Fetches HPR's own metadata, audio, and official transcripts live; falls back to local speech recognition (tiny model, matching HPR's own hpr-get-and-transcribe.bash tooling) only when needed, and to correlation against vendored reference clips or sibling episodes otherwise. License-aware, resumable, and never guesses -- flags anything it isn't confident about for manual review rather than producing a plausible-looking wrong cut.
37 lines
1.3 KiB
Python
37 lines
1.3 KiB
Python
import os
|
|
import unittest
|
|
|
|
from hpr_stripper import srt
|
|
|
|
REFERENCES_DIR = os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), "references")
|
|
|
|
|
|
class TestSrtParsing(unittest.TestCase):
|
|
def test_parses_vendored_outro_srt(self):
|
|
segments = srt.parse_file(os.path.join(REFERENCES_DIR, "outro.srt"))
|
|
self.assertEqual(len(segments), 5)
|
|
text0, start0, end0 = segments[0]
|
|
self.assertIn("Hacker Public Radio podcast", text0)
|
|
self.assertEqual(start0, 0.0)
|
|
self.assertAlmostEqual(end0, 5.56, places=2)
|
|
|
|
def test_parses_vendored_intro_srt_template(self):
|
|
segments = srt.parse_file(os.path.join(REFERENCES_DIR, "intro.srt"))
|
|
self.assertEqual(len(segments), 5)
|
|
self.assertEqual(segments[0][0], "REPLACE_LINE_1")
|
|
self.assertAlmostEqual(segments[-1][2], 27.32, places=2)
|
|
|
|
def test_handles_comma_or_dot_decimal_separator(self):
|
|
text = (
|
|
"1\n00:00:01,000 --> 00:00:02,500\nHello there\n\n"
|
|
"2\n00:00:02.500 --> 00:00:04.000\nSecond line\n"
|
|
)
|
|
segments = srt.parse(text)
|
|
self.assertEqual(len(segments), 2)
|
|
self.assertEqual(segments[0], ("Hello there", 1.0, 2.5))
|
|
self.assertEqual(segments[1], ("Second line", 2.5, 4.0))
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|