Improve `--download-sections`

[yt-dlp.git] / yt_dlp / __init__.py

diff --git a/yt_dlp/__init__.py b/yt_dlp/__init__.py
index 46edd88d3e8a74ae6d80186f4e93b04ae904d957..b81277a5723450092ac4ab1981f94898a4174ecf 100644 (file)
--- a/yt_dlp/__init__.py
+++ b/yt_dlp/__init__.py
@@ -320,26 +320,49 @@ def validate_outtmpl(tmpl, msg):
         opts.skip_download = None
         del opts.outtmpl['default']
 
-    def parse_chapters(name, value):
-        chapters, ranges = [], []
+    def parse_chapters(name, value, advanced=False):
         parse_timestamp = lambda x: float('inf') if x in ('inf', 'infinite') else parse_duration(x)
+        TIMESTAMP_RE = r'''(?x)(?:
+            (?P<start_sign>-?)(?P<start>[^-]+)
+        )?\s*-\s*(?:
+            (?P<end_sign>-?)(?P<end>[^-]+)
+        )?'''
+
+        chapters, ranges, from_url = [], [], False
         for regex in value or []:
-            if regex.startswith('*'):
-                for range_ in map(str.strip, regex[1:].split(',')):
-                    mobj = range_ != '-' and re.fullmatch(r'([^-]+)?\s*-\s*([^-]+)?', range_)
-                    dur = mobj and (parse_timestamp(mobj.group(1) or '0'), parse_timestamp(mobj.group(2) or 'inf'))
-                    if None in (dur or [None]):
-                        raise ValueError(f'invalid {name} time range "{regex}". Must be of the form "*start-end"')
-                    ranges.append(dur)
+            if advanced and regex == '*from-url':
+                from_url = True
+                continue
+            elif not regex.startswith('*'):
+                try:
+                    chapters.append(re.compile(regex))
+                except re.error as err:
+                    raise ValueError(f'invalid {name} regex "{regex}" - {err}')
                 continue
-            try:
-                chapters.append(re.compile(regex))
-            except re.error as err:
-                raise ValueError(f'invalid {name} regex "{regex}" - {err}')
-        return chapters, ranges
 
-    opts.remove_chapters, opts.remove_ranges = parse_chapters('--remove-chapters', opts.remove_chapters)
-    opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges))
+            for range_ in map(str.strip, regex[1:].split(',')):
+                mobj = range_ != '-' and re.fullmatch(TIMESTAMP_RE, range_)
+                dur = mobj and [parse_timestamp(mobj.group('start') or '0'), parse_timestamp(mobj.group('end') or 'inf')]
+                signs = mobj and (mobj.group('start_sign'), mobj.group('end_sign'))
+
+                err = None
+                if None in (dur or [None]):
+                    err = 'Must be of the form "*start-end"'
+                elif not advanced and any(signs):
+                    err = 'Negative timestamps are not allowed'
+                else:
+                    dur[0] *= -1 if signs[0] else 1
+                    dur[1] *= -1 if signs[1] else 1
+                    if dur[1] == float('-inf'):
+                        err = '"-inf" is not a valid end'
+                if err:
+                    raise ValueError(f'invalid {name} time range "{regex}". {err}')
+                ranges.append(dur)
+
+        return chapters, ranges, from_url
+
+    opts.remove_chapters, opts.remove_ranges, _ = parse_chapters('--remove-chapters', opts.remove_chapters)
+    opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges, True))
 
     # Cookies from browser
     if opts.cookiesfrombrowser: