From 53006b35ea8b26ff31a96a423ddaa3304d0a124e Mon Sep 17 00:00:00 2001 From: bashonly <88596187+bashonly@users.noreply.github.com> Date: Thu, 29 Dec 2022 15:04:09 +0000 Subject: [PATCH] [extractor/amazon] Add `AmazonReviews` extractor (#5857) Closes #5766 Authored by: bashonly --- yt_dlp/extractor/_extractors.py | 5 +- yt_dlp/extractor/amazon.py | 116 ++++++++++++++++++++++++++++++-- 2 files changed, 113 insertions(+), 8 deletions(-) diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py index e51228aff..4fed24c35 100644 --- a/yt_dlp/extractor/_extractors.py +++ b/yt_dlp/extractor/_extractors.py @@ -87,7 +87,10 @@ from .alura import ( AluraCourseIE ) from .amcnetworks import AMCNetworksIE -from .amazon import AmazonStoreIE +from .amazon import ( + AmazonStoreIE, + AmazonReviewsIE, +) from .amazonminitv import ( AmazonMiniTVIE, AmazonMiniTVSeasonIE, diff --git a/yt_dlp/extractor/amazon.py b/yt_dlp/extractor/amazon.py index 4d3170683..a03f983e0 100644 --- a/yt_dlp/extractor/amazon.py +++ b/yt_dlp/extractor/amazon.py @@ -1,5 +1,17 @@ +import re + from .common import InfoExtractor -from ..utils import ExtractorError, int_or_none +from ..utils import ( + ExtractorError, + clean_html, + float_or_none, + get_element_by_attribute, + get_element_by_class, + int_or_none, + js_to_json, + traverse_obj, + url_or_none, +) class AmazonStoreIE(InfoExtractor): @@ -9,7 +21,7 @@ class AmazonStoreIE(InfoExtractor): 'url': 'https://www.amazon.co.uk/dp/B098XNCHLD/', 'info_dict': { 'id': 'B098XNCHLD', - 'title': 'md5:dae240564cbb2642170c02f7f0d7e472', + 'title': str, }, 'playlist_mincount': 1, 'playlist': [{ @@ -20,28 +32,32 @@ class AmazonStoreIE(InfoExtractor): 'thumbnail': r're:^https?://.*\.jpg$', 'duration': 34, }, - }] + }], + 'expected_warnings': ['Unable to extract data'], }, { 'url': 'https://www.amazon.in/Sony-WH-1000XM4-Cancelling-Headphones-Bluetooth/dp/B0863TXGM3', 'info_dict': { 'id': 'B0863TXGM3', - 'title': 'md5:d1d3352428f8f015706c84b31e132169', + 'title': str, }, 'playlist_mincount': 4, + 'expected_warnings': ['Unable to extract data'], }, { 'url': 'https://www.amazon.com/dp/B0845NXCXF/', 'info_dict': { 'id': 'B0845NXCXF', - 'title': 'md5:f3fa12779bf62ddb6a6ec86a360a858e', + 'title': str, }, 'playlist-mincount': 1, + 'expected_warnings': ['Unable to extract data'], }, { 'url': 'https://www.amazon.es/Samsung-Smartphone-s-AMOLED-Quad-c%C3%A1mara-espa%C3%B1ola/dp/B08WX337PQ', 'info_dict': { 'id': 'B08WX337PQ', - 'title': 'md5:f3fa12779bf62ddb6a6ec86a360a858e', + 'title': str, }, 'playlist_mincount': 1, + 'expected_warnings': ['Unable to extract data'], }] def _real_extract(self, url): @@ -52,7 +68,7 @@ class AmazonStoreIE(InfoExtractor): try: data_json = self._search_json( r'var\s?obj\s?=\s?jQuery\.parseJSON\(\'', webpage, 'data', id, - transform_source=lambda x: x.replace(R'\\u', R'\u')) + transform_source=js_to_json) except ExtractorError as e: retry.error = e @@ -66,3 +82,89 @@ class AmazonStoreIE(InfoExtractor): 'width': int_or_none(video.get('videoWidth')), } for video in (data_json.get('videos') or []) if video.get('isVideo') and video.get('url')] return self.playlist_result(entries, playlist_id=id, playlist_title=data_json.get('title')) + + +class AmazonReviewsIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?amazon\.