}

bool IsUTF8Whitespace(const char *text) {
  return SpanUTF8Whitespace(text) == strlen(text);
}

unsigned int SpanUTF8Whitespace(const char *text) {
  int n_white = 0;
  for (UNICHAR::const_iterator it = UNICHAR::begin(text, strlen(text));

            

Reported by FlawFinder.

              
unsigned int SpanUTF8Whitespace(const char *text) {
  int n_white = 0;
  for (UNICHAR::const_iterator it = UNICHAR::begin(text, strlen(text));
       it != UNICHAR::end(text, strlen(text)); ++it) {
    if (!IsWhitespace(*it)) {
      break;
    }
    n_white += it.utf8_len();

            

Reported by FlawFinder.

              unsigned int SpanUTF8Whitespace(const char *text) {
  int n_white = 0;
  for (UNICHAR::const_iterator it = UNICHAR::begin(text, strlen(text));
       it != UNICHAR::end(text, strlen(text)); ++it) {
    if (!IsWhitespace(*it)) {
      break;
    }
    n_white += it.utf8_len();
  }

            

Reported by FlawFinder.

              
unsigned int SpanUTF8NotWhitespace(const char *text) {
  int n_notwhite = 0;
  for (UNICHAR::const_iterator it = UNICHAR::begin(text, strlen(text));
       it != UNICHAR::end(text, strlen(text)); ++it) {
    if (IsWhitespace(*it)) {
      break;
    }
    n_notwhite += it.utf8_len();

            

Reported by FlawFinder.

              unsigned int SpanUTF8NotWhitespace(const char *text) {
  int n_notwhite = 0;
  for (UNICHAR::const_iterator it = UNICHAR::begin(text, strlen(text));
       it != UNICHAR::end(text, strlen(text)); ++it) {
    if (IsWhitespace(*it)) {
      break;
    }
    n_notwhite += it.utf8_len();
  }

            

Reported by FlawFinder.

                // operator=() for each element like double_the_size() does.
  static T *double_the_size_memcpy(int current_size, T *data) {
    T *data_new = new T[current_size * 2];
    memcpy(data_new, data, sizeof(T) * current_size);
    delete[] data;
    return data_new;
  }

  // Reverses the elements of the vector.

            

Reported by FlawFinder.

              // returning false on error.
inline bool LoadDataFromFile(const char *filename, GenericVector<char> *data) {
  bool result = false;
  FILE *fp = fopen(filename, "rb");
  if (fp != nullptr) {
    fseek(fp, 0, SEEK_END);
    auto size = std::ftell(fp);
    fseek(fp, 0, SEEK_SET);
    // Trying to open a directory on Linux sets size to LONG_MAX. Catch it here.

            

Reported by FlawFinder.

              // The default FileWriter writes the vector of char to the filename file,
// returning false on error.
inline bool SaveDataToFile(const GenericVector<char> &data, const char *filename) {
  FILE *fp = fopen(filename, "wb");
  if (fp == nullptr) {
    return false;
  }
  bool result = static_cast<int>(fwrite(&data[0], 1, data.size(), fp)) == data.size();
  fclose(fp);

            

Reported by FlawFinder.

                // Returns false on error or if the callback returns false.
  // DEPRECATED. Use [De]Serialize[Classes] instead.
  bool write(FILE *f, std::function<bool(FILE *, const T &)> cb) const;
  bool read(TFile *f, std::function<bool(TFile *, T *)> cb);
  // Writes a vector of simple types to the given file. Assumes that bitwise
  // read/write of T will work. Returns false in case of error.
  // TODO(rays) Change all callers to use TFile and remove deprecated methods.
  bool Serialize(FILE *fp) const;
  bool Serialize(TFile *fp) const;

            

Reported by FlawFinder.

              }

template <typename T>
bool GenericVector<T>::read(TFile *f, std::function<bool(TFile *, T *)> cb) {
  int32_t reserved;
  if (f->FReadEndian(&reserved, sizeof(reserved), 1) != 1) {
    return false;
  }
  reserve(reserved);

            

Reported by FlawFinder.

              FILE *OpenBoxFile(const char *fname) {
  std::string filename = BoxFileName(fname);
  FILE *box_file = nullptr;
  if (!(box_file = fopen(filename.c_str(), "rb"))) {
    CANTOPENFILE.error("read_next_box", TESSEXIT, "Can't open box file %s", filename.c_str());
  }
  return box_file;
}


            

Reported by FlawFinder.

              bool ReadNextBox(int target_page, int *line_number, FILE *box_file, std::string &utf8_str,
                 TBOX *bounding_box) {
  int page = 0;
  char buff[kBoxReadBufSize]; // boxfile read buffer
  char *buffptr = buff;

  while (fgets(buff, sizeof(buff) - 1, box_file)) {
    (*line_number)++;


            

Reported by FlawFinder.

                                   TBOX *bounding_box) {
  *bounding_box = TBOX(); // Initialize it to empty.
  utf8_str = "";
  char uch[kBoxReadBufSize];
  const char *buffptr = boxfile_str;
  // Read the unichar without messing up on Tibetan.
  // According to issue 253 the utf-8 surrogates 85 and A0 are treated
  // as whitespace by sscanf, so it is more reliable to just find
  // ascii space and tab.

            

Reported by FlawFinder.

                }
  // Test for long space-delimited string label.
  if (strcmp(uch, kMultiBlobLabelCode) == 0 && (buffptr = strchr(buffptr, '#')) != nullptr) {
    strncpy(uch, buffptr + 1, kBoxReadBufSize - 1);
    uch[kBoxReadBufSize - 1] = '\0'; // Prevent buffer overrun.
    chomp_string(uch);
    uch_len = strlen(uch);
  }
  // Validate UTF8 by making unichars with it.

            

Reported by FlawFinder.

                  strncpy(uch, buffptr + 1, kBoxReadBufSize - 1);
    uch[kBoxReadBufSize - 1] = '\0'; // Prevent buffer overrun.
    chomp_string(uch);
    uch_len = strlen(uch);
  }
  // Validate UTF8 by making unichars with it.
  int used = 0;
  while (used < uch_len) {
    tesseract::UNICHAR ch(uch + used, uch_len - used);

            

Reported by FlawFinder.

                    break; // Illegal surrogate
    }
  }
  memcpy(chars, utf8_str, total_len);
  if (total_len < UNICHAR_LEN) {
    chars[UNICHAR_LEN - 1] = total_len;
    while (total_len < UNICHAR_LEN - 1) {
      chars[total_len++] = 0;
    }

            

Reported by FlawFinder.

              char *UNICHAR::utf8_str() const {
  int len = utf8_len();
  char *str = new char[len + 1];
  memcpy(str, chars, len);
  str[len] = 0;
  return str;
}

// Get the number of bytes in the first character of the given utf8 string.

            

Reported by FlawFinder.

              
// Get the number of bytes in the first character of the given utf8 string.
int UNICHAR::utf8_step(const char *utf8_str) {
  static const char utf8_bytes[256] = {
      1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
      1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
      1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
      1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
      1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,

            

Reported by FlawFinder.

                  utf8_output[0] = ' ';
    return 1;
  }
  strncpy(utf8_output, it_, len);
  return len;
}

int UNICHAR::const_iterator::utf8_len() const {
  ASSERT_HOST(it_ != nullptr);

            

Reported by FlawFinder.

              // Returns an empty vector if the input contains invalid UTF-8.
/* static */
std::vector<char32> UNICHAR::UTF8ToUTF32(const char *utf8_str) {
  const int utf8_length = strlen(utf8_str);
  std::vector<char32> unicodes;
  unicodes.reserve(utf8_length);
  const_iterator end_it(end(utf8_str, utf8_length));
  for (const_iterator it(begin(utf8_str, utf8_length)); it != end_it; ++it) {
    if (it.is_legal()) {

            

Reported by FlawFinder.

                // language-specific variables from [lang].traineddata file, so that custom
  // config files can override values in [lang].traineddata file.
  for (int i = 0; i < configs_size; ++i) {
    read_config_file(configs[i], set_params_constraint);
  }

  // Set params specified in vars_vec (done after setting params from config
  // files, so that params in vars_vec can override those from files).
  if (vars_vec != nullptr && vars_values != nullptr) {

            

Reported by Cppcheck.

                path += "configs/";
  path += filename;
  FILE *fp;
  if ((fp = fopen(path.c_str(), "rb")) != nullptr) {
    fclose(fp);
  } else {
    path = datadir;
    path += "tessconfigs/";
    path += filename;

            

Reported by FlawFinder.

                  path = datadir;
    path += "tessconfigs/";
    path += filename;
    if ((fp = fopen(path.c_str(), "rb")) != nullptr) {
      fclose(fp);
    } else {
      path = filename;
    }
  }

            

Reported by FlawFinder.

