Refactor office document handling

3 files changed, 75 insertions, 92 deletions

diff --git a/src/abstract.py b/src/abstract.py
index c2d282f..1f8ce6e 100644
--- a/src/abstract.py
+++ b/src/abstract.py
@@ -6,8 +6,8 @@ class AbstractParser(object):
         self.filename = filename
         self.output_filename = filename + '.cleaned'
 
-    def get_meta(self):
+    def get_meta(self) -> dict:
         raise NotImplementedError
 
-    def remove_all(self):
+    def remove_all(self) -> bool:
         raise NotImplementedError
diff --git a/src/libreoffice.py b/src/libreoffice.py
deleted file mode 100644
index 809ae3c..0000000
--- a/src/libreoffice.py
+++ /dev/null
@@ -1,68 +0,0 @@
-import re
-import subprocess
-import json
-import zipfile
-import tempfile
-import shutil
-import os
-
-from . import abstract, parser_factory
-
-class LibreOfficeParser(abstract.AbstractParser):
-    mimetypes = {
-            'application/vnd.oasis.opendocument.text',
-            'application/vnd.oasis.opendocument.spreadsheet',
-            'application/vnd.oasis.opendocument.presentation',
-            'application/vnd.oasis.opendocument.graphics',
-            'application/vnd.oasis.opendocument.chart'
-    }
-
-    def get_meta(self):
-        """
-        Yes, I know that parsing xml with regexp ain't pretty,
-        be my guest and fix it if you want.
-        """
-        metadata = {}
-        zipin = zipfile.ZipFile(self.filename)
-        for item in zipin.namelist():
-            if item == 'meta.xml':
-                content = zipin.read(item).decode('utf-8')
-                for (key, value) in re.findall(r"<((?:meta|dc).+?)>(.+)</\1>", content, re.I):
-                    metadata[key] = value
-                if not metadata:  # better safe than sorry
-                    metadata[item] = 'harmful content'
-        zipin.close()
-        return metadata
-
-    def __clean_zipinfo(self, zipinfo:zipfile.ZipInfo) -> zipfile.ZipInfo:
-        zipinfo.compress_type = zipfile.ZIP_DEFLATED
-        zipinfo.create_system = 3  # Linux
-        zipinfo.comment = b''
-        zipinfo.date_time = (1980, 1, 1, 0, 0, 0)
-        return zipinfo
-
-    def remove_all(self):
-        zin = zipfile.ZipFile(self.filename, 'r')
-        zout = zipfile.ZipFile(self.output_filename, 'w')
-        temp_folder = tempfile.mkdtemp()
-
-        for item in zin.infolist():
-            if item.filename[-1] == '/':
-                continue  # `is_dir` is added in Python3.6
-            elif item.filename == 'meta.xml':
-                continue  # don't keep metadata files
-
-            zin.extract(member=item, path=temp_folder)
-            tmp_parser = parser_factory.get_parser(os.path.join(temp_folder, item.filename))
-            if tmp_parser is None:
-                print("%s isn't supported" % item.filename)
-                continue
-            tmp_parser.remove_all()
-            zinfo = zipfile.ZipInfo(item.filename)
-            item = self.__clean_zipinfo(item)
-            with open(tmp_parser.output_filename, 'rb') as f:
-                zout.writestr(zinfo, f.read())
-        shutil.rmtree(temp_folder)
-        zout.close()
-        zin.close()
-        return True
diff --git a/src/office.py b/src/office.py
index a729f2f..5083308 100644
--- a/src/office.py
+++ b/src/office.py
@@ -1,14 +1,34 @@
+import json
+import os
 import re
+import shutil
 import subprocess
-import json
-import zipfile
 import tempfile
-import shutil
-import os
+import zipfile
 
 from . import abstract, parser_factory
 
-class OfficeParser(abstract.AbstractParser):
+class ArchiveBasedAbstractParser(abstract.AbstractParser):
+    def _clean_zipinfo(self, zipinfo:zipfile.ZipInfo) -> zipfile.ZipInfo:
+        zipinfo.compress_type = zipfile.ZIP_DEFLATED
+        zipinfo.create_system = 3  # Linux
+        zipinfo.comment = b''
+        zipinfo.date_time = (1980, 1, 1, 0, 0, 0)
+        return zipinfo
+
+    def _clean_internal_file(self, item:zipfile.ZipInfo, temp_folder:str, zin:zipfile.ZipFile, zout:zipfile.ZipFile):
+        zin.extract(member=item, path=temp_folder)
+        tmp_parser = parser_factory.get_parser(os.path.join(temp_folder, item.filename))
+        if tmp_parser is None:
+            print("%s isn't supported" % item.filename)
+            return
+        tmp_parser.remove_all()
+        zinfo = zipfile.ZipInfo(item.filename)
+        item = self._clean_zipinfo(item)
+        with open(tmp_parser.output_filename, 'rb') as f:
+            zout.writestr(zinfo, f.read())
+
+class MSOfficeParser(ArchiveBasedAbstractParser):
     mimetypes = {
             'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
             'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
@@ -33,12 +53,6 @@ class OfficeParser(abstract.AbstractParser):
         zipin.close()
         return metadata
 
-    def __clean_zipinfo(self, zipinfo:zipfile.ZipInfo) -> zipfile.ZipInfo:
-        zipinfo.compress_type = zipfile.ZIP_DEFLATED
-        zipinfo.create_system = 3  # Linux
-        zipinfo.comment = b''
-        zipinfo.date_time = (1980, 1, 1, 0, 0, 0)
-        return zipinfo
 
     def remove_all(self):
         zin = zipfile.ZipFile(self.filename, 'r')
@@ -52,20 +66,57 @@ class OfficeParser(abstract.AbstractParser):
                 if not item.filename.endswith('.rels'):
                     continue  # don't keep metadata files
             if item.filename in self.files_to_keep:
-                item = self.__clean_zipinfo(item)
+                item = self._clean_zipinfo(item)
                 zout.writestr(item, zin.read(item))
                 continue
 
-            zin.extract(member=item, path=temp_folder)
-            tmp_parser = parser_factory.get_parser(os.path.join(temp_folder, item.filename))
-            if tmp_parser is None:
-                print("%s isn't supported" % item.filename)
-                continue
-            tmp_parser.remove_all()
-            zinfo = zipfile.ZipInfo(item.filename)
-            item = self.__clean_zipinfo(item)
-            with open(tmp_parser.output_filename, 'rb') as f:
-                zout.writestr(zinfo, f.read())
+            self._clean_internal_file(item, temp_folder, zin, zout)
+
+        shutil.rmtree(temp_folder)
+        zout.close()
+        zin.close()
+        return True
+
+
+
+class LibreOfficeParser(ArchiveBasedAbstractParser):
+    mimetypes = {
+            'application/vnd.oasis.opendocument.text',
+            'application/vnd.oasis.opendocument.spreadsheet',
+            'application/vnd.oasis.opendocument.presentation',
+            'application/vnd.oasis.opendocument.graphics',
+            'application/vnd.oasis.opendocument.chart'
+    }
+
+    def get_meta(self):
+        """
+        Yes, I know that parsing xml with regexp ain't pretty,
+        be my guest and fix it if you want.
+        """
+        metadata = {}
+        zipin = zipfile.ZipFile(self.filename)
+        for item in zipin.namelist():
+            if item == 'meta.xml':
+                content = zipin.read(item).decode('utf-8')
+                for (key, value) in re.findall(r"<((?:meta|dc|cp).+?)>(.+)</\1>", content, re.I):
+                    metadata[key] = value
+                if not metadata:  # better safe than sorry
+                    metadata[item] = 'harmful content'
+        zipin.close()
+        return metadata
+
+    def remove_all(self):
+        zin = zipfile.ZipFile(self.filename, 'r')
+        zout = zipfile.ZipFile(self.output_filename, 'w')
+        temp_folder = tempfile.mkdtemp()
+
+        for item in zin.infolist():
+            if item.filename[-1] == '/':
+                continue  # `is_dir` is added in Python3.6
+            elif item.filename == 'meta.xml':
+                continue  # don't keep metadata files
+
+            self._clean_internal_file(item, temp_folder, zin, zout)
 
         shutil.rmtree(temp_folder)
         zout.close()