rewritten specific search for me

This commit is contained in:
Hellow 2023-05-23 16:50:54 +02:00
parent 4cc36087b6
commit 70c7d831c9
2 changed files with 115 additions and 111 deletions

View File

@ -44,13 +44,6 @@ class Page(threading.Thread):
# to search stuff # to search stuff
def search(self, query: Query) -> List[DatabaseObject]: def search(self, query: Query) -> List[DatabaseObject]:
results = []
if query.is_raw:
for search_query in query.default_search:
results.extend(self.general_search(search_query))
return results
music_object = query.music_object music_object = query.music_object
search_functions = { search_functions = {
@ -67,9 +60,9 @@ class Page(threading.Thread):
r = [] r = []
for default_query in query.default_search: for default_query in query.default_search:
results.extend(self.general_search(default_query)) r.extend(self.general_search(default_query))
return results return r
def general_search(self, search_query: str) -> List[DatabaseObject]: def general_search(self, search_query: str) -> List[DatabaseObject]:
return [] return []

View File

@ -34,111 +34,30 @@ ALBUM_TYPE_MAP: Dict[str, AlbumType] = defaultdict(lambda: AlbumType.OTHER, {
}) })
class EncyclopaediaMetallum(Page): def _song_from_json(artist_html=None, album_html=None, release_type=None, title=None,
CONNECTION: Connection = Connection( lyrics_html=None) -> Song:
host="https://www.metal-archives.com/", song_id = None
logger=ENCYCLOPAEDIA_METALLUM_LOGGER if lyrics_html is not None:
) soup = BeautifulSoup(lyrics_html, 'html.parser')
anchor = soup.find('a')
raw_song_id = anchor.get('id')
song_id = raw_song_id.replace("lyricsLink_", "")
SOURCE_TYPE = SourcePages.ENCYCLOPAEDIA_METALLUM return Song(
title=title,
LOGGER = ENCYCLOPAEDIA_METALLUM_LOGGER main_artist_list=[
_artist_from_json(artist_html=artist_html)
def __init__(self): ],
self.connection: Connection = Connection( album_list=[
host="https://www.metal-archives.com/", _album_from_json(album_html=album_html, release_type=release_type, artist_html=artist_html)
logger=ENCYCLOPAEDIA_METALLUM_LOGGER ],
) source_list=[
Source(SourcePages.ENCYCLOPAEDIA_METALLUM, song_id)
super().__init__()
def song_search(self, song: Song) -> List[Song]:
endpoint = "https://www.metal-archives.com/search/ajax-advanced/searching/songs/?songTitle={song}&bandName={" \
"artist}&releaseTitle={album}&lyrics=&genre=&sEcho=1&iColumns=5&sColumns=&iDisplayStart=0" \
"&iDisplayLength=200&mDataProp_0=0&mDataProp_1=1&mDataProp_2=2&mDataProp_3=3&mDataProp_4=4&_" \
"=1674550595663"
"""
The difficult question I am facing is, that if I try every artist, with every song, with every album,
I end up with a quadratic runtime complecety O(n^2), where every step means one web request.
This.
Is not good.
"""
r = self.connection.get(
endpoint.format(song=song.title, artist=query.artist_str, album=query.album_str)
)
if r is None:
return []
return [cls.get_song_from_json(
artist_html=raw_song[0],
album_html=raw_song[1],
release_type=raw_song[2],
title=raw_song[3],
lyrics_html=raw_song[4]
) for raw_song in r.json()['aaData']]
@classmethod
def search_for_album(cls, query: Query) -> List[Album]:
endpoint = "https://www.metal-archives.com/search/ajax-advanced/searching/albums/?bandName={" \
"artist}&releaseTitle={album}&releaseYearFrom=&releaseMonthFrom=&releaseYearTo=&releaseMonthTo" \
"=&country=&location=&releaseLabelName=&releaseCatalogNumber=&releaseIdentifiers" \
"=&releaseRecordingInfo=&releaseDescription=&releaseNotes=&genre=&sEcho=1&iColumns=3&sColumns" \
"=&iDisplayStart=0&iDisplayLength=200&mDataProp_0=0&mDataProp_1=1&mDataProp_2=2&_=1674563943747"
r = cls.CONNECTION.get(endpoint.format(artist=query.artist_str, album=query.album_str))
if r is None:
return []
return [cls.get_album_from_json(
artist_html=raw_album[0],
