/LICENSE-2.0
14  * Unless required by applicable law or agreed to in writing, software
15  * distributed under the License is distributed on an "AS IS" BASIS,
16  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
17  * See the License for the specific language governing permissions and
18  * limitations under the License.
19  *
20  **********************************************************************/
21 
22 #ifndef TRAININGDATA_LIGATURE_TABLE_H_
23 #define TRAININGDATA_LIGATURE_TABLE_H_
24 
25 #include <memory>
26 #include <string>
27 #include <unordered_map>
28 
29 #include "util.h"
30 
31 namespace tesseract {
32 
33 class PangoFontInfo; // defined in pango_font_info.h
34 
35 // Map to substitute strings for ligatures.
36 using LigHash = std::unordered_map<std::string, std::string, StringHash>;
37 
39  public:
40  // Get a static instance of this class.
41  static LigatureTable* Get();
42 
43  // Convert the utf8 string so that ligaturizable sequences, such as "fi" get
44  // replaced by the (utf8 code for) appropriate ligature characters. Only do so
45  // if the corresponding ligature character is renderable in the current font.
46  std::string AddLigatures(const std::string& str, const PangoFontInfo* font) const;
47  // Remove all ligatures.
48  std::string RemoveLigatures(const std::string& str) const;
49  // Remove only custom ligatures (eg. "ct") encoded in the private-use-area.
50  std::string RemoveCustomLigatures(const std::string& str) const;
51 
52  const LigHash& norm_to_lig_table() const {
53  return norm_to_lig_table_;
54  }
55  const LigHash& lig_to_norm_table() const {
56  return lig_to_norm_table_;
57  }
58 
59  protected:
60  LigatureTable();
61  // Initialize the hash tables mapping between ligature strings and the
62  // corresponding ligature characters.
63  void Init();
64 
65  static std::unique_ptr<LigatureTable> instance_;
72 
73  private:
75  void operator=(const LigatureTable&);
76 };
77 
78 } // namespace tesseract
79 
80 #endif // OCR_TRAININGDATA_TYPESETTING_LIGATURE_TABLE_H_
std::unordered_map< std::string, std::string, StringHash > LigHash
std::string RemoveLigatures(const std::string &str) const
static std::unique_ptr< LigatureTable > instance_
static LigatureTable * Get()
const LigHash & lig_to_norm_table() const
std::string AddLigatures(const std::string &str, const PangoFontInfo *font) const
std::string RemoveCustomLigatures(const std::string &str) const
const LigHash & norm_to_lig_table() const