music-duplicates: python3 compatibility

Open files in binary mode instead of text mode.

scripts/music-duplicates

@@ -0,0 +1,83 @@
+#!/usr/bin/env python
+# SPDX-License-Identifier: BSD-3-Clause
+#
+# Find duplicated music tracks and create a summary report of music
+# that the project needs.
+
+from glob import glob
+import hashlib
+import os
+import re
+import sys
+
+PHASE1_MATCH_RE = re.compile(r'(e\dm\d)', re.I)
+PHASE2_MATCH_RE = re.compile(r'(map\d\d)', re.I)
+FREEDM_MATCH_RE = re.compile(r'(dm\d\d)', re.I)
+
+def get_music_tracks():
+	"""Returns a dictionary mapping from MIDI file SHA1
+	   to a list of game tracks that use that MIDI."""
+	result = {}
+	musics_path = os.path.join(os.path.dirname(sys.argv[0]), '../musics')
+	for mus in glob('%s/*.mid' % musics_path):
+		with open(mus, 'rb') as f:
+			contents = f.read()
+			m = hashlib.sha1()
+			m.update(contents)
+			digest = m.digest()
+			basename = os.path.basename(mus)
+			result.setdefault(digest, []).append(basename)
+	return result
+
+def get_prime_track(tracks):
+	"""Given a list of tracks that all use the same MIDI, find the
+	   "prime" one (the one that isn't a reuse/duplicate)."""
+	# We have almost all Phase 2 tracks fulfilled. So if the same
+	# track is used in Phase 1 and Phase 2, or Phase 2 and FreeDM,
+	# the Phase 2 track is probably the leader.
+	phase2_tracks = [x for x in tracks if PHASE2_MATCH_RE.search(x)]
+	if len(phase2_tracks) == 1:
+		return phase2_tracks[0]
+
+	# FreeDM music has been hand-picked. So if it is used for both
+	# Phase 1 and FreeDM, assume it's probably a FreeDM track.
+	freedm_tracks = [x for x in tracks if FREEDM_MATCH_RE.search(x)]
+	if len(freedm_tracks) == 1:
+		return freedm_tracks[0]
+
+	# We're out of options. Pick the first one in the list.
+	#print "Warning: Don't know which of %s is the leader." % tracks
+	return sorted(tracks)[0]
+
+def find_missing_tracks(tracks):
+	"""Given a dictionary of tracks, get a list of "missing" tracks."""
+	result = []
+	for midi, tracks in tracks.items():
+		if len(tracks) < 2:
+			continue
+		prime_track = get_prime_track(tracks)
+		result.extend(x for x in tracks if x != prime_track)
+	return result
+
+def tracks_matching_regexp(tracks, regexp):
+	return set([x for x in tracks if regexp.search(x)])
+
+def print_report(title, tracks):
+	if len(tracks) == 0:
+		return
+	print(title)
+	for track in sorted(tracks):
+		print('\t%s' % track.replace('.mid', '').upper())
+	print('')
+
+missing_tracks = set(find_missing_tracks(get_music_tracks()))
+phase1_tracks = tracks_matching_regexp(missing_tracks, PHASE1_MATCH_RE)
+phase2_tracks = tracks_matching_regexp(missing_tracks, PHASE2_MATCH_RE)
+freedm_tracks = tracks_matching_regexp(missing_tracks, FREEDM_MATCH_RE)
+other_tracks = missing_tracks - phase1_tracks - phase2_tracks - freedm_tracks
+
+print('=== Missing tracks (tracks currently using duplicates):\n')
+print_report('Phase 1 tracks:', phase1_tracks)
+print_report('Phase 2 tracks:', phase2_tracks)
+print_report('FreeDM tracks:', freedm_tracks)
+print_report('Other tracks:', other_tracks)