Mirroring
/
yt-dlp
mirror of https://github.com/yt-dlp/yt-dlp
1
0
Fork 0
Code Issues Releases Wiki Activity

rework

Browse Source
pull/13395/head
doe1080 3 weeks ago
parent 7a37f183d1
commit 7cdc226150
2 changed files with 568 additions and 283 deletions
Show Stats Download Patch File Download Diff File

524
test/test_InfoExtractor.py
Unescape Escape View File

@ -132,17 +132,20 @@ class TestInfoExtractor(unittest.TestCase):
self.assertRaises(RegexNotFoundError, ie._html_search_meta, ('z', 'x'), html, None, fatal=True) self.assertRaises(RegexNotFoundError, ie._html_search_meta, ('z', 'x'), html, None, fatal=True)
def test_search_json_ld_realworld(self): def test_search_json_ld_realworld(self):
_TESTS = [ _TESTS = [(
# https://github.com/ytdl-org/youtube-dl/issues/23306 # https://github.com/ytdl-org/youtube-dl/issues/23306
( r'''
r'''<script type="application/ld+json"> <script type="application/ld+json">
{ {
"@context": "http://schema.org/", "@context": "http://schema.org/",
"@type": "VideoObject", "@type": "VideoObject",
"name": "1 On 1 With Kleio", "name": "1 On 1 With Kleio",
"url": "https://www.eporner.com/hd-porn/xN49A1cT3eB/1-On-1-With-Kleio/", "url": "https://www.eporner.com/hd-porn/xN49A1cT3eB/1-On-1-With-Kleio/",
"duration": "PT0H12M23S", "duration": "PT0H12M23S",
"thumbnailUrl": ["https://static-eu-cdn.eporner.com/thumbs/static4/7/78/780/780814/9_360.jpg", "https://imggen.eporner.com/780814/1920/1080/9.jpg"], "thumbnailUrl": [
"https://static-eu-cdn.eporner.com/thumbs/static4/7/78/780/780814/9_360.jpg",
"https://imggen.eporner.com/780814/1920/1080/9.jpg"
],
"contentUrl": "https://gvideo.eporner.com/xN49A1cT3eB/xN49A1cT3eB.mp4", "contentUrl": "https://gvideo.eporner.com/xN49A1cT3eB/xN49A1cT3eB.mp4",
"embedUrl": "https://www.eporner.com/embed/xN49A1cT3eB/1-On-1-With-Kleio/", "embedUrl": "https://www.eporner.com/embed/xN49A1cT3eB/1-On-1-With-Kleio/",
"image": "https://static-eu-cdn.eporner.com/thumbs/static4/7/78/780/780814/9_360.jpg", "image": "https://static-eu-cdn.eporner.com/thumbs/static4/7/78/780/780814/9_360.jpg",
@ -155,111 +158,140 @@ class TestInfoExtractor(unittest.TestCase):
"uploadDate": "2015-12-05T21:24:35+01:00", "uploadDate": "2015-12-05T21:24:35+01:00",
"interactionStatistic": { "interactionStatistic": {
"@type": "InteractionCounter", "@type": "InteractionCounter",
"interactionType": { "@type": "http://schema.org/WatchAction" }, "interactionType": {
"@type": "http://schema.org/WatchAction"
},
"userInteractionCount": 1120958 "userInteractionCount": 1120958
}, "aggregateRating": { },
"aggregateRating": {
"@type": "AggregateRating", "@type": "AggregateRating",
"ratingValue": "88", "ratingValue": "88",
"ratingCount": "630", "ratingCount": "630",
"bestRating": "100", "bestRating": "100",
"worstRating": "0" "worstRating": "0"
}, "actor": [{ },
"actor": [{
"@type": "Person", "@type": "Person",
"name": "Kleio Valentien", "name": "Kleio Valentien",
"url": "https://www.eporner.com/pornstar/kleio-valentien/" "url": "https://www.eporner.com/pornstar/kleio-valentien/"
}]} }]
</script>''', }
{ </script>
''', {
'ext': 'mp4',
'title': '1 On 1 With Kleio', 'title': '1 On 1 With Kleio',
'age_limit': 18,
'artists': ['1 On 1 With Kleio'],
'average_rating': 88,
'description': 'Kleio Valentien', 'description': 'Kleio Valentien',
'url': 'https://gvideo.eporner.com/xN49A1cT3eB/xN49A1cT3eB.mp4', 'duration': 743,
'height': 1080,
'thumbnails': [
{'url': 'https://static-eu-cdn.eporner.com/thumbs/static4/7/78/780/780814/9_360.jpg'},
{'url': 'https://imggen.eporner.com/780814/1920/1080/9.jpg'},
],
'timestamp': 1449347075, 'timestamp': 1449347075,
'duration': 743.0, 'url': 'https://gvideo.eporner.com/xN49A1cT3eB/xN49A1cT3eB.mp4',
'view_count': 1120958, 'view_count': 1120958,
'width': 1920, 'width': 1920,
'height': 1080, }, {},
}, ), (
{}, # https://github.com/yt-dlp/yt-dlp/pull/1983
), r'''
( <script type="application/ld+json">
r'''<script type="application/ld+json">
{ {
"@context": "https://schema.org", "@context": "https://schema.org",
"@graph": [ "@graph": [{
{
"@type": "NewsArticle", "@type": "NewsArticle",
"mainEntityOfPage": { "name": "Rewilding Sharks - Inside the lucrative trade of shark fishing in Indonesia",
"@type": "WebPage", "headline": "Rewilding Sharks - Inside the lucrative trade of shark fishing in Indonesia",
"@id": "https://www.ant1news.gr/Society/article/620286/symmoria-anilikon-dikigoros-thymaton-ithelan-na-toys-apoteleiosoyn" "description": "“Arif”, an environmental journalist investigating the shark fishing trade in Surabaya, Indonesia, speaks to industry insiders to understand the business — from the port where fishermen sell a wide variety of sharks, to a drying facility where sharks are processed before exporters pick them up. Shark fishing is l",
