Remove old code for string class (no longer needed) (#1354)

* Remove old code for string class (no longer needed)

Signed-off-by: Stefan Weil <sw@weilnetz.de>

* Add std namespace to string class

Signed-off-by: Stefan Weil <sw@weilnetz.de>

ccstruct/ratngs.cpp

@@ -201,7 +201,7 @@ WERD_CHOICE::WERD_CHOICE(const char *src_string,
     : unicharset_(&unicharset){
   GenericVector<UNICHAR_ID> encoding;
   GenericVector<char> lengths;
-  string cleaned = unicharset.CleanupString(src_string);
+  std::string cleaned = unicharset.CleanupString(src_string);
   if (unicharset.encode_string(cleaned.c_str(), true, &encoding, &lengths,
                                NULL)) {
     lengths.push_back('\0');

ccutil/platform.h

@@ -53,12 +53,6 @@
 #define SIGNED signed
 #endif
 
-// Fix to map between google use of string without std and everywhere else.
-#ifdef USE_STD_NAMESPACE
-#include <string>
-using std::string;
-#endif
-
 #if defined(_WIN32) || defined(__CYGWIN__)
 #ifndef M_PI
 #define M_PI 3.14159265358979323846

ccutil/tessdatamanager.cpp

@@ -172,13 +172,13 @@ bool TessdataManager::GetComponent(TessdataType type, TFile *fp) const {
 }
 
 // Returns the current version string.
-string TessdataManager::VersionString() const {
-  return string(&entries_[TESSDATA_VERSION][0],
-                entries_[TESSDATA_VERSION].size());
+std::string TessdataManager::VersionString() const {
+  return std::string(&entries_[TESSDATA_VERSION][0],
+                     entries_[TESSDATA_VERSION].size());
 }
 
 // Sets the version string to the given v_str.
-void TessdataManager::SetVersionString(const string &v_str) {
+void TessdataManager::SetVersionString(const std::string &v_str) {
   entries_[TESSDATA_VERSION].resize_no_init(v_str.size());
   memcpy(&entries_[TESSDATA_VERSION][0], v_str.data(), v_str.size());
 }

ccutil/tessdatamanager.h

@@ -178,9 +178,9 @@ class TessdataManager {
   bool GetComponent(TessdataType type, TFile *fp) const;
 
   // Returns the current version string.
-  string VersionString() const;
+  std::string VersionString() const;
   // Sets the version string to the given v_str.
-  void SetVersionString(const string &v_str);
+  void SetVersionString(const std::string &v_str);
 
   // Returns true if the base Tesseract components are present.
   bool IsBaseAvailable() const {

ccutil/unichar.cpp

@@ -227,8 +227,8 @@ std::vector<char32> UNICHAR::UTF8ToUTF32(const char* utf8_str) {
 }
 
 // Returns an empty string if the input contains an invalid unicode.
-string UNICHAR::UTF32ToUTF8(const std::vector<char32>& str32) {
-  string utf8_str;
+std::string UNICHAR::UTF32ToUTF8(const std::vector<char32>& str32) {
+  std::string utf8_str;
   for (char32 ch : str32) {
     UNICHAR uni_ch(ch);
     int step;

ccutil/unicharcompress.cpp

@@ -133,7 +133,7 @@ bool UnicharCompress::ComputeEncoding(const UNICHARSET& unicharset, int null_id,
     RecodedCharID code;
     // Convert to unicodes.
     std::vector<char32> unicodes;
-    string cleaned;
+    std::string cleaned;
     if (u < unicharset.size())
       cleaned = UNICHARSET::CleanupString(unicharset.id_to_unichar(u));
     if (u < unicharset.size() &&

ccutil/unicharset.cpp

@@ -205,7 +205,7 @@ void UNICHARSET::reserve(int unichars_number) {
 
 UNICHAR_ID
 UNICHARSET::unichar_to_id(const char* const unichar_repr) const {
-  string cleaned =
+  std::string cleaned =
       old_style_included_ ? unichar_repr : CleanupString(unichar_repr);
   return ids.contains(cleaned.data(), cleaned.size())
              ? ids.unichar_to_id(cleaned.data(), cleaned.size())
@@ -215,7 +215,7 @@ UNICHARSET::unichar_to_id(const char* const unichar_repr) const {
 UNICHAR_ID UNICHARSET::unichar_to_id(const char* const unichar_repr,
                                      int length) const {
   assert(length > 0 && length <= UNICHAR_LEN);
-  string cleaned(unichar_repr, length);
+  std::string cleaned(unichar_repr, length);
   if (!old_style_included_) cleaned = CleanupString(unichar_repr, length);
   return ids.contains(cleaned.data(), cleaned.size())
              ? ids.unichar_to_id(cleaned.data(), cleaned.size())
@@ -623,7 +623,7 @@ char UNICHARSET::get_chartype(UNICHAR_ID id) const {
 void UNICHARSET::unichar_insert(const char* const unichar_repr,
                                 OldUncleanUnichars old_style) {
   if (old_style == OldUncleanUnichars::kTrue) old_style_included_ = true;
-  string cleaned =
+  std::string cleaned =
       old_style_included_ ? unichar_repr : CleanupString(unichar_repr);
   if (!cleaned.empty() && !ids.contains(cleaned.data(), cleaned.size())) {
     const char* str = cleaned.c_str();
@@ -666,7 +666,7 @@ void UNICHARSET::unichar_insert(const char* const unichar_repr,
 }
 
 bool UNICHARSET::contains_unichar(const char* const unichar_repr) const {
-  string cleaned =
+  std::string cleaned =
       old_style_included_ ? unichar_repr : CleanupString(unichar_repr);
   return ids.contains(cleaned.data(), cleaned.size());
 }
@@ -676,7 +676,7 @@ bool UNICHARSET::contains_unichar(const char* const unichar_repr,
   if (length == 0) {
     return false;
   }
-  string cleaned(unichar_repr, length);
+  std::string cleaned(unichar_repr, length);
   if (!old_style_included_) cleaned = CleanupString(unichar_repr, length);
   return ids.contains(cleaned.data(), cleaned.size());
 }
@@ -1115,8 +1115,8 @@ int UNICHARSET::get_script_id_from_name(const char* script_name) const {
 // Removes/replaces content that belongs in rendered text, but not in the
 // unicharset.
 /* static */
-string UNICHARSET::CleanupString(const char* utf8_str, int length) {
-  string result;
+std::string UNICHARSET::CleanupString(const char* utf8_str, int length) {
+  std::string result;
   result.reserve(length);
   char ch;
   while ((ch = *utf8_str) != '\0' && --length >= 0) {

lstm/lstmtrainer.cpp

@@ -756,7 +756,7 @@ bool LSTMTrainer::EncodeString(const STRING& str, const UNICHARSET& unicharset,
   GenericVector<int> internal_labels;
   labels->truncate(0);
   if (!simple_text) labels->push_back(null_char);
-  string cleaned = unicharset.CleanupString(str.string());
+  std::string cleaned = unicharset.CleanupString(str.string());
   if (unicharset.encode_string(cleaned.c_str(), true, &internal_labels, NULL,
                                &err_index)) {
     bool success = true;

lstm/lstmtrainer.h

@@ -106,7 +106,7 @@ class LSTMTrainer : public LSTMRecognizer {
   // Initializes the character set encode/decode mechanism directly from a
   // previously setup traineddata containing dawgs, UNICHARSET and
   // UnicharCompress. Note: Call before InitNetwork!
-  void InitCharSet(const string& traineddata_path) {
+  void InitCharSet(const std::string& traineddata_path) {
     ASSERT_HOST(mgr_.Init(traineddata_path.c_str()));
     InitCharSet();
   }

