by default use codepage1252 for metadata including non-ascii characters while missing codepage info

* IMCtermite.pyx: add codepage conversion for windows
* bump VERSION

README.md

@@ -233,3 +233,6 @@ can be found in the `python/examples` folder.
 - https://docs.github.com/en/actions/reference/workflow-syntax-for-github-actions#jobsjob_idstepsrun
 - https://github.com/pypa/cibuildwheel/blob/main/examples/github-deploy.yml
 - https://cibuildwheel.readthedocs.io/en/stable/deliver-to-pypi/
+- https://www.gnu.org/software/libiconv/
+- https://vcpkg.io/en/packages.html
+- https://vcpkg.io/en/getting-started

lib/imc_channel.hpp

@@ -9,6 +9,12 @@
 #include <math.h>
 #include <chrono>
 #include <ctime>
+#include <time.h>
+#if defined(__linux__) || defined(__APPLE__)
+#include <iconv.h>
+#elif defined(__WIN32__) || defined(_WIN32)
+#define timegm _mkgmtime
+#endif
 
 //---------------------------------------------------------------------------//
 
@@ -142,6 +148,87 @@ namespace imc
     return sumstr;
   }
 
+  #if defined(__linux__) || defined(__APPLE__)
+  // convert encoding of any descriptions, channel-names, units etc.
+  class iconverter
+  {
+    std::string in_enc_, out_enc_;
+    iconv_t cd_;
+    size_t out_buffer_size_;
+
+    public:
+
+      iconverter(std::string in_enc, std::string out_enc, size_t out_buffer_size = 1024) :
+        in_enc_(in_enc), out_enc_(out_enc), out_buffer_size_(out_buffer_size)
+      {
+        // allocate descriptor for character set conversion
+        // (https://man7.org/linux/man-pages/man3/iconv_open.3.html)
+        cd_ = iconv_open(out_enc.c_str(), in_enc.c_str());
+
+        if ( (iconv_t)-1 == cd_ )
+        {
+          if ( errno == EINVAL )
+          {
+            std::string errmsg = std::string("The encoding conversion from ") + in_enc
+              + std::string(" to ") + out_enc + std::string(" is not supported by the implementation.");
+            throw std::runtime_error(errmsg);
+          }
+        }
+      }
+
+      void convert(std::string &astring)
+      {
+        if ( astring.empty() ) return;
+
+        std::vector<char> in_buffer(astring.begin(),astring.end());
+        char *inbuf = &in_buffer[0];
+        size_t inbytes = in_buffer.size();
+
+        std::vector<char> out_buffer(out_buffer_size_);
+        char *outbuf = &out_buffer[0];
+        size_t outbytes = out_buffer.size();
+
+        // perform character set conversion
+        // ( - https://man7.org/linux/man-pages/man3/iconv.3.html
+        //   - https://www.ibm.com/docs/en/zos/2.2.0?topic=functions-iconv-code-conversion )
+        while ( inbytes > 0 )
+        {
+          size_t res = iconv(cd_,&inbuf,&inbytes,&outbuf,&outbytes);
+
+          if ( (size_t)-1 == res )
+          {
+            std::string errmsg;
+            if ( errno == EILSEQ )
+            {
+              errmsg = std::string("An invalid multibyte sequence is encountered in the input.");
+              throw std::runtime_error(errmsg);
+            }
+            else if ( errno == EINVAL )
+            {
+              errmsg = std::string("An incomplete multibyte sequence is encountered in the input")
+                     + std::string(" and the input byte sequence terminates after it.");
+            }
+            else if ( errno == E2BIG )
+            {
+              errmsg = std::string("The output buffer has no more room for the next converted character.");
+            }
+            throw std::runtime_error(errmsg);
+          }
+        }
+
+        std::string outstring(out_buffer.begin(),out_buffer.end()-outbytes);
+        astring = outstring;
+      }
+  };
+  #elif defined(__WIN32__) || defined(_WIN32)
+  class iconverter
+  {
+    public:
+      iconverter(std::string in_enc, std::string out_enc, size_t out_buffer_size = 1024) {}
+      void convert(std::string &astring) {}
+  };
+  #endif
+
   // channel
   struct channel
   {
@@ -301,9 +388,12 @@ namespace imc
         double secs_int;
         trigger_time_frac_secs_ = modf((double)secs,&secs_int);
         tms.tm_sec = (int)secs_int;
+        //tms.tm_isdst = -1;
 
         // generate std::chrono::system_clock::time_point type
-        std::time_t ts = std::mktime(&tms);
+        // ( - https://www.gnu.org/software/libc/manual/html_node/Broken_002ddown-Time.html
+        //   - https://man7.org/linux/man-pages/man3/tzset.3.html )
+        std::time_t ts = timegm(&tms); //std::mktime(&tms);
         trigger_time_ = std::chrono::system_clock::from_time_t(ts);
       }
 
@@ -313,6 +403,9 @@ namespace imc
       // calculate absolute trigger-time
       absolute_trigger_time_ = trigger_time_ + std::chrono::seconds(addtime_);
       //                                       + std::chrono::nanoseconds((long int)(trigger_time_frac_secs_*1.e9));
+
+      // convert any non-UTF-8 codepage to UTF-8
+      convert_encoding();
     }
 
     // convert buffer to actual datatype
@@ -400,6 +493,40 @@ namespace imc
       }
     }
 
