erp5_wendelin_data_lake_ingestion: fix check md5 script

bt5/erp5_wendelin_data_lake_ingestion/SkinTemplateItem/portal_skins/erp5_wendelin_data_lake/DataSet_checkMd5DataStreamList.py

 """
-  Script to check that a filesystem md5sum of a folder (uploaded to file_system_checksum File)
-  is properly uploaded to Wendelin Data Lake.
+  Script to check that a data set is properly uploaded
+  to Wendelin Data Lake.
+
+  How to use it: create a file_system_checksum file containing md5sum
+  values of all dataset files uploaded with the following format:
 
   Format of is the same as md5sum's output:
   <md5_sum>    <filename.extension>
 
+  It can be generated in the original data set folder outside wendelin by doing md5sum * > output.txt
 """
 
+import os.path
 data =  str(context.file_system_checksum).strip()
 lines = data.split("\n")
 print "Total files = ", len(lines)
+print
+check_result = True
 for line in lines[:]:
   md5_checksum = line[:32].strip()
   full_filename = line[32:].strip()
-
-  # check Data stream for this hash exists
-  filename, extension = full_filename.split(".")
+  filename, extension = os.path.splitext(full_filename)
+  extension = extension[1:]
   reference = "%s/%s/%s" %(data_set_reference, filename, extension)
   catalog_kw = {"portal_type": "Data Stream",
                 "reference": reference}
   data_stream = context.portal_catalog.getResultValue(**catalog_kw)
   if data_stream is None:
     print "[NOT FOUND]", reference
+    check_result = False
   else:
     is_upload_ok = (data_stream.getVersion()==md5_checksum)
     print md5_checksum, filename, data_stream is not None, is_upload_ok
-
+    if not is_upload_ok:
+      check_result = False
+print
+if check_result:
+  print "[OK] Data set correctly uploaded"
+else:
+  print "[ERROR] Data set was not correctly uploaded"
 return printed