training/boxchar.cpp

@@ -318,16 +318,16 @@ void BoxChar::RotateBoxes(float rotation, int xcenter, int ycenter,
 
 const int kMaxLineLength = 1024;
 /* static */
-void BoxChar::WriteTesseractBoxFile(const string& filename, int height,
+void BoxChar::WriteTesseractBoxFile(const std::string& filename, int height,
                                     const std::vector<BoxChar*>& boxes) {
-  string output = GetTesseractBoxStr(height, boxes);
+  std::string output = GetTesseractBoxStr(height, boxes);
   File::WriteStringToFileOrDie(output, filename);
 }
 
 /* static */
-string BoxChar::GetTesseractBoxStr(int height,
-                                   const std::vector<BoxChar*>& boxes) {
-  string output;
+std::string BoxChar::GetTesseractBoxStr(int height,
+                                        const std::vector<BoxChar*>& boxes) {
+  std::string output;
   char buffer[kMaxLineLength];
   for (size_t i = 0; i < boxes.size(); ++i) {
     const Box* box = boxes[i]->box_;

training/boxchar.h

@@ -40,7 +40,7 @@ class BoxChar {
   ~BoxChar();
 
   // Accessors.
-  const string& ch() const { return ch_; }
+  const std::string& ch() const { return ch_; }
   const Box* box() const   { return box_; }
   const int& page() const  { return page_; }
   void set_rtl_index(int index) { rtl_index_ = index; }
@@ -51,7 +51,7 @@ class BoxChar {
 
   void set_page(int page) { page_ = page; }
 
-  string* mutable_ch() { return &ch_; }
+  std::string* mutable_ch() { return &ch_; }
   Box* mutable_box()   { return box_; }
 
   // Sort function for sorting by left edge of box. Note that this will not
@@ -102,15 +102,15 @@ class BoxChar {
 
   // Create a tesseract box file from the vector of boxes. The image height
   // is needed to convert to tesseract coordinates.
-  static void WriteTesseractBoxFile(const string& name, int height,
+  static void WriteTesseractBoxFile(const std::string& name, int height,
                                     const std::vector<BoxChar*>& boxes);
   // Gets the tesseract box file as a string from the vector of boxes.
   // The image height is needed to convert to tesseract coordinates.
-  static string GetTesseractBoxStr(int height,
-                                   const std::vector<BoxChar*>& boxes);
+  static std::string GetTesseractBoxStr(int height,
+                                        const std::vector<BoxChar*>& boxes);
 
  private:
-  string ch_;
+  std::string ch_;
   Box* box_;
   int page_;
   // If the box is an RTL character, contains the original position in the

training/fileio.cpp

@@ -35,12 +35,12 @@ namespace tesseract {
 ///////////////////////////////////////////////////////////////////////////////
 // File::
 ///////////////////////////////////////////////////////////////////////////////
-FILE* File::Open(const string& filename, const string& mode) {
+FILE* File::Open(const std::string& filename, const std::string& mode) {
   return fopen(filename.c_str(), mode.c_str());
 }
 
-FILE* File::OpenOrDie(const string& filename,
-                      const string& mode) {
+FILE* File::OpenOrDie(const std::string& filename,
+                      const std::string& mode) {
   FILE* stream = fopen(filename.c_str(), mode.c_str());
   if (stream == nullptr) {
     tprintf("Unable to open '%s' in mode '%s'\n", filename.c_str(),
@@ -49,8 +49,8 @@ FILE* File::OpenOrDie(const string& filename,
   return stream;
 }
 
-void File::WriteStringToFileOrDie(const string& str,
-                                  const string& filename) {
+void File::WriteStringToFileOrDie(const std::string& str,
+                                  const std::string& filename) {
   FILE* stream = fopen(filename.c_str(), "wb");
   if (stream == nullptr) {
     tprintf("Unable to open '%s' for writing\n", filename.c_str());
@@ -60,7 +60,7 @@ void File::WriteStringToFileOrDie(const string& str,
   ASSERT_HOST(fclose(stream) == 0);
 }
 
