Test int8 models (#107)

* Test int8 models

* Fix displaying help messages

* small fixes

* Fix jni test

sherpa-onnx/csrc/text-utils.cc

@@ -7,7 +7,11 @@
 
 #include <assert.h>
 
+#include <algorithm>
+#include <limits>
+#include <sstream>
 #include <string>
+#include <unordered_map>
 #include <vector>
 
 // This file is copied/modified from
@@ -15,6 +19,102 @@
 
 namespace sherpa_onnx {
 
+// copied from kaldi/src/util/text-util.cc
+template <class T>
+class NumberIstream {
+ public:
+  explicit NumberIstream(std::istream &i) : in_(i) {}
+
+  NumberIstream &operator>>(T &x) {
+    if (!in_.good()) return *this;
+    in_ >> x;
+    if (!in_.fail() && RemainderIsOnlySpaces()) return *this;
+    return ParseOnFail(&x);
+  }
+
+ private:
+  std::istream &in_;
+
+  bool RemainderIsOnlySpaces() {
+    if (in_.tellg() != std::istream::pos_type(-1)) {
+      std::string rem;
+      in_ >> rem;
+
+      if (rem.find_first_not_of(' ') != std::string::npos) {
+        // there is not only spaces
+        return false;
+      }
+    }
+
+    in_.clear();
+    return true;
+  }
+
+  NumberIstream &ParseOnFail(T *x) {
+    std::string str;
+    in_.clear();
+    in_.seekg(0);
+    // If the stream is broken even before trying
+    // to read from it or if there are many tokens,
+    // it's pointless to try.
+    if (!(in_ >> str) || !RemainderIsOnlySpaces()) {
+      in_.setstate(std::ios_base::failbit);
+      return *this;
+    }
+
+    std::unordered_map<std::string, T> inf_nan_map;
+    // we'll keep just uppercase values.
+    inf_nan_map["INF"] = std::numeric_limits<T>::infinity();
+    inf_nan_map["+INF"] = std::numeric_limits<T>::infinity();
+    inf_nan_map["-INF"] = -std::numeric_limits<T>::infinity();
+    inf_nan_map["INFINITY"] = std::numeric_limits<T>::infinity();
+    inf_nan_map["+INFINITY"] = std::numeric_limits<T>::infinity();
+    inf_nan_map["-INFINITY"] = -std::numeric_limits<T>::infinity();
+    inf_nan_map["NAN"] = std::numeric_limits<T>::quiet_NaN();
+    inf_nan_map["+NAN"] = std::numeric_limits<T>::quiet_NaN();
+    inf_nan_map["-NAN"] = -std::numeric_limits<T>::quiet_NaN();
+    // MSVC
+    inf_nan_map["1.#INF"] = std::numeric_limits<T>::infinity();
+    inf_nan_map["-1.#INF"] = -std::numeric_limits<T>::infinity();
+    inf_nan_map["1.#QNAN"] = std::numeric_limits<T>::quiet_NaN();
+    inf_nan_map["-1.#QNAN"] = -std::numeric_limits<T>::quiet_NaN();
+
+    std::transform(str.begin(), str.end(), str.begin(), ::toupper);
+
+    if (inf_nan_map.find(str) != inf_nan_map.end()) {
+      *x = inf_nan_map[str];
+    } else {
+      in_.setstate(std::ios_base::failbit);
+    }
+
+    return *this;
+  }
+};
+
+/// ConvertStringToReal converts a string into either float or double
+/// and returns false if there was any kind of problem (i.e. the string
+/// was not a floating point number or contained extra non-whitespace junk).
+/// Be careful- this function will successfully read inf's or nan's.
+template <typename T>
+bool ConvertStringToReal(const std::string &str, T *out) {
+  std::istringstream iss(str);
+
+  NumberIstream<T> i(iss);
+
+  i >> *out;
+
+  if (iss.fail()) {
+    // Number conversion failed.
+    return false;
+  }
+
+  return true;
+}
+
+template bool ConvertStringToReal<float>(const std::string &str, float *out);
+
+template bool ConvertStringToReal<double>(const std::string &str, double *out);
+
 void SplitStringToVector(const std::string &full, const char *delim,
                          bool omit_empty_strings,
                          std::vector<std::string> *out) {
@@ -43,7 +143,9 @@ bool SplitStringToFloats(const std::string &full, const char *delim,
   out->resize(split.size());
   for (size_t i = 0; i < split.size(); ++i) {
     // assume atof never fails
-    (*out)[i] = atof(split[i].c_str());
+    F f = 0;
+    if (!ConvertStringToReal(split[i], &f)) return false;
+    (*out)[i] = f;
   }
   return true;
 }