Implement --write-successful-file and --skip-file

spotdl/command_line/__main__.py

@@ -39,7 +39,6 @@ def main():
         sys.exit(5)
 
     logging_level = argument_handler.get_logging_level()
-    print(logging_level)
     logger = set_logger(logging_level)
     try:
         spotdl = Spotdl(argument_handler)

spotdl/command_line/arguments.py

@@ -189,15 +189,17 @@ def get_arguments(config_base=_CONFIG_BASE):
         "to the next track (if any)",
         action="store_true",
     )
-    # parser.add_argument(
-    #     "--processor",
-    #     default=defaults["processor"],
-    #     choices={"synchronous", "threaded"},
-    #     help='list downloading strategy: - "synchronous" downloads '
-    #     'tracks one-by-one. - "threaded" (highly experimental at the '
-    #     'moment! expect it to slash & burn) pre-fetches the next '
-    #     'track\'s metadata for more efficient downloading'
-    # )
+    parser.add_argument(
+        "--processor",
+        default="synchronous",
+        choices={"synchronous", "threaded"},
+        # help='list downloading strategy: - "synchronous" downloads '
+        # 'tracks one-by-one. - "threaded" (highly experimental at the '
+        # 'moment! expect it to slash & burn) pre-fetches the next '
+        # 'track\'s metadata for more efficient downloading'
+        # XXX: Still very experimental to be exposed
+        help=argparse.SUPPRESS,
+    )
     parser.add_argument(
         "-ns",
         "--no-spaces",

spotdl/command_line/lib.py

@@ -208,6 +208,17 @@ class Spotdl:
 
         return to_overwrite
 
+    def generate_temp_filename(self, filename, for_stdout=False):
+        if for_stdout:
+            return filename
+        return "{filename}.temp".format(filename=filename)
+
+    def output_filename_filter(self, allow_spaces):
+        replace_spaces_with_underscores = not allow_spaces
+        if replace_spaces_with_underscores:
+            return lambda s: s.replace(" ", "_")
+        return lambda s: s
+
     def download_track_from_metadata(self, metadata):
         track = Track(metadata, cache_albumart=(not self.arguments["no_metadata"]))
         stream = metadata["streams"].get(
@@ -229,10 +240,7 @@ class Spotdl:
             )
         )
         download_to_stdout = filename == "-"
-        if download_to_stdout:
-            temp_filename = filename
-        else:
-            temp_filename = "{filename}.temp".format(filename=filename)
+        temp_filename = self.generate_temp_filename(filename, for_stdout=download_to_stdout)
 
         to_skip_download = self.arguments["dry_run"]
         if os.path.isfile(filename):
@@ -249,6 +257,17 @@ class Spotdl:
         if not self.arguments["no_metadata"]:
             metadata["lyrics"].start()
 
+        filter_space_chars = self.output_filename_filter(not self.arguments["no_spaces"])
+        directory = os.path.dirname(
+            spotdl.metadata.format_string(
+                self.arguments["output_file"],
+                metadata,
+                output_extension=output_extension,
+                sanitizer=filter_space_chars
+            )
+        )
+        os.makedirs(directory or ".", exist_ok=True)
+
         logger.info('Downloading to "{filename}"'.format(filename=filename))
         if self.arguments["no_encode"]:
             track.download(stream, temp_filename)
@@ -282,18 +301,33 @@ class Spotdl:
         except TypeError:
             logger.warning("Cannot apply metadata on provided output format.")
 
+    def strip_and_filter_duplicates(self, tracks):
+        filtered_tracks = spotdl.util.remove_duplicates(
+            tracks,
+            condition=lambda x: x,
+            operation=str.strip
+        )
+        return filtered_tracks
+
+    def filter_against_skip_file(self, items, skip_file):
+        skip_items = spotdl.util.readlines_from_nonbinary_file(skip_file)
+        filtered_skip_items = self.strip_and_filter_duplicates(skip_items)
+        filtered_items = [item for item in items if not item in filtered_skip_items]
+        return filtered_items
+
     def download_tracks_from_file(self, path):
         logger.info(
             "Checking and removing any duplicate tracks in {}.".format(path)
         )
         tracks = spotdl.util.readlines_from_nonbinary_file(path)
-        # Remove duplicates and empty elements
-        # Also strip whitespaces from elements (if any)
-        tracks = spotdl.util.remove_duplicates(
-            tracks,
-            condition=lambda x: x,
-            operation=str.strip
-        )
+        tracks = self.strip_and_filter_duplicates(tracks)
+
+        if self.arguments["skip_file"]:
+            len_tracks_before = len(tracks)
+            tracks = self.filter_against_skip_file(tracks, self.arguments["skip_file"])
+            logger.info("Skipping {} tracks due to matches in skip file.".format(
+                len_tracks_before - len(tracks))
+            )
         # Overwrite file
         spotdl.util.writelines_to_nonbinary_file(path, tracks)
 
@@ -326,9 +360,9 @@ class Spotdl:
             except (NoYouTubeVideoFoundError, NoYouTubeVideoMatchError) as e:
                 logger.error("{err}".format(err=e.args[0]))
             else:
-                if self.arguments["write_sucessful_file"]:
-                    with open(self.arguments["write_sucessful_file"], "a") as fout:
-                        fout.write(track)
+                if self.arguments["write_successful_file"]:
+                    with open(self.arguments["write_successful_file"], "a") as fout:
+                        fout.write("{}\n".format(track))
             finally:
                 spotdl.util.writelines_to_nonbinary_file(path, tracks[position:])
                 print("", file=sys.stderr)

spotdl/util.py

@@ -45,14 +45,14 @@ class ThreadWithReturnValue(threading.Thread):
 
 def merge(base, overrider):
     """ Override base dict with an overrider dict, recursively. """
-    for key, value in base.items():
+    for key, value in overrider.items():
         if isinstance(value, dict):
-            subitem = overrider.setdefault(key, {})
-            merge(value, subitem)
+            subitem = base.setdefault(key, {})
+            merge(subitem, value)
         else:
-            overrider[key] = value
+            base[key] = value
 
-    return overrider
+    return base
 
 
 def prompt_user_for_selection(items):