From 9f05d4302249c8807f6fdd377a7036158fae141e Mon Sep 17 00:00:00 2001
From: lars <Lars.Noack@outlook.de>
Date: Mon, 31 Oct 2022 00:56:32 +0100
Subject: [PATCH] implemented the file and folder paths in the db

---
 src/download_links.py    | 36 ++++++++++++------------------------
 src/metadata/database.py |  3 +++
 src/musify.py            |  8 ++++----
 3 files changed, 19 insertions(+), 28 deletions(-)

diff --git a/src/download_links.py b/src/download_links.py
index 1e196d2..d9fceb8 100644
--- a/src/download_links.py
+++ b/src/download_links.py
@@ -1,57 +1,45 @@
-import json
-import os.path
 import requests
 import logging
 
 import musify
 import youtube_music
+from metadata import database
 
 
 class Download:
-    def __init__(self, metadata_csv: str = ".cache1.csv", proxies: dict = None,
-                 file: str = ".cache2.csv", temp: str = "temp") -> None:
+    def __init__(self, metadata_csv: str = ".cache1.csv", proxies: dict = None) -> None:
         if proxies is not None:
             musify.set_proxy(proxies)
 
-        self.temp = temp
-        self.metadata = pd.read_csv(os.path.join(self.temp, metadata_csv), index_col=0)
-
         self.urls = []
 
-        for idx, row in self.metadata.iterrows():
-            row['artist'] = json.loads(row['artist'].replace("'", '"'))
+        for row in database.get_tracks_to_download():
+            row['artists'] = [artist['name'] for artist in row['artists']]
+
+            id_ = row['id']
 
             # check musify
             musify_url = musify.get_musify_url(row)
             if musify_url is not None:
-                self.add_url(musify_url, 'musify', dict(row))
+                self.add_url(musify_url, 'musify', id_)
                 continue
 
             # check YouTube
             youtube_url = youtube_music.get_youtube_url(row)
             if youtube_url is not None:
-                self.add_url(youtube_url, 'youtube', dict(row))
+                self.add_url(youtube_url, 'youtube', id_)
                 continue
 
             # check musify again, but with a different methode that takes longer
             musify_url = musify.get_musify_url_slow(row)
             if musify_url is not None:
-                self.add_url(musify_url, 'musify', dict(row))
+                self.add_url(musify_url, 'musify', id_)
                 continue
 
             logging.warning(f"Didn't find any sources for {row['title']}")
 
-        self.dump_urls(file)
-
-    def add_url(self, url: str, src: str, row: dict):
-        row['url'] = url
-        row['src'] = src
-
-        self.urls.append(row)
-
-    def dump_urls(self, file: str = ".cache2.csv"):
-        df = pd.DataFrame(self.urls)
-        df.to_csv(os.path.join(self.temp, file))
+    def add_url(self, url: str, src: str, id_: str):
+        database.set_download_data(id_, url, src)
 
 
 if __name__ == "__main__":
@@ -62,4 +50,4 @@ if __name__ == "__main__":
 
     s = requests.Session()
     s.proxies = proxies
-    download = Download(session=s)
+    download = Download()
diff --git a/src/metadata/database.py b/src/metadata/database.py
index 9c42439..2063ffa 100644
--- a/src/metadata/database.py
+++ b/src/metadata/database.py
@@ -198,6 +198,9 @@ def get_tracks_to_download():
     return get_custom_track(["track.downloaded == 0"])
 
 
+def get_tracks_without_src():
+    return get_custom_track(["(track.url IS NULL OR track.src IS NULL)"])
+
 def get_tracks_without_isrc():
     return get_custom_track(["track.isrc IS NULL"])
 
diff --git a/src/musify.py b/src/musify.py
index 5a652e1..69d534a 100644
--- a/src/musify.py
+++ b/src/musify.py
@@ -16,8 +16,8 @@ def set_proxy(proxies):
 
 
 def get_musify_url(row):
-    title = row.title
-    artists = row.artist
+    title = row['title']
+    artists = row['artists']
 
     url = f"https://musify.club/search/suggestions?term={artists[0]} - {title}"
 
@@ -74,8 +74,8 @@ def get_soup_of_search(query: str):
 
 
 def search_for_track(row):
-    track = row.title
-    artist = row.artist
+    track = row['title']
+    artist = row['artists']
 
     soup = get_soup_of_search(f"{artist[0]} - {track}")
     tracklist_container_soup = soup.find_all("div", {"class": "playlist"})