diff options
Diffstat (limited to 'numpy/lib/recfunctions.py')
| -rw-r--r-- | numpy/lib/recfunctions.py | 52 |
1 files changed, 27 insertions, 25 deletions
diff --git a/numpy/lib/recfunctions.py b/numpy/lib/recfunctions.py index 40c687937..a2ee19c6b 100644 --- a/numpy/lib/recfunctions.py +++ b/numpy/lib/recfunctions.py @@ -27,11 +27,12 @@ _check_fill_value = np.ma.core._check_fill_value __all__ = [ 'append_fields', 'apply_along_fields', 'assign_fields_by_name', - 'drop_fields', 'find_duplicates', 'get_fieldstructure', + 'drop_fields', 'find_duplicates', 'flatten_descr', + 'get_fieldstructure', 'get_names', 'get_names_flat', 'join_by', 'merge_arrays', 'rec_append_fields', 'rec_drop_fields', 'rec_join', 'recursive_fill_fields', - 'rename_fields', 'require_fields', 'stack_arrays', - 'structured_to_unstructured', + 'rename_fields', 'repack_fields', 'require_fields', + 'stack_arrays', 'structured_to_unstructured', ] @@ -78,7 +79,7 @@ def recursive_fill_fields(input, output): return output -def get_fieldspec(dtype): +def _get_fieldspec(dtype): """ Produce a list of name/dtype pairs corresponding to the dtype fields @@ -93,7 +94,7 @@ def get_fieldspec(dtype): >>> dt = np.dtype([(('a', 'A'), np.int64), ('b', np.double, 3)]) >>> dt.descr [(('a', 'A'), '<i8'), ('b', '<f8', (3,))] - >>> get_fieldspec(dt) + >>> _get_fieldspec(dt) [(('a', 'A'), dtype('int64')), ('b', dtype(('<f8', (3,))))] """ @@ -211,7 +212,7 @@ def _zip_dtype_dispatcher(seqarrays, flatten=None): @array_function_dispatch(_zip_dtype_dispatcher) -def zip_dtype(seqarrays, flatten=False): +def _zip_dtype(seqarrays, flatten=False): newdtype = [] if flatten: for a in seqarrays: @@ -221,14 +222,14 @@ def zip_dtype(seqarrays, flatten=False): current = a.dtype if current.names and len(current.names) <= 1: # special case - dtypes of 0 or 1 field are flattened - newdtype.extend(get_fieldspec(current)) + newdtype.extend(_get_fieldspec(current)) else: newdtype.append(('', current)) return np.dtype(newdtype) @array_function_dispatch(_zip_dtype_dispatcher) -def zip_descr(seqarrays, flatten=False): +def _zip_descr(seqarrays, flatten=False): """ Combine the dtype description of a series of arrays. @@ -239,7 +240,7 @@ def zip_descr(seqarrays, flatten=False): flatten : {boolean}, optional Whether to collapse nested descriptions. """ - return zip_dtype(seqarrays, flatten=flatten).descr + return _zip_dtype(seqarrays, flatten=flatten).descr def get_fieldstructure(adtype, lastname=None, parents=None,): @@ -325,7 +326,7 @@ def _izip_records_dispatcher(seqarrays, fill_value=None, flatten=None): @array_function_dispatch(_izip_records_dispatcher) -def izip_records(seqarrays, fill_value=None, flatten=True): +def _izip_records(seqarrays, fill_value=None, flatten=True): """ Returns an iterator of concatenated items from a sequence of arrays. @@ -447,7 +448,7 @@ def merge_arrays(seqarrays, fill_value=-1, flatten=False, # Make sure we have named fields if not seqdtype.names: seqdtype = np.dtype([('', seqdtype)]) - if not flatten or zip_dtype((seqarrays,), flatten=True) == seqdtype: + if not flatten or _zip_dtype((seqarrays,), flatten=True) == seqdtype: # Minimal processing needed: just make sure everythng's a-ok seqarrays = seqarrays.ravel() # Find what type of array we must return @@ -470,7 +471,7 @@ def merge_arrays(seqarrays, fill_value=-1, flatten=False, sizes = tuple(a.size for a in seqarrays) maxlength = max(sizes) # Get the dtype of the output (flattening if needed) - newdtype = zip_dtype(seqarrays, flatten=flatten) + newdtype = _zip_dtype(seqarrays, flatten=flatten) # Initialize the sequences for data and mask seqdata = [] seqmask = [] @@ -498,9 +499,9 @@ def merge_arrays(seqarrays, fill_value=-1, flatten=False, seqdata.append(itertools.chain(data, [fval] * nbmissing)) seqmask.append(itertools.chain(mask, [fmsk] * nbmissing)) # Create