(?:[a-z]{2,3})(?:\.[a-z]{2})?/gp/customer-reviews/(?P[^/&#$?]+)' + _TESTS = [{ + 'url': 'https://www.amazon.com/gp/customer-reviews/R10VE9VUSY19L3/ref=cm_cr_arp_d_rvw_ttl', + 'info_dict': { + 'id': 'R10VE9VUSY19L3', + 'ext': 'mp4', + 'title': 'Get squad #Suspicious', + 'description': 'md5:7012695052f440a1e064e402d87e0afb', + 'uploader': 'Kimberly Cronkright', + 'average_rating': 1.0, + 'thumbnail': r're:^https?://.*\.jpg$', + }, + 'expected_warnings': ['Review body was not found in webpage'], + }, { + 'url': 'https://www.amazon.com/gp/customer-reviews/R10VE9VUSY19L3/ref=cm_cr_arp_d_rvw_ttl?language=es_US', + 'info_dict': { + 'id': 'R10VE9VUSY19L3', + 'ext': 'mp4', + 'title': 'Get squad #Suspicious', + 'description': 'md5:7012695052f440a1e064e402d87e0afb', + 'uploader': 'Kimberly Cronkright', + 'average_rating': 1.0, + 'thumbnail': r're:^https?://.*\.jpg$', + }, + 'expected_warnings': ['Review body was not found in webpage'], + }, { + 'url': 'https://www.amazon.in/gp/customer-reviews/RV1CO8JN5VGXV/', + 'info_dict': { + 'id': 'RV1CO8JN5VGXV', + 'ext': 'mp4', + 'title': 'Not sure about its durability', + 'description': 'md5:1a252c106357f0a3109ebf37d2e87494', + 'uploader': 'Shoaib Gulzar', + 'average_rating': 2.0, + 'thumbnail': r're:^https?://.*\.jpg$', + }, + 'expected_warnings': ['Review body was not found in webpage'], + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + for retry in self.RetryManager(): + webpage = self._download_webpage(url, video_id) + review_body = get_element_by_attribute('data-hook', 'review-body', webpage) + if not review_body: + retry.error = ExtractorError('Review body was not found in webpage', expected=True) + + formats, subtitles = [], {} + + manifest_url = self._search_regex( + r'data-video-url="([^"]+)"', review_body, 'm3u8 url', default=None) + if url_or_none(manifest_url): + fmts, subtitles = self._extract_m3u8_formats_and_subtitles( + manifest_url, video_id, 'mp4', fatal=False) + formats.extend(fmts) + + video_url = self._search_regex( + r']+\bvalue="([^"]+)"[^>]+\bclass="video-url"', review_body, 'mp4 url', default=None) + if url_or_none(video_url): + formats.append({ + 'url': video_url, + 'ext': 'mp4', + 'format_id': 'http-mp4', + }) + + if not formats: + self.raise_no_formats('No video found for this customer review', expected=True) + + return { + 'id': video_id, + 'title': (clean_html(get_element_by_attribute('data-hook', 'review-title', webpage)) + or self._html_extract_title(webpage)), + 'description': clean_html(traverse_obj(re.findall( + r'(.+?)', review_body), -1)), + 'uploader': clean_html(get_element_by_class('a-profile-name', webpage)), + 'average_rating': float_or_none(clean_html(get_element_by_attribute( + 'data-hook', 'review-star-rating', webpage) or '').partition(' ')[0]), + 'thumbnail': self._search_regex( + r'data-thumbnail-url="([^"]+)"', review_body, 'thumbnail', default=None), + 'formats': formats, + 'subtitles': subtitles, + }