本文整理汇总了C++中UNICHARSET::encodable_string方法的典型用法代码示例。如果您正苦于以下问题：C++ UNICHARSET::encodable_string方法的具体用法？C++ UNICHARSET::encodable_string怎么用？C++ UNICHARSET::encodable_string使用的例子？那么, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类UNICHARSET的用法示例。
在下文中一共展示了UNICHARSET::encodable_string方法的1个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的C++代码示例。
示例1: Main
//.........这里部分代码省略.........
const unsigned int kCharsPerLine = (FLAGS_ptsize > 20) ? 50 : 100;
std::string rand_utf8;
UNICHARSET unicharset;
if (FLAGS_render_ngrams && !FLAGS_unicharset_file.empty() &&
!unicharset.load_from_file(FLAGS_unicharset_file.c_str())) {
tprintf("Failed to load unicharset from file %s\n",
FLAGS_unicharset_file.c_str());
exit(1);
}
// If we are rendering ngrams that will be OCRed later, shuffle them so that
// tesseract does not have difficulties finding correct baseline, word
// spaces, etc.
const char *str8 = src_utf8.c_str();
int len = src_utf8.length();
int step;
std::vector<std::pair<int, int> > offsets;
int offset = SpanUTF8Whitespace(str8);
while (offset < len) {
step = SpanUTF8NotWhitespace(str8 + offset);
offsets.push_back(std::make_pair(offset, step));
offset += step;
offset += SpanUTF8Whitespace(str8 + offset);
}
if (FLAGS_render_ngrams)
std::random_shuffle(offsets.begin(), offsets.end());
for (size_t i = 0, line = 1; i < offsets.size(); ++i) {
const char *curr_pos = str8 + offsets[i].first;
int ngram_len = offsets[i].second;
// Skip words that contain characters not in found in unicharset.
std::string cleaned = UNICHARSET::CleanupString(curr_pos, ngram_len);
if (!FLAGS_unicharset_file.empty() &&
!unicharset.encodable_string(cleaned.c_str(), nullptr)) {
continue;
}
rand_utf8.append(curr_pos, ngram_len);
if (rand_utf8.length() > line * kCharsPerLine) {
rand_utf8.append(" \n");
++line;
if (line & 0x1) rand_utf8.append(kSeparator);
} else {
rand_utf8.append(kSeparator);
}
}
tlog(1, "Rendered ngram string of size %d\n", rand_utf8.length());
src_utf8.swap(rand_utf8);
}
if (FLAGS_only_extract_font_properties) {
tprintf("Extracting font properties only\n");
ExtractFontProperties(src_utf8, &render, FLAGS_outputbase.c_str());
tprintf("Done!\n");
return 0;
}
int im = 0;
std::vector<float> page_rotation;
const char* to_render_utf8 = src_utf8.c_str();
tesseract::TRand randomizer;
randomizer.set_seed(kRandomSeed);
std::vector<std::string> font_names;
// We use a two pass mechanism to rotate images in both direction.
// The first pass(0) will rotate the images in random directions and
// the second pass(1) will mirror those rotations.
int num_pass = FLAGS_bidirectional_rotation ? 2 : 1;