Extract slide notes from PowerPoint

extract_ppt_notes.py

import zipfile
import cStringIO
from xml.dom import minidom


class ExtractPptNotes(object):

    _chunks = None

    def __init__(self, file_data):

        self._chunks = self.__inflate_file_data(file_data)

    def get_all_slide_notes(self):

        notes = {}

        for slide_name in self._chunks[u'ppt/notesSlides']:
            notes[slide_name] = self.get_slide_notes(slide_name)

        return notes

    def get_slide_notes(self, slide_name):

        notes = []

        slide_xml_string = self._chunks[u'ppt/notesSlides'][slide_name]
        slide_xml = minidom.parseString(slide_xml_string)
        nodes = slide_xml.getElementsByTagName('a:t')
        for node in nodes:
            for child_node in node.childNodes:
                if child_node.nodeType == child_node.TEXT_NODE:
                    notes.append(child_node.nodeValue)

        return notes
        
    def __inflate_file_data(self, file_data):

        # Inflate the file
        file_stream = cStringIO.StringIO(file_data)
        openxml_zip = zipfile.ZipFile(file_stream, 'r', zipfile.ZIP_DEFLATED)

        # Extract each xml file contained in the openxml doc into a chunk store
        # where chunks[chunk_path][chunk_filename] = chunk_data
        chunks = {}
        for chunk_file_path in openxml_zip.namelist():
            chunk_data = openxml_zip.read(chunk_file_path)
            chunk_path_info = chunk_file_path.rsplit('/', 1)
            if len(chunk_path_info) < 2:
                chunk_name = chunk_path_info[0]
                chunk_path = '/'
            else:
                chunk_name = chunk_path_info[1]
                chunk_path = chunk_path_info[0]
            if not chunk_path in chunks:
                chunks[chunk_path] = {}
            chunks[chunk_path][chunk_name] = chunk_data

        openxml_zip.close()
        file_stream.close()

        return chunks