diff --git a/README.md b/README.md index 35229f728..efb490ab1 100644 --- a/README.md +++ b/README.md @@ -1246,7 +1246,7 @@ The field names themselves (the part inside the parenthesis) can also have some 1. **Alternatives**: Alternate fields can be specified separated with a `,`. E.g. `%(release_date>%Y,upload_date>%Y|Unknown)s` -1. **Replacement**: A replacement value can be specified using a `&` separator. If the field is *not* empty, this replacement value will be used instead of the actual field content. This is done after alternate fields are considered; thus the replacement is used if *any* of the alternative fields is *not* empty. +1. **Replacement**: A replacement value can be specified using a `&` separator according to the [`str.format` mini-language](https://docs.python.org/3/library/string.html#format-specification-mini-language). If the field is *not* empty, this replacement value will be used instead of the actual field content. This is done after alternate fields are considered; thus the replacement is used if *any* of the alternative fields is *not* empty. E.g. `%(chapters&has chapters|no chapters)s`, `%(title&TITLE={:>20}|NO TITLE)s` 1. **Default**: A literal default value can be specified for when the field is empty using a `|` separator. This overrides `--output-na-placeholder`. E.g. `%(uploader|Unknown)s` diff --git a/test/test_YoutubeDL.py b/test/test_YoutubeDL.py index 49ae9e2b1..3c26bd7c6 100644 --- a/test/test_YoutubeDL.py +++ b/test/test_YoutubeDL.py @@ -822,7 +822,10 @@ class TestYoutubeDL(unittest.TestCase): test('%(title&foo|baz)s.bar', 'baz.bar') test('%(x,id&foo|baz)s.bar', 'foo.bar') test('%(x,title&foo|baz)s.bar', 'baz.bar') - test('%(title&\n|)s', '\n') + test('%(id&a\nb|)s', ('a\nb', 'a b')) + test('%(id&hi {:>10} {}|)s', 'hi 1234 1234') + test(R'%(id&{0} {}|)s', 'NA') + test(R'%(id&{0.1}|)s', 'NA') # Laziness def gen(): diff --git a/yt_dlp/YoutubeDL.py b/yt_dlp/YoutubeDL.py index 61c149e47..dce6cf928 100644 --- a/yt_dlp/YoutubeDL.py +++ b/yt_dlp/YoutubeDL.py @@ -21,7 +21,7 @@ import tokenize import traceback import unicodedata import urllib.request -from string import ascii_letters +from string import Formatter, ascii_letters from .cache import Cache from .compat import compat_os_name, compat_shlex_quote @@ -1237,6 +1237,14 @@ class YoutubeDL: return list(obj) return repr(obj) + class _ReplacementFormatter(Formatter): + def get_field(self, field_name, args, kwargs): + if field_name.isdigit(): + return args[0], -1 + raise ValueError('Unsupported field') + + replacement_formatter = _ReplacementFormatter() + def create_key(outer_mobj): if not outer_mobj.group('has_key'): return outer_mobj.group(0) @@ -1258,7 +1266,13 @@ class YoutubeDL: if fmt == 's' and value is not None and key in field_size_compat_map.keys(): fmt = f'0{field_size_compat_map[key]:d}d' - value = default if value is None else value if replacement is None else replacement + if value is None: + value = default + elif replacement is not None: + try: + value = replacement_formatter.format(replacement, value) + except ValueError: + value = na flags = outer_mobj.group('conversion') or '' str_fmt = f'{fmt[:-1]}s'