-bool File::Readable(const string& filename) {
+bool File::Readable(const std::string& filename) {
   FILE* stream = fopen(filename.c_str(), "rb");
   if (stream == nullptr) {
     return false;
@@ -69,7 +69,7 @@ bool File::Readable(const string& filename) {
   return true;
 }
 
-bool File::ReadFileToString(const string& filename, string* out) {
+bool File::ReadFileToString(const std::string& filename, std::string* out) {
   FILE* stream = File::Open(filename.c_str(), "rb");
   if (stream == nullptr) return false;
   InputBuffer in(stream);
@@ -78,7 +78,7 @@ bool File::ReadFileToString(const string& filename, string* out) {
   return in.CloseFile();
 }
 
-string File::JoinPath(const string& prefix, const string& suffix) {
+std::string File::JoinPath(const std::string& prefix, const std::string& suffix) {
   return (prefix.empty() || prefix[prefix.size() - 1] == '/')
              ? prefix + suffix
              : prefix + "/" + suffix;
@@ -145,7 +145,7 @@ InputBuffer::~InputBuffer() {
   }
 }
 
-bool InputBuffer::Read(string* out) {
+bool InputBuffer::Read(std::string* out) {
   char buf[BUFSIZ + 1];
   int l;
   while ((l = fread(buf, 1, BUFSIZ, stream_)) > 0) {
@@ -183,7 +183,7 @@ OutputBuffer::~OutputBuffer() {
   }
 }
 
-void OutputBuffer::WriteString(const string& str) {
+void OutputBuffer::WriteString(const std::string& str) {
   fputs(str.c_str(), stream_);
 }
 

training/fileio.h

@@ -30,22 +30,22 @@ class File {
  public:
   // Try to open the file 'filename' in mode 'mode'.
   // Stop the program if it cannot open it.
-  static FILE* OpenOrDie(const string& filename, const string& mode);
-  static FILE* Open(const string& filename, const string& mode);
+  static FILE* OpenOrDie(const std::string& filename, const std::string& mode);
+  static FILE* Open(const std::string& filename, const std::string& mode);
 
   // Try to open the file 'filename' and to write 'str' in it.
   // Stop the program if it fails.
-  static void WriteStringToFileOrDie(const string& str, const string& filename);
+  static void WriteStringToFileOrDie(const std::string& str, const std::string& filename);
 
   // Return true if the file 'filename' is readable.
-  static bool Readable(const string& filename);
+  static bool Readable(const std::string& filename);
 
-  static bool ReadFileToString(const string& filename, string* out);
+  static bool ReadFileToString(const std::string& filename, std::string* out);
 
   // Helper methods
 
   // Concatenate file paths removing any extra intervening '/' symbols.
-  static string JoinPath(const string& prefix, const string& suffix);
+  static std::string JoinPath(const std::string& prefix, const std::string& suffix);
   // Delete a filename or all filenames matching a glob pattern.
   static bool Delete(const char* pathname);
   static bool DeleteMatchingFiles(const char* pattern);
@@ -63,7 +63,7 @@ class InputBuffer {
   // Read data until end-of-file.
   // The data is stored in '*out'.
   // Return false if an error occurs, true otherwise.
-  bool Read(string* out);
+  bool Read(std::string* out);
 
   // Close the FILE* used by InputBuffer.
   // Return false if an error occurs, true otherwise.
@@ -84,7 +84,7 @@ class OutputBuffer {
   ~OutputBuffer();
 
   // Write string 'str' to the open FILE*.
-  void WriteString(const string& str);
+  void WriteString(const std::string& str);
 
   // Close the FILE* used by InputBuffer.
   // Return false if an error occurs, true otherwise.