Refactor.

README.md

@@ -130,6 +130,7 @@ options, please check the **--help** command:
 Adding tasks is very simple. You can have your tasks in three formats:
 
  * JSON
+ * Excel (xlsx from 2010. It imports the first sheet)
  * CSV
  * PO (any po file that you want to translate)
  * PROPERTIES (any PROPERTIES file that you want to translate)

helpers.py

@@ -134,58 +134,70 @@ def _update_project(config, task_presenter, results,
         raise
 
 
+def _load_data(data_file, data_type):
+    """Load data from CSV, JSON, Excel, ..., formats."""
+    raw_data = data_file.read()
+    if data_type is None:
+        data_type = data_file.name.split('.')[-1]
+    # Data list to process
+    data = []
+    # JSON type
+    if data_type == 'json':
+        data = json.loads(raw_data)
+        return data
+    # CSV type
+    elif data_type == 'csv':
+        csv_data = StringIO(raw_data)
+        reader = csv.DictReader(csv_data, delimiter=',')
+        for line in reader:
+            data.append(line)
+        return data
+    elif data_type in ['xlsx', 'xlsm', 'xltx', 'xltm']:
+        excel_data = StringIO(raw_data)
+        wb = openpyxl.load_workbook(excel_data)
+        ws = wb.active
+        # First headers
+        headers = []
+        for row in ws.iter_rows(max_row=1):
+            for cell in row:
+                tmp = '_'.join(cell.value.split(" ")).lower()
+                headers.append(tmp)
+        # Simulate DictReader
+        for row in ws.iter_rows(row_offset=1):
+            values = []
+            for cell in row:
+                values.append(cell.value)
+            tmp = dict(itertools.izip(headers, values))
+            if len(values) == len(headers) and not row_empty(values):
+                data.append(tmp)
+        return data
+    # PO type
+    elif data_type == 'po':
+        po = polib.pofile(raw_data)
+        for entry in po.untranslated_entries():
+            data.append(entry.__dict__)
+        return data
+    # PROPERTIES type (used in Java and Firefox extensions)
+    elif data_type == 'properties':
+        lines = raw_data.split('\n')
+        for l in lines:
+            if l:
+                var_id, string = l.split('=')
+                tmp = dict(var_id=var_id, string=string)
+                data.append(tmp)
+        return data
+    else:
+        return data
+
+
 def _add_tasks(config, tasks_file, tasks_type, priority, redundancy):
     """Add tasks to a project."""
     try:
         project = find_project_by_short_name(config.project['short_name'],
                                              config.pbclient,
                                              config.all)
-        tasks = tasks_file.read()
-        if tasks_type is None:
-            tasks_type = tasks_file.name.split('.')[-1]
-        # Data list to process
-        data = []
-        # JSON type
-        if tasks_type == 'json':
-            data = json.loads(tasks)
-        # CSV type
-        elif tasks_type == 'csv':
-            csv_data = StringIO(tasks)
-            reader = csv.DictReader(csv_data, delimiter=',')
-            for line in reader:
-                data.append(line)
-        elif tasks_type in ['xlsx', 'xlsm', 'xltx', 'xltm']:
-            excel_data = StringIO(tasks)
-            wb = openpyxl.load_workbook(excel_data)
-            ws = wb.active
-            # First headers
-            headers = []
-            for row in ws.iter_rows(max_row=1):
-                for cell in row:
-                    tmp = '_'.join(cell.value.split(" ")).lower()
-                    headers.append(tmp)
-            # Simulate DictReader
-            for row in ws.iter_rows(row_offset=1):
-                values = []
-                for cell in row:
-                    values.append(cell.value)
-                tmp = dict(itertools.izip(headers, values))
-                if len(values) == len(headers) and not row_empty(values):
-                    data.append(tmp)
-        # PO type
-        elif tasks_type == 'po':
-            po = polib.pofile(tasks)
-            for entry in po.untranslated_entries():
-                data.append(entry.__dict__)
-        # PROPERTIES type (used in Java and Firefox extensions)
-        elif tasks_type == 'properties':
-            lines = tasks.split('\n')
-            for l in lines:
-                if l:
-                    var_id, string = l.split('=')
-                    tmp = dict(var_id=var_id, string=string)
-                    data.append(tmp)
-        else:
+        data = _load_data(tasks_file, tasks_type)
+        if len(data) == 0:
             return ("Unknown format for the tasks file. Use json, csv, po or "
                     "properties.")
         # Check if for the data we have to auto-throttle task creation