fix decoders: force indices in FST starting from one & add version check in setup

.clang_format.hook

 #!/usr/bin/env bash
 set -e
 
-readonly VERSION="3.8"
+readonly VERSION="3.6"
 
 version=$(clang-format -version)
 

decoders/swig/path_trie.cpp

@@ -52,7 +52,7 @@ PathTrie* PathTrie::get_path_trie(int new_char, bool reset) {
   } else {
     if (has_dictionary_) {
       matcher_->SetState(dictionary_state_);
-      bool found = matcher_->Find(new_char);
+      bool found = matcher_->Find(new_char + 1);
       if (!found) {
         // Adding this character causes word outside dictionary
         auto FSTZERO = fst::TropicalWeight::Zero();

decoders/swig/scorer.cpp

@@ -152,10 +152,8 @@ void Scorer::set_char_map(const std::vector<std::string>& char_list) {
   for (size_t i = 0; i < char_list_.size(); i++) {
     if (char_list_[i] == " ") {
       SPACE_ID_ = i;
-      char_map_[' '] = i;
-    } else if (char_list_[i].size() == 1) {
-      char_map_[char_list_[i][0]] = i;
     }
+    char_map_[char_list_[i]] = i + 1;  // Force index starting from zero
   }
 }
 
@@ -193,17 +191,11 @@ std::vector<std::string> Scorer::make_ngram(PathTrie* prefix) {
 
 void Scorer::fill_dictionary(bool add_space) {
   fst::StdVectorFst dictionary;
-  // First reverse char_list so ints can be accessed by chars
-  std::unordered_map<std::string, int> char_map;
-  for (size_t i = 0; i < char_list_.size(); i++) {
-    char_map[char_list_[i]] = i;
-  }
-
   // For each unigram convert to ints and put in trie
   int dict_size = 0;
   for (const auto& word : vocabulary_) {
     bool added = add_word_to_dictionary(
-        word, char_map, add_space, SPACE_ID_, &dictionary);
+        word, char_map_, add_space, SPACE_ID_ + 1, &dictionary);
     dict_size += added ? 1 : 0;
   }
 

decoders/swig/scorer.h

@@ -104,7 +104,7 @@ private:
 
   int SPACE_ID_;
   std::vector<std::string> char_list_;
-  std::unordered_map<char, int> char_map_;
+  std::unordered_map<std::string, int> char_map_;
 
   std::vector<std::string> vocabulary_;
 };

decoders/swig/setup.py

@@ -113,7 +113,7 @@ decoders_module = [
 
 setup(
     name='swig_decoders',
-    version='1.0',
+    version='1.1',
     description="""CTC decoders""",
     ext_modules=decoders_module,
     py_modules=['swig_decoders'], )

setup.sh

@@ -27,7 +27,7 @@ if [ $? != 0 ]; then
 fi
 
 # install decoders
-python -c "import swig_decoders"
+python -c "import pkg_resources; pkg_resources.require(\"swig_decoders==1.1\")"
 if [ $? != 0 ]; then
     cd decoders/swig > /dev/null
     sh setup.sh