fix(sample): stop a malformed sidx from hanging the download

pymp4 loops forever on a sidx that counts more references than it holds. Manifest errors in download threads now raise instead of sys.exit.

tests/test_sample.py

@@ -345,6 +345,12 @@ def test_parse_sidx_refuses_what_it_cannot_cut() -> None:
     assert parse_sidx(sidx_box(count=3)[:-4], 0) == (0, [])
 
 
+def test_parse_sidx_refuses_a_box_that_counts_more_references_than_it_holds() -> None:
+    box = bytearray(sidx_box(count=3))
+    box[30:32] = (5).to_bytes(2, "big")
+    assert parse_sidx(bytes(box) + bytes(100), 0) == (0, [])
+
+
 def test_a_single_file_dash_track_is_cut_by_its_segment_index(monkeypatch: pytest.MonkeyPatch) -> None:
     box = sidx_box()
     requested: list[str] = []

unshackle/core/manifests/dash.py

@@ -6,7 +6,6 @@ import logging
 import math
 import re
 import shutil
-import sys
 from concurrent.futures import Future
 from contextlib import ExitStack, closing
 from copy import deepcopy
@@ -17,6 +16,7 @@ from urllib.parse import urljoin, urlparse
 from uuid import UUID
 from zlib import crc32
 
+import click
 import requests
 from langcodes import Language, tag_is_valid
 from lxml.etree import Element, ElementTree, XMLSyntaxError
@@ -497,14 +497,13 @@ class DASH:
             if config.sample and segments:
                 break
 
+        # raise, never sys.exit: this runs in a download thread, where SystemExit escapes the failure handler
         if not segments:
-            log.error("Could not find a way to get segments from this MPD manifest.")
             log.debug(track.url)
-            sys.exit(1)
+            raise click.ClickException("Could not find a way to get segments from this MPD manifest.")
 
         if config.sample and not segment_durations and not isinstance(track, Subtitle):
-            log.error("--sample cannot cut this track: the manifest gives it as one file with no segment index.")
-            sys.exit(1)
+            raise click.ClickException("--sample cannot cut this track: the server sent no segment index for it.")
 
         # TODO: Should we floor/ceil/round, or is int() ok?
         track.data["dash"]["timescale"] = int(segment_timescale)

unshackle/core/manifests/hls.py

@@ -8,7 +8,6 @@ import os
 import re
 import shutil
 import subprocess
-import sys
 import time
 from contextlib import ExitStack, closing
 from itertools import accumulate
@@ -18,6 +17,7 @@ from urllib.parse import urljoin, urlparse
 from uuid import UUID
 from zlib import crc32
 
+import click
 import m3u8
 import requests
 from langcodes import Language, tag_is_valid
@@ -644,14 +644,12 @@ class HLS:
             if not isinstance(response, (requests.Response, RnetResponse)):
                 raise TypeError(f"Expected response to be a requests.Response or rnet.Response, not {type(response)}")
             if not response.ok:
-                log.error(f"Failed to request the invariant M3U8 playlist: {response.status_code}")
-                sys.exit(1)
+                raise click.ClickException(f"Failed to request the invariant M3U8 playlist: {response.status_code}")
 
             master = m3u8.loads(utf8_text(response), uri=track.url)
 
         if not master.segments:
-            log.error("Track's HLS playlist has no segments, expecting an invariant M3U8 playlist.")
-            sys.exit(1)
+            raise click.ClickException("Track's HLS playlist has no segments, expecting an invariant M3U8 playlist.")
 
         unwanted_segments = [
             segment for segment in master.segments if callable(track.OnSegmentFilter) and track.OnSegmentFilter(segment)

unshackle/core/tracks/track.py

@@ -360,11 +360,15 @@ def parse_sidx(data: bytes, offset: int) -> tuple[int, list[tuple[int, int, int]
     Returns the timescale and one (start time, first byte, last byte) for each subsegment. The list
     is empty when the box is not a ``sidx`` or when it points at more indexes and not at media.
     """
+    size, header = int.from_bytes(data[:4], "big"), 32 if data[8:9] == b"\x00" else 40
+    # pymp4 loops forever on a box that counts more references than its size holds
+    if data[4:8] != b"sidx" or size < header + 12 * int.from_bytes(data[header - 2 : header], "big"):
+        return 0, []
     try:
         box = Box.parse(data)
     except ConstructError:
         return 0, []
-    if box.type != b"sidx" or any(reference.reference_type != "MEDIA" for reference in box.references):
+    if any(reference.reference_type != "MEDIA" for reference in box.references):
         return 0, []
     start, byte = box.earliest_presentation_time, offset + box.end + box.first_offset
     subsegments: list[tuple[int, int, int]] = []