}, "about": [
"headline": "Συμμορία ανηλίκων – δικηγόρος θυμάτων: ήθελαν να τους αποτελειώσουν", "leopard sharks",
"name": "Συμμορία ανηλίκων – δικηγόρος θυμάτων: ήθελαν να τους αποτελειώσουν", "Rewilding Sharks",
"description": "Τα παιδιά δέχθηκαν την επίθεση επειδή αρνήθηκαν να γίνουν μέλη της συμμορίας, ανέφερε ο Γ. Ζαχαρόπουλος.", "sustainability",
"animals"
],
"image": { "image": {
"@type": "ImageObject", "@type": "ImageObject",
"url": "https://ant1media.azureedge.net/imgHandler/1100/a635c968-be71-447c-bf9c-80d843ece21e.jpg", "url": "https://dam.mediacorp.sg/image/upload/s--0VYzW7We--/c_fill,g_auto,h_338,w_600/f_auto,q_auto/v1/mediacorp/cna/image/2025/03/17/1742148440-image.jpg?itok=rav-cQ_p",
"width": 1100, "width": "100",
"height": 756 }, "height": "100"
"datePublished": "2021-11-10T08:50:00+03:00", },
"dateModified": "2021-11-10T08:52:53+03:00", "datePublished": "2025-03-17T01:58:00+08:00",
"dateModified": "2025-04-08T12:07:12+08:00",
"author": { "author": {
"@type": "Person", "@type": "Person",
"@id": "https://www.ant1news.gr/", "@id": "https://www.channelnewsasia.com/",
"name": "Ant1news", "name": "CNA",
"image": "https://www.ant1news.gr/images/logo-e5d7e4b3e714c88e8d2eca96130142f6.png", "url": "https://www.channelnewsasia.com/"
"url": "https://www.ant1news.gr/"
}, },
"publisher": { "publisher": {
"@type": "Organization", "@type": "Organization",
"@id": "https://www.ant1news.gr#publisher", "@id": "https://www.channelnewsasia.com/",
"name": "Ant1news", "name": "CNA",
"url": "https://www.ant1news.gr", "url": "https://www.channelnewsasia.com/",
"logo": { "logo": {
"@type": "ImageObject", "@type": "ImageObject",
"url": "https://www.ant1news.gr/images/logo-e5d7e4b3e714c88e8d2eca96130142f6.png", "url": "https://www.channelnewsasia.com/sites/default/themes/mc_cna_theme/images/logo.svg",
"width": 400, "width": "100",
"height": 400 }, "height": "100"
"sameAs": [
"https://www.facebook.com/Ant1news.gr",
"https://twitter.com/antennanews",
"https://www.youtube.com/channel/UC0smvAbfczoN75dP0Hw4Pzw",
"https://www.instagram.com/ant1news/"
]
},
"keywords": "μαχαίρωμα,συμμορία ανηλίκων,ΕΙΔΗΣΕΙΣ,ΕΙΔΗΣΕΙΣ ΣΗΜΕΡΑ,ΝΕΑ,Κοινωνία - Ant1news",
"articleSection": "Κοινωνία"
} }
] },
"mainEntityOfPage": "https://www.channelnewsasia.com/watch/rewilding-sharks/inside-lucrative-trade-shark-fishing-indonesia-5004256"
}, {
"@type": "VideoObject",
"thumbnailUrl": "https://cf-images.ap-southeast-1.prod.boltdns.net/v1/static/6057984932001/b49a7cc0-bbd3-4634-8049-756f0bf3d0c3/3e2f7ea5-0290-4760-889f-b084117a46e8/1280x720/match/image.jpg",
"uploadDate": "2025-04-08T12:07:12+08:00",
"description": "“Arif”, an environmental journalist investigating the shark fishing trade in Surabaya, Indonesia, speaks to industry insiders to understand the business — from the port where fishermen sell a wide variety of sharks, to a drying facility where sharks are processed before exporters pick them up. Shark fishing is legal in Indonesia, and sharks bring in good money. Some species are highly sought after. For example, leopard sharks are prized for their special skin. A fisherman revealed that he gets requests for up to 600kg of leopard shark in a month, worth about 282 million rupiah (US$17,000).",
"name": "Inside the lucrative trade of shark fishing in Indonesia",
"@id": "https://www.channelnewsasia.com/watch/rewilding-sharks/inside-lucrative-trade-shark-fishing-indonesia-5004256",
"duration": "PT472S",
"embedUrl": "https://www.channelnewsasia.com/watch/rewilding-sharks/inside-lucrative-trade-shark-fishing-indonesia-5004256?view=embed",
"contentUrl": "https://www.channelnewsasia.com/watch/rewilding-sharks/inside-lucrative-trade-shark-fishing-indonesia-5004256"
}]
} }
</script>''', </script>
''', {
'title': 'Rewilding Sharks - Inside the lucrative trade of shark fishing in Indonesia',
'creators': ['CNA'],
'description': 'md5:4ce967a72d546b32935cb98c8722346b',
'modified_timestamp': 1744085232,
'release_timestamp': 1742147880,
'thumbnails': [{
'height': 100,
'url': 'https://dam.mediacorp.sg/image/upload/s--0VYzW7We--/c_fill,g_auto,h_338,w_600/f_auto,q_auto/v1/mediacorp/cna/image/2025/03/17/1742148440-image.jpg?itok=rav-cQ_p',
'width': 100,
}],
}, {'expected_type': 'NewsArticle'},
), (
# https://github.com/yt-dlp/yt-dlp/pull/2031
r'''
<script type="application/ld+json">
{ {
'title': 'md5:91fe569e952e4d146485740ae927662b', "url":"/vrtnu/a-z/het-journaal/2021/het-journaal-het-journaal-19u-20211231/",
'categories': ['Κοινωνία'],
'creators': ['Ant1news'],
'description': 'md5:16756d0a18f33bf550e683d134a72f3c',
'modified_timestamp': 1636523573,
