chenx221/Textractor
1
0
Fork 0
forked from Public-Mirror/Textractor
Code Pull Requests Activity

optimize repetition removal at cost of some correctness (can be dealt with by extensions)

Browse Source
This commit is contained in:
Akash Mozumdar 2018-11-15 00:29:21 -05:00
parent 0492cb9549
commit 59869dc45a
1 changed files with 2 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
GUI/host/textthread.cc
View File

@ -50,14 +50,10 @@ void TextThread::Flush()
{
std::wstring sentence = buffer;
buffer.clear();
repeatingChars.clear();
locker.unlock(); // This algorithm might take a while
std::unordered_set<wchar_t> repeatingChars;
for (std::wsmatch results; std::regex_search(sentence, results, std::wregex(L"([^\\x00]{6,})\\1\\1")); sentence = results[1])
for (std::wsmatch results; std::regex_search(sentence, results, std::wregex(L"^([^]{6,})\\1\\1")); sentence = results[1])
repeatingChars = std::unordered_set(sentence.begin(), sentence.end());
locker.lock();
this->repeatingChars = repeatingChars;
locker.unlock();
AddSentence(sentence);