album_html=raw_album[1],
release_type=raw_album[2]
) for raw_album in r.json()['aaData']]
@classmethod
def search_for_artist(cls, query: Query) -> List[Artist]:
endpoint = "https://www.metal-archives.com/search/ajax-advanced/searching/bands/?bandName={" \
"artist}&genre=&country=&yearCreationFrom=&yearCreationTo=&bandNotes=&status=&themes=&location" \
"=&bandLabelName=&sEcho=1&iColumns=3&sColumns=&iDisplayStart=0&iDisplayLength=200&mDataProp_0=0" \
"&mDataProp_1=1&mDataProp_2=2&_=1674565459976"
r = cls.CONNECTION.get(endpoint.format(artist=query.artist))
if r is None:
return []
data_key = 'aaData'
parsed_data = r.json()
if data_key not in parsed_data:
return []
return [
cls.get_artist_from_json(artist_html=raw_artist[0], genre=raw_artist[1], country=raw_artist[2])
for raw_artist in r.json()['aaData']
] ]
)
@classmethod
def _raw_search(cls, query: str) -> Options:
"""
Searches the default endpoint from metal archives, which intern searches only
for bands, but it is the default, thus I am rolling with it
"""
endpoint = "https://www.metal-archives.com/search/ajax-band-search/?field=name&query={query}&sEcho=1&iColumns=3&sColumns=&iDisplayStart=0&iDisplayLength=200&mDataProp_0=0&mDataProp_1=1&mDataProp_2=2"
r = cls.CONNECTION.get(endpoint.format(query=query)) def _artist_from_json(artist_html=None, genre=None, country=None) -> Artist:
if r is None:
return Options()
return Options([
cls.get_artist_from_json(artist_html=raw_artist[0], genre=raw_artist[1], country=raw_artist[2])
for raw_artist in r.json()['aaData']
])
@classmethod
def get_artist_from_json(cls, artist_html=None, genre=None, country=None) -> Artist:
""" """
TODO parse the country to a standart TODO parse the country to a standart
""" """
@ -163,8 +82,8 @@ class EncyclopaediaMetallum(Page):
] ]
) )
@classmethod
def get_album_from_json(cls, album_html=None, release_type=None, artist_html=None) -> Album: def _album_from_json(album_html=None, release_type=None, artist_html=None) -> Album:
# parse the html # parse the html
# <a href="https://www.metal-archives.com/albums/Ghost_Bath/Self_Loather/970834">Self Loather</a>' # <a href="https://www.metal-archives.com/albums/Ghost_Bath/Self_Loather/970834">Self Loather</a>'
soup = BeautifulSoup(album_html, 'html.parser') soup = BeautifulSoup(album_html, 'html.parser')
@ -173,7 +92,7 @@ class EncyclopaediaMetallum(Page):
album_url = anchor.get('href') album_url = anchor.get('href')
album_id = album_url.split("/")[-1] album_id = album_url.split("/")[-1]
album_type = cls.ALBUM_TYPE_MAP[release_type.strip()] album_type = ALBUM_TYPE_MAP[release_type.strip()]
return Album( return Album(
title=album_name, title=album_name,
@ -182,32 +101,124 @@ class EncyclopaediaMetallum(Page):
Source(SourcePages.ENCYCLOPAEDIA_METALLUM, album_url) Source(SourcePages.ENCYCLOPAEDIA_METALLUM, album_url)
], ],
artist_list=[ artist_list=[
cls.get_artist_from_json(artist_html=artist_html) _artist_from_json(artist_html=artist_html)
] ]
) )
class EncyclopaediaMetallum(Page):
SOURCE_TYPE = SourcePages.ENCYCLOPAEDIA_METALLUM
LOGGER = ENCYCLOPAEDIA_METALLUM_LOGGER
def __init__(self):
self.connection: Connection = Connection(
host="https://www.metal-archives.com/",
logger=ENCYCLOPAEDIA_METALLUM_LOGGER
)
super().__init__()
def song_search(self, song: Song) -> List[Song]:
endpoint = "https://www.metal-archives.com/search/ajax-advanced/searching/songs/?songTitle={song}&bandName={" \
"artist}&releaseTitle={album}&lyrics=&genre=&sEcho=1&iColumns=5&sColumns=&iDisplayStart=0" \
"&iDisplayLength=200&mDataProp_0=0&mDataProp_1=1&mDataProp_2=2&mDataProp_3=3&mDataProp_4=4&_" \
"=1674550595663"
"""
The difficult question I am facing is, that if I try every artist, with every song, with every album,
I end up with a quadratic runtime complecety O(n^2), where every step means one web request.
