Adda an --input-filter argument (#100)

Also change the --filter argument to --output-filter, but keep --filter as an alias.
Also add the possibility to use file extensions etc as filter names, if they are specified (which they rarely are).

Co-authored-by: Lennart Regebro <[email protected]>

CHANGES.rst

@@ -1,7 +1,16 @@
 2.0.2 (unreleased)
 ------------------
 
-- Nothing changed yet.
+- Add a --input-filter argument to specify a different file type than the
+  one LibreOffice will guess.
+
+- For consistency renamed --filter to --output-filter, but the --filter
+  will remain for backwards compatibility.
+
+- If you specify a non-existent filter, the list of filters is now alphabetical.
+
+- You can now use both the LibreOffice name, but also internal shorter names
+  and sometimes even file suffices to specify the filter.
 
 
 2.0.1 (2024-01-12)

src/unoserver/client.py

@@ -48,6 +48,7 @@ def convert(
         filtername=None,
         filter_options=[],
         update_index=True,
+        infiltername=None,
     ):
         """Converts a file from one type to another
 
@@ -92,6 +93,7 @@ def convert(
                 filtername,
                 filter_options,
                 update_index,
+                infiltername,
             )
             if result is not None:
                 # We got the file back over xmlrpc:
@@ -192,6 +194,11 @@ def converter_main():
         help="The file type/extension of the output file (ex pdf). Required when using stdout",
     )
     parser.add_argument(
+        "--input-filter",
+        help="The LibreOffice input filter to use (ex 'writer8'), if autodetect fails",
+    )
+    parser.add_argument(
+        "--output-filter",
         "--filter",
         default=None,
         help="The export filter to use when converting. It is selected automatically if not specified.",
@@ -250,9 +257,10 @@ def converter_main():
         indata=indata,
         outpath=args.outfile,
         convert_to=args.convert_to,
-        filtername=args.filter,
+        filtername=args.output_filter,
         filter_options=args.filter_options,
         update_index=args.update_index,
+        infiltername=args.input_filter,
     )
 
     if args.outfile is None:

src/unoserver/comparer.py

@@ -247,7 +247,9 @@ def compare(
                 )
 
             logger.info(f"Exporting to {outpath}")
-            logger.info(f"Using {filtername} export filter")
+            logger.info(
+                f"Using {filtername} export filter from {new_type} to {export_type}"
+            )
 
             output_props = (
                 PropertyValue(Name="FilterName", Value=filtername),

src/unoserver/converter.py

@@ -102,6 +102,17 @@ def find_filter(self, import_type, export_type):
         # No filter found
         return None
 
+    def get_available_import_filters(self):
+        # List import filters. You can only search on module, iflags and eflags,
+        # so the import and export types we have to test in a loop
+        import_filters = self.filter_service.createSubSetEnumerationByQuery(
+            "getSortedFilterList():iflags=1"
+        )
+
+        while import_filters.hasMoreElements():
+            # Filter DocumentService here
+            yield prop2dict(import_filters.nextElement())
+
     def get_available_export_filters(self):
         # List export filters. You can only search on module, iflags and eflags,
         # so the import and export types we have to test in a loop
@@ -113,8 +124,20 @@ def get_available_export_filters(self):
             # Filter DocumentService here
             yield prop2dict(export_filters.nextElement())
 
-    def get_available_filter_names(self):
-        return [filter["Name"] for filter in self.get_available_export_filters()]
+    def get_filter_names(self, filters):
+        names = {}
+        for flt in filters:
+            # Add all names and exstensions, etc in a mapping to the internal
+            # Libreoffice name, so we can map it.
+            # The actual name:
+            names[flt["Name"]] = flt["Name"]
+            # UserData sometimes has file extensions, etc.
+            # Skip empty data, and those weird file paths, and "true"...
+            for name in filter(
+                lambda x: x and x != "true" and "." not in x, flt["UserData"]
+            ):
+                names[name] = flt["Name"]
+        return names
 
     def convert(
         self,
@@ -125,6 +148,7 @@ def convert(
         filtername=None,
         filter_options=[],
         update_index=True,
+        infiltername=None,
     ):
         """Converts a file from one type to another
 
@@ -139,12 +163,25 @@ def convert(
 
         filtername: The name of the export filter to use for conversion. If None, it is auto-detected.
 
+        filter_options: A list of output filter options as strings, in a "OptionName=Value" format.
+
         update_index: Updates the index before conversion
 
+        infiltername: The name of the input filter, ie "writer8", "PowerPoint 3", etc.
+
         You must specify the inpath or the indata, and you must specify and outpath or a convert_to.
         """
-
         input_props = (PropertyValue(Name="ReadOnly", Value=True),)
+        if infiltername:
+            infilters = self.get_filter_names(self.get_available_import_filters())
+            if infiltername in infilters:
+                input_props += (
+                    PropertyValue(Name="FilterName", Value=infilters[infiltername]),
+                )
+            else:
+                raise ValueError(
+                    f"There is no '{infiltername}' import filter. Available filters: {sorted(infilters.keys())}"
+                )
 
         if inpath:
             # TODO: Verify that inpath exists and is openable, and that outdir exists, because uno's
@@ -171,6 +208,17 @@ def convert(
             import_path, "_default", 0, input_props
         )
 
+        if document is None:
+            # Could not load document, fail
+            if not inpath:
+                inpath = "<remote file>"
+            if not infiltername:
+                infiltername = "default"
+
+            error = f"Could not load document {inpath} using the {infiltername} filter."
+            logger.error(error)
+            raise RuntimeError(error)
+
         if update_index:
             # Update document indexes
             for ii in range(2):
@@ -216,10 +264,12 @@ def convert(
                 )
 
             if filtername is not None:
-                available_filter_names = self.get_available_filter_names()
+                available_filter_names = self.get_filter_names(
+                    self.get_available_export_filters()
+                )
                 if filtername not in available_filter_names:
                     raise RuntimeError(
-                        f"'{filtername}' is not a valid filter name. Valid filters are {available_filter_names}"
+                        f"There is no '{filtername}' export-filter. Available filters: {sorted(available_filter_names)}"
                     )
             else:
                 filtername = self.find_filter(import_type, export_type)
@@ -229,7 +279,9 @@ def convert(
                     )
 
             logger.info(f"Exporting to {outpath}")
-            logger.info(f"Using {filtername} export filter")
+            logger.info(
+                f"Using {filtername} export filter from {infiltername} to {export_type}"
+            )
 
             filter_data = []
             for option in filter_options:

src/unoserver/server.py

@@ -100,6 +100,7 @@ def convert(
                 filtername=None,
                 filter_options=[],
                 update_index=True,
+                infiltername=None,
             ):
                 if indata is not None:
                     indata = indata.data
@@ -114,6 +115,7 @@ def convert(
                     filtername,
                     filter_options,
                     update_index,
+                    infiltername,
                 )
                 return result