通过命令行指定爬取参数，不用再修改代码了

netease_music/sql/statistic.sql

@@ -1,13 +1,42 @@
- -- 查看需要爬取的音乐的分布
- SELECT cast( format( t_tmp.song_id / 10000000, 0) * 10000000 as UNSIGNED ) as s, count(*) as count
- FROM (
+-- 查看需要爬取的 song 的分布
+SELECT cast( format( t_tmp.song_id / 10000000, 0) * 10000000 as UNSIGNED ) as s, count(*) as count
+FROM (
     SELECT DISTINCT song_id FROM song_album_relation
     UNION
     SELECT DISTINCT song_id FROM song_artist_relation
 ) as t_tmp
- WHERE song_id NOT IN ( SELECT song_id FROM song )
- GROUP BY s
- ORDER BY s DESC
+WHERE song_id NOT IN ( SELECT song_id FROM song )
+GROUP BY s
+ORDER BY s DESC
+
+-- 查看需要爬取的 album 的分布
+SELECT cast( format( album_id / 1000000, 0) * 1000000 as UNSIGNED ) as s, count(*) as count
+FROM song_album_relation
+WHERE album_id NOT IN ( SELECT album_id FROM album )
+GROUP BY s
+ORDER BY s DESC
+
+-- 查看需要爬取的 artist 的分布
+SELECT cast( format( artist_id / 2000000, 0) * 2000000 as UNSIGNED ) as s, count(*) as count
+FROM song_artist_relation
+WHERE artist_id NOT IN ( SELECT artist_id FROM artist )
+GROUP BY s
+ORDER BY s DESC
+
+-- 查看需要爬取的 comment 的分布
+SELECT cast( format( song_id / 10000000, 0) * 10000000 as UNSIGNED ) as s, count(*) as count
+FROM comment_progress
+WHERE current_status != 2
+GROUP BY s
+ORDER BY s DESC
+
+-- 查看需要爬取的 lyric 的分布
+SELECT cast( format( song_id / 10000000, 0) * 10000000 as UNSIGNED ) as s, count(*) as count
+FROM song
+WHERE song_id NOT IN ( SELECT song_id FROM lyric )
+GROUP BY s
+ORDER BY s DESC
+
 
 -- optimize table
 optimize table album;