chore: import upstream snapshot with attribution

2026-07-13 13:25:10 +08:00
commit c397331b1e
3684 changed files with 990993 additions and 0 deletions
@@ -0,0 +1,95 @@
+#ifndef CPPJIEBA_QUERYSEGMENT_H
+#define CPPJIEBA_QUERYSEGMENT_H
+
+#include <algorithm>
+#include <set>
+#include <cassert>
+#include "limonp/Logging.hpp"
+#include "DictTrie.hpp"
+#include "SegmentBase.hpp"
+#include "FullSegment.hpp"
+#include "MixSegment.hpp"
+#include "Unicode.hpp"
+
+namespace cppjieba {
+class QuerySegment: public SegmentBase {
+ public:
+  QuerySegment(const string& dict, const string& model, const string& userDict = "")
+    : mixSeg_(dict, model, userDict),
+      trie_(mixSeg_.GetDictTrie()) {
+  }
+  QuerySegment(const DictTrie* dictTrie, const HMMModel* model)
+    : mixSeg_(dictTrie, model), trie_(dictTrie) {
+  }
+  QuerySegment() {
+    trie_ = NULL;
+  }
+  ~QuerySegment() {
+  }
+  void setRes(DictTrie *&dictTrie, HMMModel *&model) {
+    mixSeg_.setRes(dictTrie, model);
+    trie_ = dictTrie;
+  }
+  void Cut(const string& sentence, vector<string>& words) const {
+    Cut(sentence, words, true);
+  }
+  void Cut(const string& sentence, vector<string>& words, bool hmm) const {
+    vector<Word> tmp;
+    Cut(sentence, tmp, hmm);
+    GetStringsFromWords(tmp, words);
+  }
+  void Cut(const string& sentence, vector<Word>& words, bool hmm = true) const {
+    PreFilter pre_filter(symbols_, sentence);
+    PreFilter::Range range;
+    vector<WordRange> wrs;
+    wrs.reserve(sentence.size()/2);
+    while (pre_filter.HasNext()) {
+      range = pre_filter.Next();
+      Cut(range.begin, range.end, wrs, hmm);
+    }
+    words.clear();
+    words.reserve(wrs.size());
+    GetWordsFromWordRanges(sentence, wrs, words);
+  }
+  void Cut(RuneStrArray::const_iterator begin, RuneStrArray::const_iterator end, vector<WordRange>& res, bool hmm) const {
+    //use mix Cut first
+    vector<WordRange> mixRes;
+    mixSeg_.Cut(begin, end, mixRes, hmm);
+
+    vector<WordRange> fullRes;
+    for (vector<WordRange>::const_iterator mixResItr = mixRes.begin(); mixResItr != mixRes.end(); mixResItr++) {
+      if (mixResItr->Length() > 2) {
+        for (size_t i = 0; i + 1 < mixResItr->Length(); i++) {
+          WordRange wr(mixResItr->left + i, mixResItr->left + i + 1);
+          if (trie_->Find(wr.left, wr.right + 1) != NULL) {
+            res.push_back(wr);
+          }
+        }
+      }
+      if (mixResItr->Length() > 3) {
+        for (size_t i = 0; i + 2 < mixResItr->Length(); i++) {
+          WordRange wr(mixResItr->left + i, mixResItr->left + i + 2);
+          if (trie_->Find(wr.left, wr.right + 1) != NULL) {
+            res.push_back(wr);
+          }
+        }
+      }
+      res.push_back(*mixResItr);
+    }
+  }
+ private:
+  bool IsAllAscii(const Unicode& s) const {
+   for(size_t i = 0; i < s.size(); i++) {
+     if (s[i] >= 0x80) {
+       return false;
+     }
+   }
+   return true;
+  }
+  MixSegment mixSeg_;
+  const DictTrie* trie_;
+}; // QuerySegment
+
+} // namespace cppjieba
+
+#endif