小编the*_*own的帖子

有一个带值的 df

df：

165  156  1    test    greater 56gsa
-------------------------------------
spin 201  2    normal  lesser  12asgs
pine 202  3    fast    greater 5sasgs

所需输出：

0     1   2     3        4       5
-------------------------------------
165  156  1    test    greater 56gsa
spin 201  2    normal  lesser  12asgs
pine 202  3    fast    greater 5sasgs

有一个带有值的集合

_id:ObjectId('......')
ton_id :ObjectId('abcd')
value:587900
date:2019-12-13T07:09:40.075+00:00


_id:ObjectId('......')
ton_id :ObjectId('abcd')
value:50540
date:2018-1-13T07:09:40.075+00:00

_id:ObjectId('......')
ton_id :ObjectId('abcd1')
value:55400
date:2019-5-13T07:09:40.075+00:00


_id:ObjectId('......')
ton_id :ObjectId('abcd1')
value:22500
date:2018-12-13T07:09:40.075+00:00

对于 ton_idsabcd并abcd1删除除最旧记录之外的所有记录。

所需的输出

_id:ObjectId('......')
ton_id :ObjectId('abcd')
value:50540
date:2018-1-13T07:09:40.075+00:00



_id:ObjectId('......')
ton_id :ObjectId('abcd1')
value:22500
date:2018-12-13T07:09:40.075+00:00

有一个带值的 df

df

0      1
sun   NaN
moon  123
cat   NaN
dog   yatch

将已经存在的值转为通过，将 NaN 转为失败

预期输出

0      1
sun   fail
moon  pass
cat   fail
dog   pass