Question

非常简短..这是一个错误还是我错过了什么？ tmp_j是一个包含单个项目和6个分区的包。但是，我明白了类似的回应与更大的袋子。

这个特殊的包用以下构造：

>>> tmp_j = jnode_b.filter(lambda r: (r['node']['attrib']['uid'] == '8909') & 
               (r['node']['attrib']['version'] == '1')).pluck('node').pluck('attire')

看起来像：

>>> tmp_j.compute()

[{'changeset': '39455176',
  'id': '4197394169',
  'lat': '53.4803608',
  'lon': '-113.4955328',
  'timestamp': '2016-05-20T16:43:02Z',
  'uid': '8909',
  'user': 'mvexel',
  'version': '1'}]

再次感谢..

>>> tmp_j.repartition(1).map(json.dumps).to_textfiles('tmpA*.json')

正常工作，（写入文件），但

>>> tmp_j.map(json.dumps).to_textfiles('tmpA*.json')

给出

StopIteration                             Traceback (most recent call last)
<ipython-input-28-a77a33e2ff26> in <module>()
----> 1 tmp_j.map(json.dumps).to_textfiles('tmp*.json')

/Users/jlatmann/anaconda/envs/python3/lib/python3.5/site-packages/dask/bag/core.py in to_textfiles(self, path, name_function, compression, encoding, compute)
    469     def to_textfiles(self, path, name_function=str, compression='infer',
    470                      encoding=system_encoding, compute=True):
--> 471         return to_textfiles(self, path, name_function, compression, encoding, compute)
    472 
    473     def fold(self, binop, combine=None, initial=no_default, split_every=None):

/Users/jlatmann/anaconda/envs/python3/lib/python3.5/site-packages/dask/bag/core.py in to_textfiles(b, path, name_function, compression, encoding, compute)
    167     result = Bag(merge(b.dask, dsk), name, b.npartitions)
    168     if compute:
--> 169         result.compute()
    170     else:
    171         return result

/Users/jlatmann/anaconda/envs/python3/lib/python3.5/site-packages/dask/base.py in compute(self, **kwargs)
     35 
     36     def compute(self, **kwargs):
---> 37         return compute(self, **kwargs)[0]
     38 
     39     @classmethod

/Users/jlatmann/anaconda/envs/python3/lib/python3.5/site-packages/dask/base.py in compute(*args, **kwargs)
    108                 for opt, val in groups.items()])
    109     keys = [var._keys() for var in variables]
--> 110     results = get(dsk, keys, **kwargs)
    111 
    112     results_iter = iter(results)

/Users/jlatmann/anaconda/envs/python3/lib/python3.5/site-packages/dask/multiprocessing.py in get(dsk, keys, optimizations, num_workers, func_loads, func_dumps, **kwargs)
     76         # Run
     77         result = get_async(apply_async, len(pool._pool), dsk3, keys,
---> 78                            queue=queue, get_id=_process_get_id, **kwargs)
     79     finally:
     80         if cleanup:

/Users/jlatmann/anaconda/envs/python3/lib/python3.5/site-packages/dask/async.py in get_async(apply_async, num_workers, dsk, result, cache, queue, get_id, raise_on_exception, rerun_exceptions_locally, callbacks, **kwargs)
    486                 _execute_task(task, data)  # Re-execute locally
    487             else:
--> 488                 raise(remote_exception(res, tb))
    489         state['cache'][key] = res
    490         finish_task(dsk, key, state, results, keyorder.get)

StopIteration: 

Traceback
---------
  File "/Users/jlatmann/anaconda/envs/python3/lib/python3.5/site-packages/dask/async.py", line 267, in execute_task
    result = _execute_task(task, data)
  File "/Users/jlatmann/anaconda/envs/python3/lib/python3.5/site-packages/dask/async.py", line 249, in _execute_task
    return func(*args2)
  File "/Users/jlatmann/anaconda/envs/python3/lib/python3.5/site-packages/dask/bag/core.py", line 1024, in write
    firstline = next(data)

注意：是那个

>>> tmp_b = db.from_sequence(tmp_j,partition_size=3)
>>> tmp_b.map(json.dumps).to_textfiles('tmp*.json')

工作正常（但同样，tmp_b.npartitions == 1）。

再次感谢您的见解 - 我确实查看了源代码，但后来意识到我的智能/懒惰比率太低了。

当我确信自己掌握了这一点时，我会提交文档。

Answer 1

这是一个真正的错误，现在已在主人

中解决

In [1]: import dask.bag as db

In [2]: db.range(5, npartitions=5).filter(lambda x: x == 1).map(str).to_textfiles('*.txt')

In [3]: ls *.txt
0.txt  1.txt  2.txt  3.txt  4.txt  C:\nppdf32Log\debuglog.txt  foo.txt

Dask Bag.to_textfiles适用于单个分区但不适用于多个分区

1 个答案: