prepare for the new release (#69)

* add the manual shift mode for the subaligner cli

* support alignment on a single video and multiple subtitles

.github/workflows/ci-pipeline.yml

@@ -10,7 +10,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: [3.7]
+        python-version: [3.8]
 
     steps:
       - uses: actions/checkout@v2
@@ -31,6 +31,9 @@ jobs:
           cat requirements-dev.txt | xargs -L 1 pip install
           cat requirements-site.txt | xargs -L 1 pip install
           pip install -e . --ignore-installed
+      - name: Type checking
+        run: |
+          python -m mypy --follow-imports=skip subaligner
       - name: Linting
         run: |
           pycodestyle subaligner tests examples misc bin/subaligner bin/subaligner_1pass bin/subaligner_2pass bin/subaligner_batch bin/subaligner_convert bin/subaligner_train  bin/subaligner_tune setup.py --ignore=E203,E501,W503 --exclude="subaligner/lib"

Pipfile

@@ -36,7 +36,7 @@ click = "==5.1"
 cloudpickle = "==0.5.3"
 cycler = "==0.10.0"
 Cython = "~=0.29.22"
-dask = "==0.15.0"
+dask = ">=2021.10.0"
 decorator = "==4.3.0"
 distributed = "==1.13.0"
 filelock = "==3.0.12"

README.md

@@ -123,6 +123,12 @@ $ subaligner -m script -v test.mp4 -s subtitle.txt -o subtitle_aligned.srt
 $ subaligner -m script -v https://example.com/video.mp4 -s https://example.com/subtitle.txt -o subtitle_aligned.srt
 ```
 ```
+# Alignment on multiple subtitles against the single media file
+
+$ subaligner -m script -v test.mp4 -s subtitle_lang_1.txt -s subtitle_lang_2.txt
+$ subaligner -m script -v test.mp4 -s subtitle_lang_1.txt subtitle_lang_2.txt
+```
+```
 # Translative alignment with the ISO 639-3 language code pair (src,tgt)
 
 $ subaligner_1pass --languages
@@ -135,6 +141,12 @@ $ subaligner -m dual -v video.mp4 -s subtitle.srt -t src,tgt
 $ subaligner -m script -v test.mp4 -s subtitle.txt -o subtitle_aligned.srt -t src,tgt
 ```
 ```
+# Shift subtitle manually by offset in seconds
+
+$ subaligner -m shift --subtitle_path subtitle.srt -os 5.5
+$ subaligner -m shift --subtitle_path subtitle.srt -os -5.5 -o subtitle_shifted.srt
+```
+```
 # Run batch alignment against directories
 
 $ subaligner_batch -m single -vd videos/ -sd subtitles/ -od aligned_subtitles/

mypy.ini

@@ -0,0 +1,8 @@
+# Global options:
+
+[mypy]
+ignore_missing_imports = True
+no_implicit_optional = True
+allow_redefinition = True
+
+# Per-module options:

requirements-dev.txt

@@ -8,7 +8,9 @@ line-profiler==3.1.0
 scikit-build==0.11.1
 radish-bdd~=0.13.3
 pex==2.1.34
-mypy==0.910
+mypy==0.931
+types-requests==2.27.9
+types-setuptools==57.4.9
 parameterized==0.8.1
 pylint~=2.8.2
 pygments==2.7.4

site/source/usage.rst

@@ -31,6 +31,11 @@ Make sure you have got the virtual environment activated upfront.
     (.venv) $ subaligner -m script -v test.mp4 -s subtitle.txt -o subtitle_aligned.srt
     (.venv) $ subaligner -m script -v https://example.com/video.mp4 -s https://example.com/subtitle.txt -o subtitle_aligned.srt
 
+**Alignment on multiple subtitles against the single media file**::
+
+    (.venv) $ subaligner -m script -v test.mp4 -s subtitle_lang_1.txt -s subtitle_lang_2.txt
+    (.venv) $ subaligner -m script -v test.mp4 -s subtitle_lang_1.txt subtitle_lang_2.txt
+
 **Translative alignment with the ISO 639-3 language code pair (src,tgt)**::
 
     (.venv) $ subaligner_1pass --languages
@@ -42,6 +47,11 @@ Make sure you have got the virtual environment activated upfront.
     (.venv) $ subaligner -m dual -v video.mp4 -s subtitle.srt -t src,tgt
     (.venv) $ subaligner -m script -v test.mp4 -s subtitle.txt -o subtitle_aligned.srt -t src,tgt
 
+**Shift subtitle manually by offset in seconds**::
+
+    (.venv) $ subaligner -m shift --subtitle_path subtitle.srt -os 5.5
+    (.venv) $ subaligner -m shift --subtitle_path subtitle.srt -os -5.5 -o subtitle_shifted.srt
+
 **Run batch alignment against directories**::
 
     (.venv) $ subaligner_batch -m single -vd videos/ -sd subtitles/ -od aligned_subtitles/

subaligner/lib/to_srt.py

@@ -598,32 +598,3 @@ def parseChildTree(element_list):
 
     def __iter__(self):
         return iter(self.subs)
-
-if __name__ == '__main__':
-
-    from optparse import OptionParser
-    import sys
-
-    parser = OptionParser(usage = 'usage: %prog [options] input output')
-    parser.set_defaults(reader_class=STL)
-    parser.add_option('-d', '--debug', dest='debug_level', action='store_const', const=logging.DEBUG, default=logging.ERROR)
-    parser.add_option('-r', '--rich', dest='rich_formatting', action='store_true', default=False, help='Output text with some formatting, the following HTML tags are used: b i u font(color)')
-    parser.add_option("-s", "--stl", dest="reader_class", action="store_const", const=STL,
-                            help="Set input file format as STL (default)")
-    parser.add_option("-t", "--tt", dest="reader_class", action="store_const", const=TT,
-                            help="Set input file format as TT, handles both the EBU and SMPTE variants")
-
-    (options, args) = parser.parse_args()
-
-    if len(args) != 2:
-        parser.print_help()
-        sys.exit(1)
-
-    logging.basicConfig(level=options.debug_level)
-
-    input = options.reader_class(args[0], options.rich_formatting)
-    c = SRT(args[1])
-    for sub in input:
-        (tci, tco, txt) = sub
-        c.write(tci, tco, txt)
-    c.file.close()

subaligner/logger.py

@@ -6,7 +6,7 @@
 absl_logging._warn_preinit_stderr = 0
 
 
-class Logger(Singleton):
+class Logger(metaclass=Singleton):
     """Common logging."""
 
     VERBOSE = False

subaligner/predictor.py

@@ -8,7 +8,7 @@
 import logging
 import numpy as np
 import multiprocessing as mp
-from typing import Tuple, List, Optional, Dict, Any
+from typing import Tuple, List, Optional, Dict, Any, Iterable
 from pysrt import SubRipTime, SubRipItem, SubRipFile
 from sklearn.metrics import log_loss
 from copy import deepcopy
@@ -23,7 +23,7 @@
 from .logger import Logger
 
 
-class Predictor(Singleton):
+class Predictor(metaclass=Singleton):
     """ Predictor for working out the time to shift subtitles
     """
     __MAX_SHIFT_IN_SECS = (
@@ -456,8 +456,8 @@ def _predict_in_multithreads(
                 os.remove(segment_path)
 
     @staticmethod
-    def __minibatch(total, batch_size):
-        batch = []
+    def __minibatch(total: int, batch_size: int) -> Iterable[List[int]]:
+        batch: List = []
         for i in range(total):
             if len(batch) == batch_size:
                 yield batch
@@ -708,8 +708,8 @@ def __predict(
             subtitles: Optional[SubRipFile] = None,
             max_shift_secs: Optional[float] = None,
             previous_gap: Optional[float] = None,
-            lock: threading.RLock = None,
-            network: Network = None
+            lock: Optional[threading.RLock] = None,
+            network: Optional[Network] = None
     ) -> Tuple[List[SubRipItem], str, "np.ndarray[float]"]:
         """Shift out-of-sync subtitle cues by sending the audio track of an video to the trained network.
 

subaligner/singleton.py

@@ -1,18 +1,14 @@
 from typing import Dict, Any
 
 
-class _Singleton(type):  # type: ignore
+class Singleton(type):  # type: ignore
     """ A metaclass that creates a Singleton base class when called. """
 
     _instances: Dict[Any, Any] = {}
 
     def __call__(cls, *args, **kwargs) -> Any:
         if cls not in cls._instances:
-            cls._instances[cls] = super(_Singleton, cls).__call__(
+            cls._instances[cls] = super(Singleton, cls).__call__(
                 *args, **kwargs
             )
         return cls._instances[cls]
-
-
-class Singleton(_Singleton("SingletonMeta", (object,), {})):  # type: ignore
-    pass

subaligner/subaligner_batch/__main__.py

@@ -4,7 +4,7 @@
                         [-sil {afr,amh,ara,arg,asm,aze,ben,bos,bul,cat,ces,cmn,cym,dan,deu,ell,eng,epo,est,eus,fas,fin,fra,gla,gle,glg,grc,grn,guj,heb,hin,hrv,hun,hye,ina,ind,isl,ita,jbo,jpn,kal,kan,kat,kir,kor,kur,lat,lav,lfn,lit,mal,mar,mkd,mlt,msa,mya,nah,nep,nld,nor,ori,orm,pan,pap,pol,por,ron,rus,sin,slk,slv,spa,sqi,srp,swa,swe,tam,tat,tel,tha,tsn,tur,ukr,urd,vie,yue,zho}]
                         [-fos] [-tod TRAINING_OUTPUT_DIRECTORY] [-od OUTPUT_DIRECTORY] [-t TRANSLATE] [-lgs] [-d] [-q] [-ver]
 
-Batch align multiple subtitle files and audiovisual files (v0.1.4)
+Batch align multiple subtitle files and audiovisual files
 
 Subtitle files and their companion audiovisual files need to be stored in two separate directories.
 Each file pair needs to share the same base filename, the part before the extension.

subaligner/subtitle.py

@@ -324,6 +324,10 @@ def shift_subtitle(
             string -- The path to the shifted subtitle file.
         """
         _, file_extension = os.path.splitext(subtitle_file_path)
+        if shifted_subtitle_file_path is None:
+            shifted_subtitle_file_path = subtitle_file_path.replace(
+                file_extension, "{}{}".format(suffix, file_extension)
+            )
         if file_extension.lower() in cls.TTML_EXTENSIONS:
             subs = cls(cls.__secret, subtitle_file_path, "ttml").subs
             subs.shift(seconds=seconds)
@@ -333,10 +337,6 @@ def shift_subtitle(
             for index, cue in enumerate(cues):
                 cue.attrib["begin"] = str(subs[index].start).replace(",", ".")
                 cue.attrib["end"] = str(subs[index].end).replace(",", ".")
-            if shifted_subtitle_file_path is None:
-                shifted_subtitle_file_path = subtitle_file_path.replace(
-                    file_extension, "{}{}".format(suffix, file_extension)
-                )
             encoding = Utils.detect_encoding(subtitle_file_path)
             tree.write(shifted_subtitle_file_path, encoding=encoding)
         elif file_extension.lower() in cls.STL_EXTENSIONS:
@@ -741,7 +741,7 @@ def __save_subtitle_by_extension(file_extension: str,
                 # Change single quotes in the XML header to double quotes
                 with open(target_file_path, "w", encoding=encoding) as target:
                     if "xml_declaration" in inspect.getfullargspec(ElementTree.tostring).kwonlyargs:  # for >= python 3.8
-                        encoded = ElementTree.tostring(tt, encoding=encoding, method="xml", xml_declaration=True)
+                        encoded = ElementTree.tostring(tt, encoding=encoding, method="xml", xml_declaration=True)  # type: ignore
                     else:
                         encoded = ElementTree.tostring(tt, encoding=encoding, method="xml")
                     normalised = encoded.decode(encoding) \

subaligner/translator.py

@@ -10,7 +10,7 @@
 from .logger import Logger
 
 
-class Translator(Singleton):
+class Translator(metaclass=Singleton):
     """Translate subtitles.
     """
 

tests/integration/feature/subaligner.feature

@@ -175,7 +175,7 @@ Feature: Subaligner CLI
         |  subaligner_2pass |  <NULL>   |
         |  subaligner       |  dual     |
 
-    @quality-management
+    @quality-control
     Scenario Outline: Test exit when alignment log loss is too high
         Given I have a video file "test.mp4"
         And I have a subtitle file "test.srt"
@@ -302,3 +302,25 @@ Feature: Subaligner CLI
         |  subaligner_1pass |
         |  subaligner_2pass |
         |  subaligner       |
+
+    @manual_shift
+    Scenario Outline: Shift the subtitle by offset in seconds
+        Given I have a subtitle file <subtitle-in>
+        When I run the manual shift with offset of <offset> in seconds
+        Then a new subtitle file <subtitle-out> is generated
+    Examples:
+        |   subtitle-in |   subtitle-out             |  offset  |
+        |   "test.srt"  |   "test_shifted.srt"       |  1.1     |
+        |   "test.ttml" |   "test_shifted.ttml"      |  2.2     |
+        |   "test.xml"  |   "test_shifted.xml"       |  3       |
+        |   "test.dfxp" |   "test_shifted.dfxp"      |  4.25    |
+        |   "test.vtt"  |   "test_shifted.vtt"       |  +0      |
+        |   "test.sami" |   "test_shifted.sami"      |  0       |
+        |   "test.ssa"  |   "test_shifted.ssa"       |  -0      |
+        |   "test.ass"  |   "test_shifted.ass"       |  -1.1    |
+        |   "test.sub"  |   "test_shifted.sub"       |  -2.2    |
+        |   "test.tmp"  |   "test_shifted.tmp"       |  -3      |
+        |   "test.smi"  |   "test_shifted.smi"       |  -4.25   |
+        |   "test.scc"  |   "test_shifted.scc"       |  1.1     |
+        |   "test.sbv"  |   "test_shifted.sbv"       |  2.2     |
+        |   "test.ytt"  |   "test_shifted.ytt"       |  3       |

tests/integration/radish/step.py

@@ -26,6 +26,29 @@ def subtitle_file(step, file_name):
         step.context.subtitle_path_or_selector = os.path.join(PWD, "..", "..", "subaligner", "resource", file_name).replace("[]", " ")
 
 
+@given('I have a list of subtitle files "{file_names:S}"')
+def subtitle_file_list(step, file_names):
+    step.context.subtitle_path_or_selector = [os.path.join(PWD, "..", "..", "subaligner", "resource", file_name).replace("[]", " ") for file_name in file_names.split(",")]
+
+
+@when('I run the alignment with subaligner on all of them')
+def run_subaligner_on_multi_subtitles(step):
+    process = subprocess.Popen([
+        os.path.join(PWD, "..", "..", "..", "bin", "subaligner"),
+        "-m", "single",
+        "-v", step.context.video_file_path,
+        "-q"] + [["-s", path] for path in step.context.subtitle_path_or_selector], shell=False)
+    step.context.exit_code = process.wait(timeout=WAIT_TIMEOUT_IN_SECONDS)
+
+
+@then('a list of subtitle files "{file_names:S}" are generated')
+def expect_result_list(step, file_names):
+    for file_name in file_names.split(","):
+        output_file_path = os.path.join(step.context.aligning_output, file_name)
+        assert os.path.isfile(output_file_path) is True
+    assert step.context.exit_code == 0
+
+
 @given('I have selector "{selector:S}" for the embedded subtitle')
 def subtitle_selector(step, selector):
     step.context.subtitle_path_or_selector = selector
@@ -49,6 +72,17 @@ def run_subaligner(step, aligner, mode):
     step.context.exit_code = process.wait(timeout=WAIT_TIMEOUT_IN_SECONDS)
 
 
+@when("I run the manual shift with offset of {offset_seconds:g} in seconds")
+def run_subaligner_manual_shift(step, offset_seconds):
+    process = subprocess.Popen([
+        os.path.join(PWD, "..", "..", "..", "bin", "subaligner"),
+        "-m", "shift",
+        "-s", step.context.subtitle_path_or_selector,
+        "-os", str(offset_seconds),
+        "-q"], shell=False)
+    step.context.exit_code = process.wait(timeout=WAIT_TIMEOUT_IN_SECONDS)
+
+
 @when("I run the alignment with {aligner:S} on them with {mode:S} stage and {language_pair:S} for translation")
 def run_subaligner_with_translation(step, aligner, mode, language_pair):
     if mode == "<NULL>":

tests/subaligner/test_singleton.py

@@ -4,7 +4,7 @@
 
 class SingletonTests(unittest.TestCase):
     def test_singleton(self):
-        class Single(Singleton):
+        class Single(metaclass=Singleton):
             pass
 
         a = Single()