Cleaned

read.py

@@ -34,6 +34,7 @@ toolPath = os.path.dirname(os.path.abspath(__file__))
 schema_lib = os.path.join(toolPath,'schemas','lib')
 
 def ERV(TextParser,read_sections_list, schema, code_tables_path):
+
     data_buffer = StringIO()
     valid_buffer = StringIO()
 
@@ -69,7 +70,8 @@ def ERV(TextParser,read_sections_list, schema, code_tables_path):
     # (source is either pd.io.parsers.TextFileReader or a file with chunksize specified on input):
     # This way it supports direct chunksize property inheritance if the input source was a pd.io.parsers.TextFileReader
     chunksize = TextParser.orig_options['chunksize'] if isinstance(TextParser,pd.io.parsers.TextFileReader) else None
-    # 'datetime' is not a valid pandas dtype: Only on output (on reading) will be then converted (via parse_dates) to datetime64[ns] type, cannot specify 'datetime' (of any kind) here: will fail
+    # 'datetime' is not a valid pandas dtype: Only on output (on reading) will be then converted (via parse_dates) to datetime64[ns] type,
+    # cannot specify 'datetime' (of any kind) here: would fail
     date_columns = [] # Needs to be the numeric index of the column, as seems not to be able to work with tupples....
     for i,element in enumerate(list(out_dtypes)):
         if out_dtypes.get(element) == 'datetime':
@@ -104,7 +106,7 @@ def read(source, data_model = None, data_model_path = None, sections = None,chun
             return
         elif not os.path.isfile(source):
             logging.error('Could not open data source file {}'.format(source))
-            logging.info('Otherwise, supported in-memory data sources are {}'.format(",".join([ str(x) for x in properties.supported_sources])))
+            logging.info('If input source was not a file: supported in-memory data sources are {}'.format(",".join([ str(x) for x in properties.supported_sources])))
             return
     if not validate_arg('sections',sections,list):
         return
@@ -124,6 +126,7 @@ def read(source, data_model = None, data_model_path = None, sections = None,chun
     else:
         model_path = data_model_path
     code_tables_path = os.path.join(model_path,'code_tables')
+
     # For future use: some work already done in schema reading
     if schema['header'].get('multiple_reports_per_line'):
         logging.error('File format not yet supported')
@@ -133,8 +136,7 @@ def read(source, data_model = None, data_model_path = None, sections = None,chun
     imodel = data_model if data_model else data_model_path
     logging.info("EXTRACTING DATA FROM MODEL: {}".format(imodel))
 
-    # 2.1. Define output 
-    # Subset data model sections to requested sections
+    # 2.1. Subset data model sections to requested sections
     parsing_order = schema['header'].get('parsing_order')
     if not sections:
         sections = [ x.get(y) for x in parsing_order for y in x ]
@@ -152,7 +154,7 @@ def read(source, data_model = None, data_model_path = None, sections = None,chun
     data,valid = ERV(TextParser,read_sections_list, schema, code_tables_path)
 
     # 3. Create out data attributes
-    logging.info("CREATING OUTPUT DATA ATTRIBUTES FROM DATA MODEL(S)")
+    logging.info("CREATING OUTPUT DATA ATTRIBUTES FROM DATA MODEL")
     data_columns = [ x for x in data ] if isinstance(data,pd.DataFrame) else data.orig_options['names']
     out_atts = schemas.df_schema(data_columns, schema)