Add support for getting an office document page count.

mayan/apps/converter/backends/python.py

@@ -30,6 +30,16 @@ Image.init()
 logger = logging.getLogger(__name__)
 
 
+class IteratorIO(object):
+    def __init__(self, iterator):
+        self.file_buffer = StringIO()
+
+        for chunk in iterator:
+            self.file_buffer.write(chunk)
+
+        self.file_buffer.seek(0)
+
+
 class Python(ConverterBase):
 
     def convert(self, *args, **kwargs):
@@ -53,19 +63,26 @@ class Python(ConverterBase):
                 fs_cleanup(input_filepath)
 
     def get_page_count(self):
+        super(Python, self).get_page_count()
+
         page_count = 1
 
-        if self.mime_type == 'application/pdf':
+        if self.mime_type == 'application/pdf' or self.soffice_file:
             # If file is a PDF open it with slate to determine the page count
+            if self.soffice_file:
+                file_object = IteratorIO(self.soffice_file).file_buffer
+            else:
+                file_object = self.file_object
+
             try:
-                pages = slate.PDF(self.file_object)
+                pages = slate.PDF(file_object)
             except Exception as exception:
                 logger.error('Slate exception; %s', exception)
                 raise
             else:
                 return len(pages)
             finally:
-                self.file_object.seek(0)
+                file_object.seek(0)
         else:
             try:
                 image = Image.open(self.file_object)

mayan/apps/converter/classes.py

@@ -135,7 +135,7 @@ class ConverterBase(object):
         self.file_object = file_object
         self.image = None
         self.mime_type = mime_type or get_mimetype(file_object=file_object, mimetype_only=False)[0]
-        self.soffice_file_object = None
+        self.soffice_file = None
 
     def to_pdf(self):
         if self.mime_type in CONVERTER_OFFICE_FILE_MIMETYPES:
@@ -183,7 +183,10 @@ class ConverterBase(object):
             self.image = transformation.execute_on(self.image)
 
     def get_page_count(self):
-        raise NotImplementedError
+        try:
+            self.soffice_file = self.to_pdf()
+        except InvalidOfficeFormat:
+            pass
 
 
 class BaseTransformation(object):