This.
Is not good.
"""
song_title = song.title
album_titles = ["*"] if song.album_collection.empty else [album.title for album in song.album_collection]
artist_titles = ["*"] if song.main_artist_collection.empty else [artist.name for artist in song.main_artist_collection]
search_results = []
for artist in artist_titles:
for album in album_titles:
r = self.connection.get(
endpoint.format(song=song_title, artist=artist, album=album)
)
if r is None:
return []
search_results.extend(_song_from_json(
artist_html=raw_song[0],
album_html=raw_song[1],
release_type=raw_song[2],
title=raw_song[3],
lyrics_html=raw_song[4]
) for raw_song in r.json()['aaData'])
return search_results
def album_search(self, album: Album) -> List[Album]:
endpoint = "https://www.metal-archives.com/search/ajax-advanced/searching/albums/?bandName={" \
"artist}&releaseTitle={album}&releaseYearFrom=&releaseMonthFrom=&releaseYearTo=&releaseMonthTo" \
"=&country=&location=&releaseLabelName=&releaseCatalogNumber=&releaseIdentifiers" \
"=&releaseRecordingInfo=&releaseDescription=&releaseNotes=&genre=&sEcho=1&iColumns=3&sColumns" \
"=&iDisplayStart=0&iDisplayLength=200&mDataProp_0=0&mDataProp_1=1&mDataProp_2=2&_=1674563943747"
album_title = album.title
artist_titles = ["*"] if album.artist_collection.empty else [artist.name for artist in album.artist_collection]
search_results = []
for artist in artist_titles:
r = self.connection.get(endpoint.format(artist=artist, album=album_title))
if r is None:
return []
search_results.extend(_album_from_json(
artist_html=raw_album[0],
album_html=raw_album[1],
release_type=raw_album[2]
) for raw_album in r.json()['aaData'])
def artist_search(self, artist: Artist) -> List[Artist]:
endpoint = "https://www.metal-archives.com/search/ajax-advanced/searching/bands/?bandName={" \
"artist}&genre=&country=&yearCreationFrom=&yearCreationTo=&bandNotes=&status=&themes=&location" \
"=&bandLabelName=&sEcho=1&iColumns=3&sColumns=&iDisplayStart=0&iDisplayLength=200&mDataProp_0=0" \
"&mDataProp_1=1&mDataProp_2=2&_=1674565459976"
r = self.connection.get(endpoint.format(artist=artist.name))
if r is None:
return []
data_key = 'aaData'
parsed_data = r.json()
if data_key not in parsed_data:
return []
return [
_artist_from_json(artist_html=raw_artist[0], genre=raw_artist[1], country=raw_artist[2])
for raw_artist in r.json()['aaData']
]
@classmethod @classmethod
def get_song_from_json(cls, artist_html=None, album_html=None, release_type=None, title=None, def _raw_search(cls, query: str) -> Options:
lyrics_html=None) -> Song: """
song_id = None Searches the default endpoint from metal archives, which intern searches only
if lyrics_html is not None: for bands, but it is the default, thus I am rolling with it
soup = BeautifulSoup(lyrics_html, 'html.parser') """
anchor = soup.find('a') endpoint = "https://www.metal-archives.com/search/ajax-band-search/?field=name&query={query}&sEcho=1&iColumns=3&sColumns=&iDisplayStart=0&iDisplayLength=200&mDataProp_0=0&mDataProp_1=1&mDataProp_2=2"
raw_song_id = anchor.get('id')
song_id = raw_song_id.replace("lyricsLink_", "") r = cls.CONNECTION.get(endpoint.format(query=query))
if r is None:
return Options()
return Options([
cls.get_artist_from_json(artist_html=raw_artist[0], genre=raw_artist[1], country=raw_artist[2])
for raw_artist in r.json()['aaData']
])
return Song(
title=title,
main_artist_list=[
cls.get_artist_from_json(artist_html=artist_html)
],
album_list=[
cls.get_album_from_json(album_html=album_html, release_type=release_type, artist_html=artist_html)
],
source_list=[
Source(SourcePages.ENCYCLOPAEDIA_METALLUM, song_id)
]
)
@classmethod @classmethod
def _fetch_artist_discography(cls, ma_artist_id: str) -> List[Album]: def _fetch_artist_discography(cls, ma_artist_id: str) -> List[Album]: