Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Need to a void a horrible stack dump from asking for chunk addresses from an unchunked file #1

Open
bnlawrence opened this issue Mar 25, 2024 · 1 comment
Assignees
Labels
bug Something isn't working

Comments

@bnlawrence
Copy link
Collaborator

Currently we get a horrible stack dump for

import pyfive
f = pyfive.File('zero.nc')
v = f['var']
d = v._dataobjects
d._get_chunk_addresses()

which gives:

File ~/Repositories/pyfive/pyfive/dataobjects.py:743, in DatasetDataObject._get_chunk_addresses(self)
    739 if self._zchunk_index == {}:
    741     self._get_chunk_params()
--> 743     self.chunk_btree = BTreeV1RawDataChunks(
    744         self.fh, self._chunk_address, self._chunk_dims)
    746     count = np.prod(self.shape)
    747     itemsize = np.dtype(self.dtype).itemsize

File ~/Repositories/pyfive/pyfive/btree.py:129, in BTreeV1RawDataChunks.__init__(self, fh, offset, dims)
    127 """ initalize. """
    128 self.dims = dims
--> 129 super().__init__(fh, offset)

File ~/Repositories/pyfive/pyfive/btree.py:25, in AbstractBTree.__init__(self, fh, offset)
     22 self.depth = None
     23 self.all_nodes = {}
---> 25 self._read_root_node()
     26 self._read_children()

File ~/Repositories/pyfive/pyfive/btree.py:39, in AbstractBTree._read_root_node(self)
     38 def _read_root_node(self):
---> 39     root_node = self._read_node(self.offset, None)
     40     self._add_node(root_node)
     41     self.depth = root_node['node_level']

File ~/Repositories/pyfive/pyfive/btree.py:133, in BTreeV1RawDataChunks._read_node(self, offset, node_level)
    131 def _read_node(self, offset, node_level):
    132     """ Return a single node in the b-tree located at a give offset. """
--> 133     node = self._read_node_header(offset, node_level)
    134     keys = []
    135     addresses = []

File ~/Repositories/pyfive/pyfive/btree.py:81, in BTreeV1._read_node_header(self, offset, node_level)
     79 def _read_node_header(self, offset, node_level):
     80     """ Return a single node header in the b-tree located at a give offset. """
---> 81     self.fh.seek(offset)
     82     node = _unpack_struct_from_file(self.B_LINK_NODE, self.fh)
     83     assert node['signature'] == b'TREE'

TypeError: 'NoneType' object cannot be interpreted as an integer

for a file with the following signature:

netcdf zero {
dimensions:
	y = 3 ;
	x = 4 ;
variables:
	float var(y, x) ;
		var:_Storage = "contiguous" ;
		var:_Endianness = "little" ;

// global attributes:
		:_NCProperties = "version=2,netcdf=4.9.2,hdf5=1.14.3" ;
		:_SuperblockVersion = 2 ;
		:_IsNetcdf4 = 1 ;
		:_Format = "netCDF-4" ;
}
@bnlawrence bnlawrence added the bug Something isn't working label Mar 25, 2024
@bnlawrence bnlawrence self-assigned this Mar 25, 2024
@bnlawrence
Copy link
Collaborator Author

In fixing this we might want to change the name of this method to _load_chunk_indexes to better represent what it does. We might also ask, do we need to go to the b-tree given it seems the b-tree is already loaded at this point?

@bnlawrence bnlawrence added this to the h5netcdf ready milestone Nov 11, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

1 participant