Python-无法从dict解析对象

我有两个版本的Python脚本。第一个版本运行良好，但看起来很难看。从我的观点来看，第二种方法应该有效，但实际上不行。在第二个版本中，我总是遇到以下错误：

new_object=Media.objects.create（filename=i.get（'filename'）， AttributeError:“str”对象没有属性“get”

我也试过I.get['filename']，但运气不好，我做错了什么？！ 我不明白的一点是，“在clean_result['streams']”中的s可以很好地工作，但是“在clean_result['format']”中的I不可以，但不是相同的，一个我可以拔出键的dict？！我可能有一个列表和一个dict混合在一起吗

第一个工作版本：

def import_media(arg):
    """Takes a single media file as argument from disk or http/https (descriptor="/path/to/file.mp4/m4a/mp3...")
        and scrapes it for metadata. A utf8 encoded JSON object will be returned"""

    def get_plist_names(name_dict):
        return [o["string"] for o in (name_dict if isinstance(name_dict, list) else [name_dict])]

    def path_leaf(path):
        filename = path.split("?")[0]
        return filename

descriptor = str("path_to_my_media_file")
scrape_cmd = join(['ffprobe', '-print_format', 'json',
                   '-show_streams', '-show_format', '-loglevel',
                   'quiet', '-hide_banner', descriptor])
scrapped_assets = subprocess.getoutput(scrape_cmd)
clean_result = json.loads(scrapped_assets)
metadata = json.dumps(clean_result, indent=4, ensure_ascii=False, sort_keys=True).encode('utf8').strip().decode()
print(metadata)
filename_value = None
bitrate_value = None
duration_value = None
size_value = None
nb_streams_value = None
media_type_value = None
title_value = None
artist_value = None
release_date_value = None
genre_value = None
synopsis_value = None
disc_value = None
total_discs_value = None
track_value = None
total_tracks_value = None
copyright_value = None
quality_lvl_value = None
album_value = None
show_value = None
season_number_value = None
episode_id_value = None
episode_sort_value = None
rating_unit_value = None
rating_value = None
cast_value = None
directors_value = None
producers_value = None
screenwriters_value = None
studio_value = None
# Parse Metadata
if 'filename' in clean_result['format']:
    filename_value = path_leaf(str(clean_result['format']['filename']))
if 'bit_rate' in clean_result['format']:
    bitrate_value = int(clean_result['format']['bit_rate'])
if 'duration' in clean_result['format']:
    duration_value = clean_result['format']['duration']
if 'size' in clean_result['format']:
    size_value = int(clean_result['format']['size'])
if 'nb_streams' in clean_result['format']:
    nb_streams_value = int(clean_result['format']['nb_streams'])
if 'media_type' in clean_result['format']['tags']:
    media_type_value = int(clean_result['format']['tags']['media_type'])
if 'title' in clean_result['format']['tags']:
    title_value = clean_result['format']['tags']['title']
if 'artist' in clean_result['format']['tags']:
    artist_value = clean_result['format']['tags']['artist']
if 'date' in clean_result['format']['tags']:
    release_date_value = str(clean_result['format']['tags']['date'][:10])
if 'genre' in clean_result['format']['tags']:
    genre_value = str(clean_result['format']['tags']['genre'])
if 'synopsis' in clean_result['format']['tags']:
    synopsis_value = str(clean_result['format']['tags']['synopsis'])
if 'disc' in clean_result['format']['tags']:
    try:
        disc_value = int(clean_result['format']['tags']['disc'].rsplit('/', 1)[0])
        total_discs_value = int(clean_result['format']['tags']['disc'].rsplit('/', 1)[1])
    except Exception as e:
        pass
if 'track' in clean_result['format']['tags']:
    try:
        track_value = int(clean_result['format']['tags']['track'].rsplit('/', 1)[0])
        total_tracks_value = int(clean_result['format']['tags']['track'].rsplit('/', 1)[1])
    except Exception as e:
        pass
if 'copyright' in clean_result['format']['tags']:
    copyright_value = str(clean_result['format']['tags']['copyright'])
if 'hd_video' in clean_result['format']['tags']:
    quality_lvl_value = int(clean_result['format']['tags']['hd_video'])
# TV-Show Specifics Metadata
if 'album' in clean_result['format']['tags']:
    album_value = str(clean_result['format']['tags']['album'])
if 'season_number' in clean_result['format']['tags']:
    season_number_value = int(clean_result['format']['tags']['season_number'])
if 'show' in clean_result['format']['tags']:
    show_value = str(clean_result['format']['tags']['show'])
if 'episode_id' in clean_result['format']['tags']:
    episode_id_value = str(clean_result['format']['tags']['episode_id'])
if 'episode_sort' in clean_result['format']['tags']:
    episode_sort_value = int(clean_result['format']['tags']['episode_sort'])
# Packet Metadata
if 'iTunEXTC' in clean_result['format']['tags']:
    rating_unit_value = str(clean_result['format']['tags']['iTunEXTC'].rsplit('|', 3)[0])
    rating_value = str(clean_result['format']['tags']['iTunEXTC'].rsplit('|', 3)[1])
if 'iTunMOVI' in clean_result['format']['tags']:
    xml_parse = xmltodict.parse(clean_result['format']['tags']['iTunMOVI'])  # Parse given XML data to dict.
    dump_json = json.dumps(xml_parse)  # dump given dict as JSON
    plist_metadata = json.loads(dump_json)
    dict = plist_metadata['plist']['dict']
    zipped = zip(dict["key"], dict["array"])
    result = {"cast": [], "directors": [], "screenwriters": [], "studio": [], "producers": [], } | {
        k: get_plist_names(v["dict"]) for k, v in zipped}
    cast_value = str(', '.join(result["cast"]))
    directors_value = str(', '.join(result["directors"]))
    producers_value = str(', '.join(result["producers"]))
    screenwriters_value = str(', '.join(result["screenwriters"]))
    if 'studio' in plist_metadata['plist']['dict']['key']:
        studio_value = plist_metadata['plist']['dict']['string']
if not Media.objects.filter(filename=filename_value).exists():
    new_object = Media.objects.create(filename=filename_value, bitrate=bitrate_value, duration=duration_value,
                                      size=size_value, nb_streams=nb_streams_value, media_type=media_type_value,
                                      title=title_value, artist=artist_value, release_date=release_date_value,
                                      genre=genre_value, synopsis=synopsis_value, disc=disc_value,
                                      total_discs=total_discs_value, track=track_value, total_tracks=total_tracks_value,
                                      copyright=copyright_value, quality_lvl=quality_lvl_value, album=album_value,
                                      season_number=season_number_value, show=show_value, episode_id=episode_id_value,
                                      episode_sort=episode_sort_value, rating_unit=rating_unit_value,
                                      rating=rating_value,
                                      cast=cast_value, directors=directors_value, producers=producers_value,
                                      screenwriters=screenwriters_value, studio=studio_value
                                      )
    new_object.save()
    for s in clean_result['streams']:
        stream = MediaStreams.objects.create(
            index=s.get('index'),
            stream_bitrate=s.get('bit_rate'),
            codec_name=s.get('codec_name'),
            codec_type=s.get('codec_type'),
            width=s.get('width'),
            height=s.get('height'),
            channel_layout=s.get('channel_layout'),
            language=s.get("tags", {'x': None}).get("language"),
            media=new_object
        )
        stream.save()

