added group_name and obtain_channel_id to cython, added extract_all.py producing .csv dump of all .tdm and .tdx data

README.md

@@ -2,7 +2,7 @@
 
 ## C++ core
 
-- XML parser employs pugixml  https://pugixml.org/
+- XML parser employs pugixml https://pugixml.org/ and it's repository https://github.com/zeux/pugixml
 - analyse .tdm file
 - buffer binary .tdx data
 - convert binary data to datatypes according to given .tdm information

extract_all.py

@@ -0,0 +1,62 @@
+import tdm_ripper
+import numpy as np
+import argparse
+import re
+
+parser = argparse.ArgumentParser(description='provide path of both .tdm and corresponding .tdx file')
+parser.add_argument('tdm_file',type=str,help='path of .tdm file')
+parser.add_argument('tdx_file',type=str,help='path of .tdx file')
+#parser.add_argument('--tdx_file',type=str,help='path of .tdx file',default='')
+parser.add_argument('--out_directory',type=str,help='choose directory where to write data',default='./')
+parser.add_argument('--prefix_name',type=str,help='provide dataset name used as filename prefix',default='')
+args = parser.parse_args()
+#print(args)
+
+# process arguments
+tdmpath = args.tdm_file #"samples/SineData.tdm"
+tdxpath = args.tdx_file #"samples/SineData.tdx"
+outdirx = args.out_directory
+fprefix = args.prefix_name
+
+# if no prefix is given, .tdm filename will be used
+if fprefix == '' :
+    fprefix = tdmpath.rstrip('.tdm').split('/')[-1]
+    # TODO better use os.path staff !!
+
+#print(fprefix)
+
+# create instance of ripper class
+RP = tdm_ripper.pytdmripper(tdmpath.encode('utf-8'),tdxpath.encode('utf-8'))
+
+# provide overview over available groups and channels
+#RP.show_groups()
+#RP.show_channels()
+
+# obtain number of available groups and channels
+numgr = RP.num_groups()
+numch = RP.num_channels()
+
+# dump all meta information
+RP.print_meta((outdirx+fprefix+'.csv').encode('utf-8'))
+
+# dump all available groups and channels
+for g in range(0,numgr):
+    numgrch = RP.no_channels(g)
+    for c in range(0,numgrch):
+        #print(str(g).rjust(10)+str(c).rjust(10))
+        # print(str(RP.channel_length(g,c)))
+        # print(RP.channel_name(g,c))
+        #print(RP.channel(g,c))
+        # obtained overall channel id
+        chid = RP.obtain_channel_id(g,c)
+        # get group's and channel's name
+        gname = RP.group_name(g)
+        cname = RP.channel_name(g,c)
+        #print(gname.rjust(30)+cname.rjust(30))
+        # use regular expression replacement to sanitize group and channel names
+        gname = re.sub('[!@#$%^&*()-+= ,]','',gname)
+        cname = re.sub('[!@#$%^&*()-+= ,]','',cname)
+        # generate filename
+        fichan = fprefix + '_' + str(g+1) + '_' + str(c+1) + '_' + gname + '_' + cname + '.csv'
+        # print channel
+        RP.print_channel(chid,(outdirx+fichan).encode('utf-8'))

lib/tdm_ripper.cpp

@@ -542,13 +542,13 @@ std::vector<double> tdm_ripper::get_channel(int channelid)
   std::vector<double> chann = convert_channel(channel_ext_[channelid]);
 
   // check if converted value is within expected range
-  // for ( int i = 0; i < (int)chann.size(); i++ )
-  // {
-  //   if ( chann[i] < minmax_[channelid-1].first
-  //     || chann[i] > minmax_[channelid-1].second ) std::cout<<std::setw(20)<<chann[i]<<std::setw(20)<<minmax_[channelid-1].first<<std::setw(20)<<minmax_[channelid-1].second<<"\n";
-  //   assert( chann[i] >= minmax_[channelid-1].first  - 1.0e-6
-  //        && chann[i] <= minmax_[channelid-1].second + 1.0e-6 );
-  // }
+  for ( int i = 0; i < (int)chann.size(); i++ )
+  {
+    // if ( chann[i] < minmax_[channelid-1].first
+    //   || chann[i] > minmax_[channelid-1].second ) std::cout<<std::setw(20)<<chann[i]<<std::setw(20)<<minmax_[channelid-1].first<<std::setw(20)<<minmax_[channelid-1].second<<"\n";
+    assert( chann[i] >= minmax_[channelid-1].first  - 1.0e-6
+         && chann[i] <= minmax_[channelid-1].second + 1.0e-6 );
+  }
 
   return chann;
 }

pytdm_ripper.pyx

@@ -35,12 +35,18 @@ cdef class pytdmripper:
     def channel_name(self,int groupid,int channelid):
         return self.cripp.channel_name(groupid,channelid).decode('utf-8')
 
+    def group_name(self,int groupid):
+        return self.cripp.group_name(groupid).decode('utf-8')
+
     def channel_unit(self,int groupid,int channelid):
         return (self.cripp.channel_unit(groupid,channelid))
 
     def channel_exists(self,int groupid, string channelname):
         return self.cripp.channel_exists(groupid,channelname)
 
+    def obtain_channel_id(self,int groupid, int channelid):
+      return self.cripp.obtain_channel_id(groupid,channelid)
+
     def get_channel(self, int channelid):
         return np.asarray(self.cripp.get_channel(channelid))
 

tdm_ripper.pxd

@@ -19,8 +19,10 @@ cdef extern from "tdm_ripper.hpp":
         int num_groups()
         int no_channel_groups()
         string channel_name(int,int)
+        string group_name(int)
         string channel_unit(int,int)
         int channel_exists(int,string)
+        int obtain_channel_id(int,int)
         vector[double] get_channel(int)
         int channel_length(int,int)
         string time_stamp(int,bool)