'release_timestamp': 1636523400,
'tags': 'count:6',
'thumbnails': [{'url': 'https://ant1media.azureedge.net/imgHandler/1100/a635c968-be71-447c-bf9c-80d843ece21e.jpg'}],
'uploader': 'Ant1news',
},
{'expected_type': 'NewsArticle'},
),
(
r'''<script type="application/ld+json">
{"url":"/vrtnu/a-z/het-journaal/2021/het-journaal-het-journaal-19u-20211231/",
"name":"Het journaal 19u", "name":"Het journaal 19u",
"description":"Het journaal 19u van vrijdag 31 december 2021.", "description":"Het journaal 19u van vrijdag 31 december 2021.",
"potentialAction":{"url":"https://vrtnu.page.link/pfVy6ihgCAJKgHqe8","@type":"ShareAction"}, "potentialAction":{
"mainEntityOfPage":{"@id":"1640092242445","@type":"WebPage"}, "url":"https://vrtnu.page.link/pfVy6ihgCAJKgHqe8",
"@type":"ShareAction"
},
"mainEntityOfPage":{
"@id":"1640092242445",
"@type":"WebPage"
},
"publication":[{ "publication":[{
"startDate":"2021-12-31T19:00:00.000+01:00", "startDate":"2021-12-31T19:00:00.000+01:00",
"endDate":"2022-01-30T23:55:00.000+01:00", "endDate":"2022-01-30T23:55:00.000+01:00",
"publishedBy":{"name":"een","@type":"Organization"}, "publishedBy":{
"publishedOn":{"url":"https://www.vrt.be/vrtnu/","name":"VRT NU","@type":"BroadcastService"}, "name":"een",
"@type":"Organization"
},
"publishedOn":{
"url":"https://www.vrt.be/vrtnu/",
"name":"VRT NU",
"@type":"BroadcastService"
},
"@id":"pbs-pub-3a7ec233-da95-4c1e-9b2b-cf5fdfebcbe8", "@id":"pbs-pub-3a7ec233-da95-4c1e-9b2b-cf5fdfebcbe8",
"@type":"BroadcastEvent" "@type":"BroadcastEvent"
}], }],
@ -287,11 +319,24 @@ class TestInfoExtractor(unittest.TestCase):
}, },
"genre":["Nieuws en actua"], "genre":["Nieuws en actua"],
"episodeNumber":365, "episodeNumber":365,
"partOfSeries":{"name":"Het journaal","@id":"222831405527","@type":"TVSeries"}, "partOfSeries":{
"partOfSeason":{"name":"Seizoen 2021","@id":"961809365527","@type":"TVSeason"}, "name":"Het journaal",
"@context":"https://schema.org","@id":"961685295527","@type":"TVEpisode"}</script> "@id":"222831405527",
''', "@type":"TVSeries"
{ },
"partOfSeason":{
"name":"Seizoen 2021",
"@id":"961809365527",
"@type":"TVSeason"
},
"@context":"https://schema.org",
"@id":"961685295527",
"@type":"TVEpisode"
}
</script>
''', {
'title': 'Het journaal - Aflevering 365 (Seizoen 2021)',
'artists': ['Het journaal - Aflevering 365 (Seizoen 2021)'],
'chapters': [ 'chapters': [
{'title': 'Explosie Turnhout', 'start_time': 70, 'end_time': 440}, {'title': 'Explosie Turnhout', 'start_time': 70, 'end_time': 440},
{'title': 'Jaarwisseling', 'start_time': 440, 'end_time': 1179}, {'title': 'Jaarwisseling', 'start_time': 440, 'end_time': 1179},
@ -304,98 +349,279 @@ class TestInfoExtractor(unittest.TestCase):
{'title': 'Olympische Winterspelen', 'start_time': 1728, 'end_time': 1873}, {'title': 'Olympische Winterspelen', 'start_time': 1728, 'end_time': 1873},
{'title': 'Sober oudjaar in Nederland', 'start_time': 1873, 'end_time': 2079.23}, {'title': 'Sober oudjaar in Nederland', 'start_time': 1873, 'end_time': 2079.23},
], ],
'title': 'Het journaal - Aflevering 365 (Seizoen 2021)', 'description': 'Het journaal 19u van vrijdag 31 december 2021. Bekijk aflevering 365 van seizoen 2021 met VRT NU via de site of app.',
'duration': 2079.23,
'episode': 'Het journaal 19u',
'episode_number': 365,
'genres': ['Nieuws en actua'],
'season': 'Het journaal 19u',
'series': 'Het journaal',
'thumbnails': [{'url': 'https://images.vrt.be/width1280/2021/12/31/80d5ed00-6a64-11ec-b07d-02b7b76bf47f.jpg'}],
'timestamp': 1640973600,
}, {}, }, {},
), ), (
( # thumbnailUrl, {str}
# test multiple thumbnails in a list
r''' r'''
<script type="application/ld+json"> <script type="application/ld+json">
{"@context":"https://schema.org",
"@type":"VideoObject",
"thumbnailUrl":["https://www.rainews.it/cropgd/640x360/dl/img/2021/12/30/1640886376927_GettyImages.jpg"]}
</script>''',
{ {
'thumbnails': [{'url': 'https://www.rainews.it/cropgd/640x360/dl/img/2021/12/30/1640886376927_GettyImages.jpg'}], "@context":"https://schema.org",
},
{},
),
(
# test single thumbnail
r'''
<script type="application/ld+json">
{"@context":"https://schema.org",
"@type":"VideoObject", "@type":"VideoObject",
"thumbnailUrl":"https://www.rainews.it/cropgd/640x360/dl/img/2021/12/30/1640886376927_GettyImages.jpg"} "thumbnailUrl":"https://www.rainews.it/cropgd/640x360/dl/img/2021/12/30/1640886376927_GettyImages.jpg"
</script>''', }
{ </script>
''', {
'thumbnails': [{'url': 'https://www.rainews.it/cropgd/640x360/dl/img/2021/12/30/1640886376927_GettyImages.jpg'}], 'thumbnails': [{'url': 'https://www.rainews.it/cropgd/640x360/dl/img/2021/12/30/1640886376927_GettyImages.jpg'}],
}, }, {},
{}, ), (
), # no scheme URL
