Even more typing errors fixes.

src/tscut/h264/bitstream.py

@@ -18,7 +18,7 @@ from tscut.exceptions import UnimplementedFeatureError
 logger = logging.getLogger(__name__)
 
 @typechecked
-def read_bit(buf:bytes, bit_position: int) -> tuple[int, int]:
+def read_bit(buf:bytes|bytearray, bit_position: int) -> tuple[int, int]:
     """
     Read a single bit from a byte buffer.
 
@@ -41,7 +41,7 @@ def read_bit(buf:bytes, bit_position: int) -> tuple[int, int]:
     return bit_position+1, bit
 
 @typechecked
-def read_boolean(buf:bytes, bit_position: int) -> tuple[int, bool]:
+def read_boolean(buf:bytes|bytearray, bit_position: int) -> tuple[int, bool]:
     """
     Read a boolean value from a byte buffer.
 
@@ -61,7 +61,7 @@ def read_boolean(buf:bytes, bit_position: int) -> tuple[int, bool]:
     return bit_position, b==1
 
 @typechecked
-def read_bits(buf:bytes, bit_position: int, nb_bits: int)  -> tuple[int, int]:
+def read_bits(buf:bytes|bytearray, bit_position: int, nb_bits: int)  -> tuple[int, int]:
     v = 0
     for _ in range(nb_bits):
         bit_position, bit = read_bit(buf, bit_position)
@@ -69,22 +69,22 @@ def read_bits(buf:bytes, bit_position: int, nb_bits: int)  -> tuple[int, int]:
     return bit_position, v
 
 @typechecked
-def read_byte(buf:bytes, bit_position: int)  -> tuple[int, int]:
+def read_byte(buf:bytes|bytearray, bit_position: int)  -> tuple[int, int]:
     bit_position, b = read_bits(buf, bit_position, 8)
     return bit_position, b
 
 @typechecked
-def read_word(buf:bytes, bit_position: int) -> tuple[int, int]:
+def read_word(buf:bytes|bytearray, bit_position: int) -> tuple[int, int]:
     bit_position, w = read_bits(buf, bit_position, 16)
     return bit_position, w
 
 @typechecked
-def read_long(buf:bytes, bit_position: int) -> tuple[int, int]:
+def read_long(buf:bytes|bytearray, bit_position: int) -> tuple[int, int]:
     bit_position, value = read_bits(buf, bit_position, 32)
     return bit_position, value
 
 @typechecked
-def read_unsigned_exp_golomb(buf:bytes, bit_position: int) -> tuple[int, int]:
+def read_unsigned_exp_golomb(buf:bytes|bytearray, bit_position: int) -> tuple[int, int]:
     nb_zeroes=0
     while True:
         bit_position, b = read_bit(buf, bit_position)
@@ -98,7 +98,7 @@ def read_unsigned_exp_golomb(buf:bytes, bit_position: int) -> tuple[int, int]:
     return bit_position, v-1
 
 @typechecked
-def read_signed_exp_golomb(buf:bytes, bit_position: int) -> tuple[int, int]:
+def read_signed_exp_golomb(buf:bytes|bytearray, bit_position: int) -> tuple[int, int]:
     bit_position, v = read_unsigned_exp_golomb(buf, bit_position)
     match v%2:
         case 0:
@@ -173,7 +173,7 @@ def write_signed_exp_golomb(buf:bytearray, bit_position: int, v) -> int:
     return bit_position
 
 @typechecked
-def parse_rbsp_trailing_bits(buf:bytes, bit_position: int) -> int:
+def parse_rbsp_trailing_bits(buf:bytes|bytearray, bit_position: int) -> int:
     bit_position, one = read_bit(buf, bit_position)
     if one==0:
         raise ValueError(f'Stop bit should be equal to one. Read: {one:d}')
@@ -193,7 +193,7 @@ def write_rbsp_trailing_bits(buf:bytearray, bit_position: int) -> int:
     return bit_position
 
 @typechecked
-def more_rbsp_data(buf:bytes, bit_position: int) -> bool:
+def more_rbsp_data(buf:bytes|bytearray, bit_position: int) -> bool:
     logger.debug('Is there more data in buffer of length: %d at bit position: %d',
                  len(buf), bit_position)
 
@@ -250,7 +250,7 @@ def sodb_to_rbsp(buf:bytes) -> bytes:
 
 # Useful for SPS and PPS
 @typechecked
-def parse_scaling_list(buf:bytes, bit_position: int, size) -> tuple[int,list[int]]:
+def parse_scaling_list(buf:bytes|bytearray, bit_position: int, size) -> tuple[int,list[int]]:
     res = []
     last_scale = 8
     next_scale = 8

src/tscut/pipeline.py

@@ -8,6 +8,7 @@ from os import unlink
 from shutil import copyfile, move
 from sys import exit
 import logging
+from typing import IO, Any
 
 import hexdump
 
@@ -55,7 +56,7 @@ logger = logging.getLogger(__name__)
 
 def process_recording(options: ProcessingOptions) -> None:
     nb_parts = len(options.parts)
-    temporaries = []
+    temporaries : list[IO[Any]] = []
 
     basename = os.path.splitext(os.path.basename(options.input_file))[0]
     mp4filename = basename+'.mp4'
@@ -460,7 +461,7 @@ def process_recording(options: ProcessingOptions) -> None:
             logger.info('Find subtitles tracks and language.')
             subtitles = find_subtitles_tracks(options.tools_paths['ffprobe'], final_with_video)
             logger.info(subtitles)
-            sts = {}
+            sts: dict[str, list[int]] = {}
             for subtitle in subtitles:
                 index = subtitle['index']
                 if 'tags' in subtitle:

src/tscut/subtitles/ocr.py

@@ -18,6 +18,7 @@ from os import (
 import re
 import os
 from subprocess import PIPE, Popen
+from typing import IO, Any
 
 from iso639 import Lang
 from iso639.exceptions import InvalidLanguageValue
@@ -29,7 +30,7 @@ from tscut.exceptions import ExternalToolError
 logger = logging.getLogger(__name__)
 
 @typechecked
-def get_tesseract_supported_lang(tesseract_path:str) -> dict[Lang, str]|None:
+def get_tesseract_supported_lang(tesseract_path:str) -> dict[Lang, str]:
     """
     Retrieves the set of natural languages supported by the Tesseract OCR tool.
 
@@ -48,6 +49,7 @@ def get_tesseract_supported_lang(tesseract_path:str) -> dict[Lang, str]|None:
     res = {}
 
     with Popen([tesseract_path, '--list-langs'], stdout=PIPE) as tesseract:
+        assert tesseract.stdout is not None
         for line in tesseract.stdout:
             line = line.decode('utf8')
             p = re.compile('(?P<lang>[a-z]{3})\n')
@@ -63,8 +65,7 @@ def get_tesseract_supported_lang(tesseract_path:str) -> dict[Lang, str]|None:
     tesseract.wait()
 
     if tesseract.returncode != 0:
-        logger.error("Tesseract returns an error code: %d",tesseract.returncode)
+        raise ExternalToolError("Tesseract returns an error code: %d",tesseract.returncode)
-        return None
 
     return res
 
@@ -100,6 +101,7 @@ def extract_srt(mkvextract:str, filename:str, subtitles:dict[str, list[int]],
 
     env = {**os.environ, 'LANG': 'C'}
     with Popen(params, stdout=PIPE, close_fds=False, env=env) as extract:
+        assert extract.stdout is not None
         pb = tqdm(TextIOWrapper(extract.stdout, encoding="utf-8"), total=100, unit='%',
                   desc='Extraction:')
         for line in pb:
@@ -108,7 +110,8 @@ def extract_srt(mkvextract:str, filename:str, subtitles:dict[str, list[int]],
                 m = p.match(line)
                 if m is None:
                     logger.error('Impossible to parse progress')
-                pb.update(int(m['progress'])-pb.n)
+                else:
+                    pb.update(int(m['progress'])-pb.n)
         pb.update(100-pb.n)
         pb.refresh()
         pb.close()
@@ -128,7 +131,7 @@ def extract_srt(mkvextract:str, filename:str, subtitles:dict[str, list[int]],
 
 @typechecked
 def do_ocr(vobsubocr:str, idxs: Sequence[tuple[str,str,str,str]], duration:timedelta,
-           temporaries:list[IO[bytes]], dump_mem_fd:bool=False):
+           temporaries:list[IO[Any]], dump_mem_fd:bool=False):
     res = []
 
     for idx_name, _, lang, iso in idxs:
@@ -141,6 +144,7 @@ def do_ocr(vobsubocr:str, idxs: Sequence[tuple[str,str,str,str]], duration:timed
 
         srtfd = memfd_create(srtname, flags=0)
         with Popen([vobsubocr, '--lang', iso, idx_name], stdout=PIPE) as ocr:
+            assert ocr.stdout is not None
             pb = tqdm(TextIOWrapper(ocr.stdout, encoding="utf-8"), total=
                       int(duration/timedelta(seconds=1)), unit='s', desc='OCR')
             for line in pb:
@@ -153,7 +157,7 @@ def do_ocr(vobsubocr:str, idxs: Sequence[tuple[str,str,str,str]], duration:timed
                 m = re.match(timestamps, line)
                 if m is not None:
                     hours = int(m.group('hours'))
-                    minutes = int(m.group('hours'))
+                    minutes = int(m.group('minutes'))
                     seconds = int(m.group('seconds'))
                     ts = timedelta(hours=hours, minutes=minutes, seconds=seconds)
                     pb.n = int(ts/timedelta(seconds=1))
@@ -174,7 +178,7 @@ def do_ocr(vobsubocr:str, idxs: Sequence[tuple[str,str,str,str]], duration:timed
                     pos = 0
                     while pos < srt_length:
                         pos+=write(outfd, buf[pos:])
-                        temporaries.append(dump_srt)
+                temporaries.append(dump_srt)
             except OSError:
                 logger.error('Impossible to create file: %s', srtname)
                 return None