第二个版本，不工作：

def import_media(arg):
    """Takes a single media file as argument from disk or http/https (descriptor="/path/to/file.mp4")
        and scrapes it for metadata. A utf8 encoded JSON object will be returned"""

    def get_plist_names(name_dict):
        return [o["string"] for o in (name_dict if isinstance(name_dict, list) else [name_dict])]

    def path_leaf(path):
        filename = path.split("?")[0]
        return filename

    descriptor = str("path_to_my_media_file")
    scrape_cmd = join(['ffprobe', '-print_format', 'json',
                       '-show_streams', '-show_format', '-loglevel',
                       'quiet', '-hide_banner', descriptor])
    scrapped_assets = subprocess.getoutput(scrape_cmd)
    clean_result = json.loads(scrapped_assets)
    metadata = json.dumps(clean_result, indent=4, ensure_ascii=False, sort_keys=True).encode('utf8').strip().decode()
    print(metadata)
    # General Meta
    filename_value = path_leaf(str(clean_result['format']['filename']))
    cast_value = None
    directors_value = None
    producers_value = None
    screenwriters_value = None
    studio_value = None
    # Parse Metadata
    if 'iTunMOVI' in clean_result['format']['tags']:
        xml_parse = xmltodict.parse(clean_result['format']['tags']['iTunMOVI'])  # Parse given XML data to dict.
        dump_json = json.dumps(xml_parse)  # dump given dict as JSON
        plist_metadata = json.loads(dump_json)
        dict = plist_metadata['plist']['dict']
        zipped = zip(dict["key"], dict["array"])
        result = {"cast": [], "directors": [], "screenwriters": [], "studio": [], "producers": [], } | {k: get_plist_names(v["dict"]) for k, v in zipped}
        cast_value = str(', '.join(result["cast"]))
        directors_value = str(', '.join(result["directors"]))
        producers_value = str(', '.join(result["producers"]))
        screenwriters_value = str(', '.join(result["screenwriters"]))
        if 'studio' in plist_metadata['plist']['dict']['key']:
            studio_value = plist_metadata['plist']['dict']['string']
    if not Media.objects.filter(filename=filename_value).exists():
        for i in clean_result['format']:
            new_object = Media.objects.create(filename=i.get('filename'),
                                              bitrate=i.get('bit_rate'),
                                              duration=i.get('duration'),
                                              size=i.get('size'),
                                              nb_streams=i.get('nb_streams'),
                                              media_type=i.get("tags", {'x': None}).get(int(("media_type"))),
                                              title=i.get("tags", {'x': None}).get(str(("title"))),
                                              artist=i.get("tags", {'x': None}).get(str(("artist"))),
                                              release_date=i.get("tags", {'x': None}).get(str(("release_date")[:10])),
                                              genre=i.get("tags", {'x': None}).get(str(("genre"))),
                                              synopsis=i.get("tags", {'x': None}).get(str(("synopsis"))),
                                              disc=i.get("tags", {'x': None}).get(int(("disc").rsplit('/', 1)[0])),
                                              total_discs=i.get("tags", {'x': None}).get(int(("disc").rsplit('/', 1)[1])),
                                              track=i.get("tags", {'x': None}).get(int(("track").rsplit('/', 1)[0])),
                                              total_tracks=i.get("tags", {'x': None}).get(int(("total_tracks").rsplit('/', 1)[1])),
                                              copyright=i.get("tags", {'x': None}).get(str(("copyright"))),
                                              quality_lvl=i.get("tags", {'x': None}).get(int(("hd_video"))),
                                              album=i.get("tags", {'x': None}).get(str(("album"))),
                                              season_number=i.get("tags", {'x': None}).get(int(("season_number"))),
                                              show=i.get("tags", {'x': None}).get(str(("show"))),
                                              episode_id=i.get("tags", {'x': None}).get(int(("episode_id"))),
                                              episode_sort=i.get("tags", {'x': None}).get(int(("episode_sort"))),
                                              rating_unit=i.get("tags", {'x': None}).get(str(("iTunEXTC").rsplit('|', 3)[0])),
                                              rating=i.get("tags", {'x': None}).get(str(("iTunEXTC").rsplit('|', 3)[1])),
                                              cast=i.get(str(cast_value)),
                                              directors=i.get(str(directors_value)),
                                              producers=i.get(str(producers_value)),
                                              screenwriters=i.get(str(screenwriters_value)),
                                              studio=i.get(str(studio_value))
                                              )
            new_object.save()
            for s in clean_result['streams']:
                stream = MediaStreams.objects.create(
                    index=s.get('index'),
                    stream_bitrate=s.get('bit_rate'),
                    codec_name=s.get('codec_name'),
                    codec_type=s.get('codec_type'),
                    width=s.get('width'),
                    height=s.get('height'),
                    channel_layout=s.get('channel_layout'),
                    language=s.get("tags", {'x': None}).get("language"),
                    media=new_object
                )
                stream.save()

---

您的问题太多了。如果您将问题缩减到重现错误所需的最小行数，这将很有帮助。如果您包含

clean\u result

的内容，这也会很有帮助

您的错误告诉您，

i

是一个字符串。因此，它没有方法

get

。在我看来，在您的第一个版本中，clean_result['format']中没有与i对应的

，因此您应该在第二个版本中查找错误

在下面的几行中，您试图从
i
获取
字段（
filename=i.get（'filename'）
等等），但是如果
clean_result['format']
是一个dict（似乎是您代码中的一个），则
i
可能是dict的键。查看在
中使用
迭代dict时发生的情况

您在第一个版本中没有遇到此错误，因为您正在逐个访问字段，例如

if 'show' in clean_result['format']['tags']:
    show_value = str(clean_result['format']['tags']['show'])

clean_result['format']的数据类型是什么？我猜是dict还是list。里面是int、decimal和stringsEach“I”的混合体，从clean_result[“format”]中提取必须是一个dict，因为.get函数正在应用于此。好吧，问题是为什么这不起作用？我想使用get提取字符串应该不是问题，但是str由于某种原因不支持此位置的get方法。如果我们讨论数据类型I，那么代码中的for s和for I之间的区别在哪里t仅仅可以是输入格式，对吗？使用for s循环可以提取clean_结果['streams']，而使用i循环可以提取clean_结果['format']。从streams和format中提取的数据可以是不同的数据类型，因此您面临的不匹配