( # thumbnail_url, {str}
# test thumbnail_url key without URL scheme
r''' r'''
<script type="application/ld+json"> <script type="application/ld+json">
{ {
"@context": "https://schema.org", "@context": "https://schema.org",
"@type": "VideoObject", "@type": "VideoObject",
"thumbnail_url": "//www.nobelprize.org/images/12693-landscape-medium-gallery.jpg" "thumbnail_url": "//www.nobelprize.org/images/12693-landscape-medium-gallery.jpg"
}</script>''', }
{ </script>
''', {
'thumbnails': [{'url': 'https://www.nobelprize.org/images/12693-landscape-medium-gallery.jpg'}], 'thumbnails': [{'url': 'https://www.nobelprize.org/images/12693-landscape-medium-gallery.jpg'}],
}, }, {},
{}, ), (
), # no scheme URL
( # thumbnailURL, {str}
r'''
<script type="application/ld+json">
{
"@context": "https://schema.org/",
"@type": "VideoObject",
"thumbnailURL": "//images.ctfassets.net/o78em1y1w4i4/2XrBpSdjPK1OJXAAFYU8iw/76ffd0f25465502c9a704dcfc2aa6c64/Teaser-Elsevier-careers-video-thumbnail.jpg"
}
</script>
''', {
'thumbnails': [{'url': 'https://images.ctfassets.net/o78em1y1w4i4/2XrBpSdjPK1OJXAAFYU8iw/76ffd0f25465502c9a704dcfc2aa6c64/Teaser-Elsevier-careers-video-thumbnail.jpg'}],
}, {},
), (
# escaped URL
# image, {"url": {str}}
r''' r'''
<script type="application/ld+json"> <script type="application/ld+json">
{"@context":"https://schema.org", {
"@context": "https:\/\/schema.org",
"@type": "NewsArticle", "@type": "NewsArticle",
"mainEntityOfPage":{ "image": {
"@type":"WebPage", "@context": "https:\/\/schema.org",
"@id":"https://www.telemb.be/actu/frameries-un-concours-pour-conducteurs-dengins-de-chantier/37879" "@type": "ImageObject",
"height": 630,
"width": 1200,
"url": "https:\/\/assets1.cbsnewsstatic.com\/hub\/i\/r\/2025\/08\/29\/597b721d-9c95-424d-8720-05df6b8a4a4e\/thumbnail\/1200x630\/45e846ad5f209972ab225651b40d0b4d\/cbsn-fusion-al-gore-on-20-years-since-hurricane-katrina-thumbnail.jpg"
}
}
</script>
''', {
'thumbnails': [{
'height': 630,
'url': 'https://assets1.cbsnewsstatic.com/hub/i/r/2025/08/29/597b721d-9c95-424d-8720-05df6b8a4a4e/thumbnail/1200x630/45e846ad5f209972ab225651b40d0b4d/cbsn-fusion-al-gore-on-20-years-since-hurricane-katrina-thumbnail.jpg',
'width': 1200,
}],
}, {},
), (
# nested width/height
# image, {"url": {str}}
r'''
<script type="application/ld+json">
{
"@context": "http://schema.org",
"@type": "NewsArticle",
"image": {
"@type": "ImageObject",
"width": {
"@type": "QuantitativeValue",
"unitText": "px",
"value": 1024
}, },
"headline":"Frameries - Un concours pour conducteurs d'engins de chantier", "height": {
"image":[ "@type": "QuantitativeValue",
"//www.telemb.be/cdn/ff/pKwkkhB7a5GqSf98QdDUcn9WlvGTYyilvXisHO3fHpI/1747320854/public/2025-05/00006554_avc-tmb-093031.jpeg" "unitText": "px",
], "value": 576
"articleSection":"Reportages",
"keywords":"enseignement secondaire",
"datePublished":"2025-05-15T16:32:00+02:00",
"dateCreated":"2025-05-15T16:32:00+02:00",
"dateModified":"2025-05-15T16:32:00+02:00",
"author":{
"@type":"Person",
"name":"Sabine Dupont"
}, },
"publisher":{ "url": "https://ichef.bbci.co.uk/ace/standard/1024/cpsprodpb/398e/live/1cc4dab0-8689-11f0-9cf6-cbf3e73ce2b9.jpg"
"@type":"Organization", }
"name":"Tele MB", }
"logo":{ </script>
''', {
'thumbnails': [{
'height': 576,
'url': 'https://ichef.bbci.co.uk/ace/standard/1024/cpsprodpb/398e/live/1cc4dab0-8689-11f0-9cf6-cbf3e73ce2b9.jpg',
'width': 1024,
}],
}, {},
), (
# image, {"url": [{str}]}
r'''
<script type="application/ld+json">
{
"@context": "https://schema.org/",
"@type": [
"NewsArticle",
"Article"
],
"image": {
"@type": "ImageObject", "@type": "ImageObject",
"url":"https://www.telemb.be/modules/custom/local_tvs/modules/tele_mb/assets/logoTLMB_BE_2022.svg" "url": [
"https://imagenes.elpais.com/resizer/v2/JUFK7HEHNNAIVJXC3WVUXNZT7M.jpg?auth=350ef4714331cf2e29299b840f86da4eb8b0ddde3a3bdb3f4302f0c90a9ae2d6&width=1960&height=1103&smart=true",
"https://imagenes.elpais.com/resizer/v2/JUFK7HEHNNAIVJXC3WVUXNZT7M.jpg?auth=350ef4714331cf2e29299b840f86da4eb8b0ddde3a3bdb3f4302f0c90a9ae2d6&width=1960&height=1470&smart=true",
"https://imagenes.elpais.com/resizer/v2/JUFK7HEHNNAIVJXC3WVUXNZT7M.jpg?auth=350ef4714331cf2e29299b840f86da4eb8b0ddde3a3bdb3f4302f0c90a9ae2d6&width=980&height=980&smart=true"
]
} }
}, }
"description":"Conduire des engins de chantier c'est un métier! Ce métier s'apprend dans 7 écoles techniques en Wallonie. 6 d’entre elles se sont retrouvées au centre de compétence du Forem à Sars-la-Bruyère pour une première édition d'un concours baptisé Engin Défi. ", </script>