                }

  if (!tessedit_write_params_to_file.empty()) {
    FILE *params_file = fopen(tessedit_write_params_to_file.c_str(), "wb");
    if (params_file != nullptr) {
      ParamUtils::PrintParams(params_file, this->params());
      fclose(params_file);
    } else {
      tprintf("Failed to open %s for writing params.\n", tessedit_write_params_to_file.c_str());

            

Reported by FlawFinder.

                    ++start;
    }
    // Find the index of the end of the lang code in string start.
    int end = strlen(start);
    const char *plus = strchr(start, '+');
    if (plus != nullptr && plus - start < end) {
      end = plus - start;
    }
    std::string lang_code(start);

            

Reported by FlawFinder.

                  if (ev_type == SVET_POPUP) {
      if (ev->command_id == IDA_SHAPE_INDEX) {
        if (shape_table_ != nullptr) {
          *shape_id = atoi(ev->parameter);
          *adaptive_on = false;
          *pretrained_on = true;
          if (*shape_id >= 0 && *shape_id < shape_table_->NumShapes()) {
            int font_id;
            shape_table_->GetFirstUnicharAndFont(*shape_id, &unichar_id, &font_id);

            

Reported by FlawFinder.

                }
  if (version_id >= 4) {
    using namespace std::placeholders; // for _1, _2
    this->fontinfo_table_.read(fp, std::bind(read_info, _1, _2));
    if (version_id >= 5) {
      this->fontinfo_table_.read(fp, std::bind(read_spacing_info, _1, _2));
    }
    this->fontset_table_.read(fp, [](auto *f, auto *fs) { return f->DeSerialize(*fs); } );
  }

            

Reported by FlawFinder.

                  using namespace std::placeholders; // for _1, _2
    this->fontinfo_table_.read(fp, std::bind(read_info, _1, _2));
    if (version_id >= 5) {
      this->fontinfo_table_.read(fp, std::bind(read_spacing_info, _1, _2));
    }
    this->fontset_table_.read(fp, [](auto *f, auto *fs) { return f->DeSerialize(*fs); } );
  }

  return (Templates);

            

Reported by FlawFinder.

                  if (version_id >= 5) {
      this->fontinfo_table_.read(fp, std::bind(read_spacing_info, _1, _2));
    }
    this->fontset_table_.read(fp, [](auto *f, auto *fs) { return f->DeSerialize(*fs); } );
  }

  return (Templates);
} /* ReadIntTemplates */


            

Reported by FlawFinder.

              
static void PrintCommandLineFlags() {
  const char *kFlagNamePrefix = "FLAGS_";
  const int kFlagNamePrefixLen = strlen(kFlagNamePrefix);
  for (auto &param : GlobalParams()->int_params) {
    if (!strncmp(param->name_str(), kFlagNamePrefix, kFlagNamePrefixLen)) {
      printf("  --%s  %s  (type:int default:%d)\n",
             param->name_str() + kFlagNamePrefixLen,
             param->info_str(), int32_t(*param));

            

Reported by FlawFinder.

                  int32_t int_val;
    if (IntFlagExists(lhs.c_str(), &int_val)) {
      if (rhs != nullptr) {
        if (!strlen(rhs)) {
          // Bad input of the format --int_flag=
          tprintf("ERROR: Bad argument: %s\n", (*argv)[i]);
          exit(1);
        }
        if (!SafeAtoi(rhs, &int_val)) {

            

Reported by FlawFinder.

                  double double_val;
    if (DoubleFlagExists(lhs.c_str(), &double_val)) {
      if (rhs != nullptr) {
        if (!strlen(rhs)) {
          // Bad input of the format --double_flag=
          tprintf("ERROR: Bad argument: %s\n", (*argv)[i]);
          exit(1);
        }
        if (!SafeAtod(rhs, &double_val)) {

            

Reported by FlawFinder.

                      // --flag form
        bool_val = true;
      } else {
        if (!strlen(rhs)) {
          // Bad input of the format --bool_flag=
          tprintf("ERROR: Bad argument: %s\n", (*argv)[i]);
          exit(1);
        }
        if (!strcmp(rhs, "false") || !strcmp(rhs, "0")) {

            

Reported by FlawFinder.

              UNICHAR_ID WERD_RES::BothQuotes(UNICHAR_ID id1, UNICHAR_ID id2) {
  const char *ch = uch_set->id_to_unichar(id1);
  const char *next_ch = uch_set->id_to_unichar(id2);
  if (is_simple_quote(ch, strlen(ch)) && is_simple_quote(next_ch, strlen(next_ch))) {
    return uch_set->unichar_to_id("\"");
  }
  return INVALID_UNICHAR_ID;
}


            

Reported by FlawFinder.

              UNICHAR_ID WERD_RES::BothQuotes(UNICHAR_ID id1, UNICHAR_ID id2) {
  const char *ch = uch_set->id_to_unichar(id1);
  const char *next_ch = uch_set->id_to_unichar(id2);
  if (is_simple_quote(ch, strlen(ch)) && is_simple_quote(next_ch, strlen(next_ch))) {
    return uch_set->unichar_to_id("\"");
  }
  return INVALID_UNICHAR_ID;
}


            

Reported by FlawFinder.

              UNICHAR_ID WERD_RES::BothHyphens(UNICHAR_ID id1, UNICHAR_ID id2) {
  const char *ch = uch_set->id_to_unichar(id1);
  const char *next_ch = uch_set->id_to_unichar(id2);
  if (strlen(ch) == 1 && strlen(next_ch) == 1 && (*ch == '-' || *ch == '~') &&
      (*next_ch == '-' || *next_ch == '~')) {
    return uch_set->unichar_to_id("-");
  }
  return INVALID_UNICHAR_ID;
}

            

Reported by FlawFinder.

              UNICHAR_ID WERD_RES::BothHyphens(UNICHAR_ID id1, UNICHAR_ID id2) {
  const char *ch = uch_set->id_to_unichar(id1);
  const char *next_ch = uch_set->id_to_unichar(id2);
  if (strlen(ch) == 1 && strlen(next_ch) == 1 && (*ch == '-' || *ch == '~') &&
      (*next_ch == '-' || *next_ch == '~')) {
    return uch_set->unichar_to_id("-");
  }
  return INVALID_UNICHAR_ID;
}

            

Reported by FlawFinder.

              // returning false on error.
bool LoadDataFromFile(const char *filename, std::vector<char> *data) {
  bool result = false;
  FILE *fp = fopen(filename, "rb");
  if (fp != nullptr) {
    fseek(fp, 0, SEEK_END);
    auto size = std::ftell(fp);
    fseek(fp, 0, SEEK_SET);
    // Trying to open a directory on Linux sets size to LONG_MAX. Catch it here.

            

Reported by FlawFinder.

              // The default FileWriter writes the vector of char to the filename file,
// returning false on error.
bool SaveDataToFile(const std::vector<char> &data, const char *filename) {
  FILE *fp = fopen(filename, "wb");
  if (fp == nullptr) {
    return false;
  }
  bool result = static_cast<int>(fwrite(&data[0], 1, data.size(), fp)) == data.size();
  fclose(fp);

            

Reported by FlawFinder.

                is_writing_ = false;
  swap_ = false;
  data_->resize(size); // TODO: optimize no init
  memcpy(&(*data_)[0], data, size);
  return true;
}

bool TFile::Open(FILE *fp, int64_t end_offset) {
  offset_ = 0;

            

Reported by FlawFinder.

                  }
  }
  if (required_size > 0 && buffer != nullptr) {
    memcpy(buffer, &(*data_)[offset_], required_size);
  }
  offset_ += required_size;
  return required_size / size;
}


            

Reported by FlawFinder.

                int thresh = (topline_onpixel_count * 70) / 100;
  int ulimit = topline_ylevel;
  int llimit = topline_ylevel;
  while (ulimit > 0 && hist_horiz.hist()[ulimit] >= thresh) {
    --ulimit;
  }
  while (llimit < pixGetHeight(word_pix) && hist_horiz.hist()[llimit] >= thresh) {
    ++llimit;
  }

            

Reported by FlawFinder.

                int thresh = (topline_onpixel_count * 70) / 100;
  int ulimit = topline_ylevel;
  int llimit = topline_ylevel;
  while (ulimit > 0 && hist_horiz.hist()[ulimit] >= thresh) {
    --ulimit;
  }
  while (llimit < pixGetHeight(word_pix) && hist_horiz.hist()[llimit] >= thresh) {
    ++llimit;
  }

            

Reported by FlawFinder.

                int ulimit = topline_ylevel;
  int llimit = topline_ylevel;
  while (ulimit > 0 && hist_horiz.hist()[ulimit] >= thresh) {
    --ulimit;
  }
  while (llimit < pixGetHeight(word_pix) && hist_horiz.hist()[llimit] >= thresh) {
    ++llimit;
  }


            

Reported by FlawFinder.

                }

  if (shirorekha_top) {
    *shirorekha_top = ulimit;
  }
  if (shirorekha_bottom) {
    *shirorekha_bottom = llimit;
  }
  if (shirorekha_ylevel) {

            

Reported by FlawFinder.