+    // convert any description, units etc. to UTF-8 (by default)
+    void convert_encoding()
+    {
+      // actual input codepage
+      std::string cpn;
+
+      if ( !codepage_.empty() )
+      {
+        // construct iconv-compatible name for respective codepage
+        cpn = std::string("CP") + codepage_;
+      }
+      else {
+	// assume codepage 1252 by default
+	cpn = std::string("CP1252");
+      }
+
+      // set up converter
+      std::string utf = std::string("UTF-8");
+      iconverter conv(cpn,utf);
+
+      conv.convert(name_);
+      conv.convert(comment_);
+      conv.convert(origin_);
+      conv.convert(origin_comment_);
+      conv.convert(text_);
+      conv.convert(language_code_);
+      conv.convert(yname_);
+      conv.convert(yunit_);
+      conv.convert(xname_);
+      conv.convert(xunit_);
+      conv.convert(group_name_);
+      conv.convert(group_comment_);
+    }
+
     // get info string
     std::string get_info(int width = 20)
     {
@@ -413,8 +540,8 @@ namespace imc
         <<std::setw(width)<<std::left<<"comment:"<<comment_<<"\n"
         <<std::setw(width)<<std::left<<"origin:"<<origin_<<"\n"
         <<std::setw(width)<<std::left<<"description:"<<text_<<"\n"
-        <<std::setw(width)<<std::left<<"trigger-time-nt:"<<std::put_time(std::localtime(&tt),"%FT%T")<<"\n"
-        <<std::setw(width)<<std::left<<"trigger-time:"<<std::put_time(std::localtime(&att),"%FT%T")<<"\n"
+        <<std::setw(width)<<std::left<<"trigger-time-nt:"<<std::put_time(std::gmtime(&tt),"%FT%T")<<"\n"
+        <<std::setw(width)<<std::left<<"trigger-time:"<<std::put_time(std::gmtime(&att),"%FT%T")<<"\n"
         <<std::setw(width)<<std::left<<"language-code:"<<language_code_<<"\n"
         <<std::setw(width)<<std::left<<"codepage:"<<codepage_<<"\n"
         <<std::setw(width)<<std::left<<"yname:"<<yname_<<"\n"
@@ -451,8 +578,8 @@ namespace imc
              <<"\",\"comment\":\""<<comment_
              <<"\",\"origin\":\""<<origin_
              <<"\",\"description\":\""<<text_
-             <<"\",\"trigger-time-nt\":\""<<std::put_time(std::localtime(&tt),"%FT%T")
-             <<"\",\"trigger-time\":\""<<std::put_time(std::localtime(&att),"%FT%T")
+             <<"\",\"trigger-time-nt\":\""<<std::put_time(std::gmtime(&tt),"%FT%T")
+             <<"\",\"trigger-time\":\""<<std::put_time(std::gmtime(&att),"%FT%T")
              <<"\",\"language-code\":\""<<language_code_
              <<"\",\"codepage\":\""<<codepage_
              <<"\",\"yname\":\""<<prepjsonstr(yname_)

makefile

@@ -10,8 +10,11 @@ SRC = src/
 LIB = lib/
 PYT = python/
 
-# list headers
+# list headers and include directories
 HPP = $(wildcard $(LIB)/*.hpp)
+IPP = $(shell find $(LIB) -type f -name '*.hpp')
+KIB = $(shell find $(LIB) -type d)
+MIB = $(foreach dir,$(KIB),-I $(dir))
 
 # choose compiler and its options
 CC = g++ -std=c++17
@@ -36,12 +39,12 @@ $(EXE): check-tags $(GVSN) main.o
 	$(CC) $(OPT) main.o -o $@
 
 # build main.cpp and include git version/commit tag
-main.o: src/main.cpp $(HPP)
+main.o: src/main.cpp $(IPP)
 	@cp $< $<.cpp
 	@sed -i 's/TAGSTRING/$(GTAG)/g' $<.cpp
 	@sed -i 's/HASHSTRING/$(GHSH)/g' $<.cpp
 	@sed -i 's/TIMESTAMPSTRING/$(TMS)/g' $<.cpp
-	$(CC) -c $(OPT) -I $(LIB) $<.cpp -o $@
+	$(CC) -c $(OPT) $(MIB) $<.cpp -o $@
 	@rm $<.cpp
 
 install: $(EXE)

python/IMCtermite.pyx

@@ -5,6 +5,16 @@ from IMCtermite cimport cppimctermite
 
 import json as jn
 import decimal
+import platform
+
+# auxiliary function for codepage conversion
+def get_codepage(chn) :
+    if platform == 'Windows' :
+        chndec = jn.loads(chn.decode(errors="ignore"))
+        chncdp = chndec["codepage"]
+        return 'utf-8' if chncdp is None else chncdp
+    else :
+        return 'utf-8'
 
 cdef class imctermite:
 
@@ -22,7 +32,7 @@ cdef class imctermite:
   # get JSON list of channels
   def get_channels(self, bool include_data):
     chnlst = self.cppimc.get_channels(True,include_data)
-    chnlstjn = [jn.loads(chn.decode(errors="ignore")) for chn in chnlst]
+    chnlstjn = [jn.loads(chn.decode(get_codepage(chn),errors="ignore")) for chn in chnlst]
     return chnlstjn
 
   # print single channel/all channels

python/MANIFEST.in

@@ -2,3 +2,4 @@ include lib/*.hpp
 include *.cpp
 include *.pyx
 include *.pxd
+include VERSION

python/VERSION

@@ -1 +1 @@
-2.0.7
+2.0.16