"articleBody":"Creuser une tranchée, charger un camion, niveler un terrain, quelques-uns des défis proposés aux écoles techniques de Wallonie qui participent à ce premier concours Engin Défi. Par binôme, les 26 élèves sélectionnés doivent réaliser différentes tâches et démontrer ainsi leur savoir-faire en matière de manipulation d’engins de chantier.« On devait charger le gros dumper. Ca s’est bien passé mais je ne sais pas encore ce que le jury en pense » explique Clément Hennuy, Elève en 6ème conducteur d’engins de chantier aux Instituts Saint-Luc.Car Engin Defi est un concours ! A chaque épreuve, des professionnels sont là pour coter individuellement les élèves mais surtout pour les encadrer.« Ils sont évalués par poste de travail. Il y a des cotes de 5 à 20 mais on n’est pas là pour les casser. On est là pour leur donner des conseils avant et pendant l’épreuve » sourit Guy Laidoum, Conseiller en prévention et jury d’Engin Défi.Les professeurs des élèves choisis sont également présents pour les coacher.« Ils sont à l’épreuve de la tranchée qu’ils doivent faire bien plate, bien propre avec des bords bien tranchants. Quelque chose de nickel comme on leur a appris » insiste Loïc Cimino, professeur au Lycée Technique Hornu-Colfontaine. Engin Défi est une première organisation mise sur pied par plusieurs partenaires du secteur de la construction. Tous n’ont qu’un but, valoriser des métiers très recherchés et demandés.« On veut leur montrer que derrière l’école, il y a un secteur d’activités à la recherche de main d’œuvre qualifiée. Ils seront bientôt qualifiés, on veut leur montrer qu’on a besoin d’eux » insiste Xavier Maes, Manager chez Constructiv. Et pour susciter des vocations, une centaine d’élèves de deuxième secondaire viennent aussi découvrir, de manière plus ludique, ces métiers du secteur de la construction. Certains sont plus intéressés que d’autres…« Je ne ferais pas ces métiers là, non. Ce n’est pas dans mes délires ! » sourit Alissa, élève de 2ème secondaire à la Sainte-Union.« Moi j’aime les métiers manuels. Travailler dans la mécanique, c’est ce qui me passionne » souligne Mohamed, un autre élève. Des passionnés, c’est ce que le secteur recherche. Clément en tout cas semble avoir trouvé sa voie.« Je suis en 6ème. Je vais faire une 7ème pour mon diplôme et ma gestion, puis je vais travailler dans une entreprise. Plus tard, j’aimerais créer ma propre entreprise »" ''', {
}</script> 'thumbnails': [{
''', 'height': 1103,
'url': 'https://imagenes.elpais.com/resizer/v2/JUFK7HEHNNAIVJXC3WVUXNZT7M.jpg?auth=350ef4714331cf2e29299b840f86da4eb8b0ddde3a3bdb3f4302f0c90a9ae2d6&width=1960&height=1103&smart=true',
'width': 1960,
}, {
'height': 1470,
'url': 'https://imagenes.elpais.com/resizer/v2/JUFK7HEHNNAIVJXC3WVUXNZT7M.jpg?auth=350ef4714331cf2e29299b840f86da4eb8b0ddde3a3bdb3f4302f0c90a9ae2d6&width=1960&height=1470&smart=true',
'width': 1960,
}, {
'height': 980,
'url': 'https://imagenes.elpais.com/resizer/v2/JUFK7HEHNNAIVJXC3WVUXNZT7M.jpg?auth=350ef4714331cf2e29299b840f86da4eb8b0ddde3a3bdb3f4302f0c90a9ae2d6&width=980&height=980&smart=true',
'width': 980,
}],
}, {},
), (
# image, [{"contentUrl": {str}}]
r'''
<script type="application/ld+json">
{
"@context": "https://schema.org/",
"@type": "VideoObject",
"image": [
{ {
'title': 'md5:3f077843a74f01f768bbf0853c210855', "@type": "ImageObject",
'categories': ['Reportages'], "contentUrl": "https://www.planet-wissen.de/sendungen/lebensglueck-frau-sprung-strand-100~_v-gseagaleriexl.jpg",
'creators': ['Sabine Dupont'], "height": 900,
'description': 'md5:1dc04a3aa56c5228503071baa8b4cc97', "width": 1600
'modified_timestamp': 1747319520,
'release_timestamp': 1747319520,
'tags': 'count:1',
'timestamp': 1747319520,
'thumbnails': [{'url': 'https://www.telemb.be/cdn/ff/pKwkkhB7a5GqSf98QdDUcn9WlvGTYyilvXisHO3fHpI/1747320854/public/2025-05/00006554_avc-tmb-093031.jpeg'}],
'uploader': 'Tele MB',
}, },
{}, {
), "@type": "ImageObject",
"contentUrl": "https://www.planet-wissen.de/sendungen/lebensglueck-frau-sprung-strand-100~_v-HintergrundL.jpg",
"height": 1152,
"width": 1536
}
] ]
}
</script>
''', {
'thumbnails': [{
'height': 900,
'url': 'https://www.planet-wissen.de/sendungen/lebensglueck-frau-sprung-strand-100~_v-gseagaleriexl.jpg',
'width': 1600,
}, {
'height': 1152,
'url': 'https://www.planet-wissen.de/sendungen/lebensglueck-frau-sprung-strand-100~_v-HintergrundL.jpg',
'width': 1536,
}],
}, {},
), (
# duplicate thumbnails
# image, [{"url": {str}}],
# thumbnail, [{"url": {str}}]
# thumbnailUrl, [{str}]
r'''
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "VideoObject",
"image": [{
"@type": "ImageObject",
"url": "https://images.sportschau.de/image/031d1445-9620-4b24-b1f5-9cbbff422f21/AAABmQQq_vA/AAABmKJiLwU/16x9-1920/leverkusen-322.jpg",
"width": 1920,
"height": 1080
}, {
"@type": "ImageObject",
"url": "https://images.sportschau.de/image/031d1445-9620-4b24-b1f5-9cbbff422f21/AAABmQQq_vA/AAABmKJibh0/1x1-1400/leverkusen-322.jpg",
"width": 1400,
"height": 1400
}, {
"@type": "ImageObject",
"url": "https://images.sportschau.de/image/031d1445-9620-4b24-b1f5-9cbbff422f21/AAABmQQq_vA/AAABmKJiwCE/4x3/leverkusen-322.jpg?width=1280",
"width": 1280,
"height": 960
}],
"thumbnail": [{
"@type": "ImageObject",
"url": "https://images.sportschau.de/image/031d1445-9620-4b24-b1f5-9cbbff422f21/AAABmQQq_vA/AAABmKJiLwU/16x9-1920/leverkusen-322.jpg",
"width": 1920,
"height": 1080
}, {
"@type": "ImageObject",
"url": "https://images.sportschau.de/image/031d1445-9620-4b24-b1f5-9cbbff422f21/AAABmQQq_vA/AAABmKJibh0/1x1-1400/leverkusen-322.jpg",
"width": 1400,
"height": 1400
}, {
"@type": "ImageObject",
"url": "https://images.sportschau.de/image/031d1445-9620-4b24-b1f5-9cbbff422f21/AAABmQQq_vA/AAABmKJiwCE/4x3/leverkusen-322.jpg?width=1280",
"width": 1280,
"height": 960
}],
"thumbnailUrl": [
"https://images.sportschau.de/image/031d1445-9620-4b24-b1f5-9cbbff422f21/AAABmQQq_vA/AAABmKJiLwU/16x9-1920/leverkusen-322.jpg",
"https://images.sportschau.de/image/031d1445-9620-4b24-b1f5-9cbbff422f21/AAABmQQq_vA/AAABmKJibh0/1x1-1400/leverkusen-322.jpg",
"https://images.sportschau.de/image/031d1445-9620-4b24-b1f5-9cbbff422f21/AAABmQQq_vA/AAABmKJiwCE/4x3/leverkusen-322.jpg?width=1280"
]
}
</script>
''', {
'thumbnails': [{
'height': 1080,
'url': 'https://images.sportschau.de/image/031d1445-9620-4b24-b1f5-9cbbff422f21/AAABmQQq_vA/AAABmKJiLwU/16x9-1920/leverkusen-322.jpg',
'width': 1920,
}, {
'height': 1400,
'url': 'https://images.sportschau.de/image/031d1445-9620-4b24-b1f5-9cbbff422f21/AAABmQQq_vA/AAABmKJibh0/1x1-1400/leverkusen-322.jpg',
'width': 1400,
}, {
'height': 960,
'url': 'https://images.sportschau.de/image/031d1445-9620-4b24-b1f5-9cbbff422f21/AAABmQQq_vA/AAABmKJiwCE/4x3/leverkusen-322.jpg?width=1280',
'width': 1280,
}],
}, {},
), (
# dateCreated, \d{4}
r'''
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Movie",
"dateCreated": "2025"
}
</script>
''', {'release_year': 2025}, {},
), (
# dateCreated, \d{4}-\d{2}-\d{2}
r'''
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Movie",
"dateCreated": "2025-09-01"
}
</script>
''', {'upload_date': '20250901'}, {},
), (
# dateCreated, ISO 8601
r'''
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Movie",
"dateCreated": "2025-09-01T00:00:00Z"
}
</script>
''', {'timestamp': 1756684800}, {},
)]
for html, expected_dict, search_json_ld_kwargs in _TESTS: for html, expected_dict, search_json_ld_kwargs in _TESTS:
expect_dict( expect_dict(
self, self,

167
yt_dlp/extractor/common.py
Unescape Escape View File

@ -72,11 +72,13 @@ from ..utils import (
mimetype2ext, mimetype2ext,
netrc_from_content, netrc_from_content,
orderedSet, orderedSet,
parse_age_limit,
parse_bitrate, parse_bitrate,
parse_codecs, parse_codecs,
parse_duration, parse_duration,
parse_iso8601, parse_iso8601,
parse_m3u8_attributes, parse_m3u8_attributes,
parse_qs,
parse_resolution, parse_resolution,
qualities, qualities,
sanitize_url, sanitize_url,
@ -84,13 +86,11 @@ from ..utils import (
str_or_none, str_or_none,
str_to_int, str_to_int,
strip_or_none, strip_or_none,
traverse_obj,
truncate_string, truncate_string,
try_call, try_call,
try_get, try_get,
unescapeHTML, unescapeHTML,
unified_strdate, unified_strdate,
unified_timestamp,
url_basename, url_basename,
url_or_none, url_or_none,
urlhandle_detect_ext, urlhandle_detect_ext,
@ -102,6 +102,7 @@ from ..utils import (
) )
from ..utils._utils import _request_dump_filename from ..utils._utils import _request_dump_filename
from ..utils.jslib import devalue from ..utils.jslib import devalue
from ..utils.traversal import traverse_obj
class InfoExtractor: class InfoExtractor:
@ -1673,38 +1674,76 @@ class InfoExtractor:
chapters[-1]['end_time'] = chapters[-1]['end_time'] or info['duration'] chapters[-1]['end_time'] = chapters[-1]['end_time'] or info['duration']
info['chapters'] = chapters info['chapters'] = chapters
def extract_thumbnail_information(e):
thumbnails = traverse_obj(e, ((
'image', 'thumbnail', 'thumbnailUrl', 'thumbnailURL', 'thumbnail_url',
), (
({str}, {url_or_none}, {'url': None}, filter),
({dict}, 'url', {list}, ..., {'url': None}, filter),
(({list}, ({dict}, all)), lambda _, v:
url_or_none(v.get('url')) or url_or_none(v.get('contentUrl'))),
({list}, ..., {str}, {url_or_none}, {'url': None}, filter),
), {
'height': ('height', (None, 'value'), {int_or_none}, any),
'url': (('contentUrl', 'url'), {str}, {unescapeHTML}, {self._proto_relative_url}, any),
'width': ('width', (None, 'value'), {int_or_none}, any),
}, all, {orderedSet}, lambda _, v: url_or_none(v['url'])))
dim_keys, url_table = {'height', 'width'}, {}
for thumbnail in thumbnails:
url = thumbnail['url']
query = parse_qs(thumbnail['url'])
for key, alt_keys in (
('height', ('height', 'h')),
('width', ('width', 'w')),
):
val = traverse_obj(query, (alt_keys, -1, {int_or_none}, any))
if val is not None:
thumbnail.setdefault(key, val)
res = parse_resolution(url_basename(url))
for key in dim_keys:
val = res.get(key)
if val is not None:
thumbnail.setdefault(key, val)
current = url_table.get(url)
if not current or len(dim_keys & thumbnail.keys()) > len(dim_keys & current.keys()):
url_table[url] = thumbnail
info['thumbnails'] = list(url_table.values()) or None
def extract_video_object(e): def extract_video_object(e):
author = e.get('author') info.update(traverse_obj(e, {
info.update({ 'ext': ('encodingFormat', {mimetype2ext}),
'url': url_or_none(e.get('contentUrl')), 'title': ('name', {clean_html}, filter),
'ext': mimetype2ext(e.get('encodingFormat')), 'age_limit': ('isFamilyFriendly', {str}, {lambda x: 18 if x.lower() in ('false', '0') else None}),
'title': unescapeHTML(e.get('name')), 'artists': (('byArtist', 'name'), {clean_html}, filter, all, {orderedSet}, filter),
'description': unescapeHTML(e.get('description')), 'description': ('description', {clean_html}, filter),
'thumbnails': traverse_obj(e, (('thumbnailUrl', 'thumbnailURL', 'thumbnail_url'), (None, ...), { 'duration': ('duration', {parse_duration}),
'url': ({str}, {unescapeHTML}, {self._proto_relative_url}, {url_or_none}), 'filesize': ('contentSize', {float_or_none}, {int_or_none}),
})), 'genres': ('genre', {clean_html}, filter, all, {orderedSet}, filter),
'duration': parse_duration(e.get('duration')), 'height': ('height', {int_or_none}),
'timestamp': unified_timestamp(e.get('uploadDate')), 'is_live': ('publication', 'isLiveBroadcast', {bool}),
# author can be an instance of 'Organization' or 'Person' types. 'release_timestamp': ('datePublished', {parse_iso8601}),
# both types can have 'name' property(inherited from 'Thing' type). [1] 'tags': ('keywords', (None, ...), {clean_html},
# however some websites are using 'Text' type instead. {lambda x: x.split(',')}, ..., {str.strip}, filter, all, {orderedSet}, filter),
# 1. https://schema.org/VideoObject 'tbr': ('bitrate', {int_or_none}),
'uploader': author.get('name') if isinstance(author, dict) else author if isinstance(author, str) else None, 'timestamp': ('uploadDate', {parse_iso8601}),
'artist': traverse_obj(e, ('byArtist', 'name'), expected_type=str), 'uploader': ('author', (None, 'name'), {clean_html}, filter, any),
'filesize': int_or_none(float_or_none(e.get('contentSize'))), 'url': ('contentUrl', {self._proto_relative_url}, {url_or_none}),
'tbr': int_or_none(e.get('bitrate')), 'view_count': ('interactionCount', {int_or_none}),
'width': int_or_none(e.get('width')), 'width': ('width', {int_or_none}),
'height': int_or_none(e.get('height')), }))
'view_count': int_or_none(e.get('interactionCount')),
'tags': try_call(lambda: e.get('keywords').split(',')),
})
if is_type(e, 'AudioObject'): if is_type(e, 'AudioObject'):
info.update({ info.update({
'abr': traverse_obj(e, ('bitrate', {int_or_none})),
'vcodec': 'none', 'vcodec': 'none',
'abr': int_or_none(e.get('bitrate')),
}) })
extract_interaction_statistic(e) extract_interaction_statistic(e)
extract_chapter_information(e) extract_chapter_information(e)
extract_thumbnail_information(e)
def traverse_json_ld(json_ld, at_top_level=True): def traverse_json_ld(json_ld, at_top_level=True):
for e in variadic(json_ld): for e in variadic(json_ld):
@ -1717,50 +1756,70 @@ class InfoExtractor:
continue continue
if expected_type is not None and not is_type(e, expected_type): if expected_type is not None and not is_type(e, expected_type):
continue continue
rating = traverse_obj(e, ('aggregateRating', 'ratingValue'), expected_type=float_or_none) rating = traverse_obj(e, ('aggregateRating', 'ratingValue', {float_or_none}))
if rating is not None: if rating is not None:
info['average_rating'] = rating info['average_rating'] = rating
if is_type(e, 'TVEpisode', 'Episode', 'PodcastEpisode'): if is_type(e, 'TVEpisode', 'Episode', 'PodcastEpisode'):
episode_name = unescapeHTML(e.get('name')) info.update(traverse_obj(e, {
info.update({ 'id': ('identifier', {str_or_none}),
'episode': episode_name, 'ext': ('encodingFormat', {mimetype2ext}),
'episode_number': int_or_none(e.get('episodeNumber')), 'title': (('title', 'name'), {clean_html}, filter, any),
'description': unescapeHTML(e.get('description')), 'creators': ('productionCompany', {clean_html}, filter, all, {orderedSet}, filter),
}) 'description': ('description', {clean_html}, filter),