an iterator for the data - data = tuple(izip_records(seqdata, flatten=flatten)) + data = tuple(_izip_records(seqdata, flatten=flatten)) output = ma.array(np.fromiter(data, dtype=newdtype, count=maxlength), - mask=list(izip_records(seqmask, flatten=flatten))) + mask=list(_izip_records(seqmask, flatten=flatten))) if asrecarray: output = output.view(MaskedRecords) else: @@ -518,7 +519,7 @@ def merge_arrays(seqarrays, fill_value=-1, flatten=False, else: fval = None seqdata.append(itertools.chain(data, [fval] * nbmissing)) - output = np.fromiter(tuple(izip_records(seqdata, flatten=flatten)), + output = np.fromiter(tuple(_izip_records(seqdata, flatten=flatten)), dtype=newdtype, count=maxlength) if asrecarray: output = output.view(recarray) @@ -745,7 +746,7 @@ def append_fields(base, names, data, dtypes=None, # output = ma.masked_all( max(len(base), len(data)), - dtype=get_fieldspec(base.dtype) + get_fieldspec(data.dtype)) + dtype=_get_fieldspec(base.dtype) + _get_fieldspec(data.dtype)) output = recursive_fill_fields(base, output) output = recursive_fill_fields(data, output) # @@ -836,17 +837,18 @@ def repack_fields(a, align=False, recurse=False): Examples -------- + >>> from numpy.lib import recfunctions as rfn >>> def print_offsets(d): ... print("offsets:", [d.fields[name][1] for name in d.names]) ... print("itemsize:", d.itemsize) ... - >>> dt = np.dtype('u1,<i4,<f4', align=True) + >>> dt = np.dtype('u1, <i8, <f8', align=True) >>> dt dtype({'names':['f0','f1','f2'], 'formats':['u1','<i8','<f8'], 'offsets':[0,8,16], 'itemsize':24}, align=True) >>> print_offsets(dt) offsets: [0, 8, 16] itemsize: 24 - >>> packed_dt = repack_fields(dt) + >>> packed_dt = rfn.repack_fields(dt) >>> packed_dt dtype([('f0', 'u1'), ('f1', '<i8'), ('f2', '<f8')]) >>> print_offsets(packed_dt) @@ -990,7 +992,7 @@ def _unstructured_to_structured_dispatcher(arr, dtype=None, names=None, return (arr,) @array_function_dispatch(_unstructured_to_structured_dispatcher) -def unstructured_to_structured(arr, dtype=None, names=None, align=False, +def _unstructured_to_structured(arr, dtype=None, names=None, align=False, copy=False, casting='unsafe'): """ Converts and n-D unstructured array into an (n-1)-D structured array. @@ -1037,7 +1039,7 @@ def unstructured_to_structured(arr, dtype=None, names=None, align=False, [ 5, 6, 7, 8, 9], [10, 11, 12, 13, 14], [15, 16, 17, 18, 19]]) - >>> unstructured_to_structured(a, dt) + >>> _unstructured_to_structured(a, dt) array([( 0, ( 1., 2), [ 3., 4.]), ( 5, ( 6., 7), [ 8., 9.]), (10, (11., 12), [13., 14.]), (15, (16., 17), [18., 19.])], dtype=[('a', '<i4'), ('b', [('f0', '<f4'), ('f1', '<u2')]), ('c', '<f4', (2,))]) @@ -1275,10 +1277,10 @@ def stack_arrays(arrays, defaults=None, usemask=True, asrecarray=False, fldnames = [d.names for d in ndtype] # dtype_l = ndtype[0] - newdescr = get_fieldspec(dtype_l) + newdescr = _get_fieldspec(dtype_l) names = [n for n, d in newdescr] for dtype_n in ndtype[1:]: - for fname, fdtype in get_fieldspec(dtype_n): + for fname, fdtype in _get_fieldspec(dtype_n): if fname not in names: newdescr.append((fname, fdtype)) names.append(fname) @@ -1501,15 +1503,15 @@ def join_by(key, r1, r2, jointype='inner', r1postfix='1', r2postfix='2', # # Build the new description of the output array ....... # Start with the key fields - ndtype = get_fieldspec(r1k.dtype) + ndtype = _get_fieldspec(r1k.dtype) # Add the fields from r1 - for fname, fdtype in get_fieldspec(r1.dtype): + for fname, fdtype in _get_fieldspec(r1.dtype): if fname not in key: ndtype.append((fname, fdtype)) # Add the fields from r2 - for fname, fdtype in get_fieldspec(r2.dtype): + for fname, fdtype in _get_fieldspec(r2.dtype): # Have we seen the current name already ? # we need to rebuild this list every time names = list(name for name, dtype in ndtype) |