if not info.get('title') and episode_name: 'duration': ((('duration', {parse_duration}), ('timeRequired', {int_or_none})), any),
info['title'] = episode_name 'episode': ('name', {clean_html}, filter),
'episode_number': ('episodeNumber', {int_or_none}),
'genres': ('genre', ..., {clean_html}, filter, all, {orderedSet}, filter),
'release_timestamp': ('datePublished', {parse_iso8601}),
}))
extract_thumbnail_information(e)
part_of_season = e.get('partOfSeason') part_of_season = e.get('partOfSeason')
if is_type(part_of_season, 'TVSeason', 'Season', 'CreativeWorkSeason'): if is_type(part_of_season, 'TVSeason', 'Season', 'CreativeWorkSeason'):
info.update({ info.update(traverse_obj(e, {
'season': unescapeHTML(part_of_season.get('name')), 'season': ('name', {clean_html}, filter),
'season_number': int_or_none(part_of_season.get('seasonNumber')), 'season_number': ('seasonNumber', {int_or_none}),
}) }))
part_of_series = e.get('partOfSeries') or e.get('partOfTVSeries') part_of_series = e.get('partOfSeries') or e.get('partOfTVSeries')
if is_type(part_of_series, 'TVSeries', 'Series', 'CreativeWorkSeries'): if is_type(part_of_series, 'TVSeries', 'Series', 'CreativeWorkSeries'):
info['series'] = unescapeHTML(part_of_series.get('name')) info['series'] = traverse_obj(part_of_series, ('name', {clean_html}, filter))
elif is_type(e, 'Movie'): elif is_type(e, 'Movie'):
info.update({ info.update(traverse_obj(e, {
'title': unescapeHTML(e.get('name')), 'title': ('name', {clean_html}, filter),
'description': unescapeHTML(e.get('description')), 'age_limit': ('contentRating', {parse_age_limit}),
'duration': parse_duration(e.get('duration')), 'cast': ('actor', ..., 'name', {clean_html}, filter, all, {orderedSet}, filter),
'timestamp': unified_timestamp(e.get('dateCreated')), 'creators': ('director', (None, ((None, ...), 'name')), {clean_html}, filter, all, {orderedSet}, filter),
}) 'description': ('description', {clean_html}, filter),
'duration': ('duration', {parse_duration}),
'genres': ('genre', (None, ...), {clean_html}, filter, all, {orderedSet}, filter),
'release_timestamp': ('datePublished', {parse_iso8601}),
}))
extract_thumbnail_information(e)
if date := traverse_obj(e, ('dateCreated', {str_or_none})):
if re.fullmatch(r'\d{4}', date):
info['release_year'] = int_or_none(date)
elif re.fullmatch(r'\d{4}-\d{2}-\d{2}', date):
info['upload_date'] = unified_strdate(date)
else:
info['timestamp'] = parse_iso8601(date)
elif is_type(e, 'Article', 'NewsArticle'): elif is_type(e, 'Article', 'NewsArticle'):
info.update(**traverse_obj(e, { info.update(traverse_obj(e, {
'title': ('headline', {clean_html}, filter), 'title': ('headline', {clean_html}, filter),
'alt_title': ('alternativeHeadline', {clean_html}, filter), 'alt_title': ('alternativeHeadline', {clean_html}, filter),
'categories': ('articleSection', {clean_html}, filter, all, filter), 'categories': ('articleSection', (None, ...), {clean_html}, filter, all, {orderedSet}, filter),
'creators': ('author', (None, 'name'), {clean_html}, filter, all, filter), 'comment_count': ('commentCount', {int_or_none}),
'creators': ('author', (None, ...), 'name', {clean_html}, filter, all, {orderedSet}, filter),
'description': (('description', 'articleBody'), {clean_html}, filter, any), 'description': (('description', 'articleBody'), {clean_html}, filter, any),
'duration': ('timeRequired', {int_or_none}),
'genres': ('genre', (None, ...), {clean_html}, filter, all, {orderedSet}, filter),
'location': ('contentLocation', 'name', {clean_html}, filter),
'modified_timestamp': ('dateModified', {parse_iso8601}), 'modified_timestamp': ('dateModified', {parse_iso8601}),
'release_timestamp': ('datePublished', {parse_iso8601}), 'release_timestamp': ('datePublished', {parse_iso8601}),
'tags': ('keywords', {clean_html}, {lambda x: x.split(',')}, ..., {str.strip}, filter, all, filter), 'tags': ('keywords', (None, ...), {clean_html},
'thumbnails': ('image', ..., { {lambda x: x.split(',')}, ..., {str.strip}, filter, all, {orderedSet}, filter),
'url': ({str}, {unescapeHTML}, {self._proto_relative_url}, {url_or_none}),
}),
'timestamp': ('dateCreated', {parse_iso8601}), 'timestamp': ('dateCreated', {parse_iso8601}),
'uploader': ('publisher', 'name', {clean_html}, filter), 'uploader': ('publisher', 'name', {clean_html}, filter),
})) }))
extract_thumbnail_information(e)
if is_type(traverse_obj(e, ('video', 0)), 'VideoObject'): if is_type(traverse_obj(e, ('video', 0)), 'VideoObject'):
extract_video_object(e['video'][0]) extract_video_object(e['video'][0])
elif is_type(traverse_obj(e, ('subjectOf', 0)), 'VideoObject'): elif is_type(traverse_obj(e, ('subjectOf', 0)), 'VideoObject'):

Write Preview
Loading…
Cancel
Save