trimesh.caching module¶

caching.py¶

Functions and classes that help with tracking changes in numpy.ndarray and clearing cached values based on those changes.

You should really pip install xxhash:

``` In [23]: %timeit int(blake2b(d).hexdigest(), 16) 102 us +/- 684 ns per loop

In [24]: %timeit int(sha256(d).hexdigest(), 16) 142 us +/- 3.73 us

In [25]: %timeit xxh3_64_intdigest(d) 3.37 us +/- 116 ns per loop ```

class trimesh.caching.Cache(id_function, force_immutable=False)¶

Bases: object

Class to cache values which will be stored until the result of an ID function changes.

__init__(id_function, force_immutable=False)¶

Create a cache object.

Parameters:

id_function (function) – Returns hashable value
force_immutable (bool) – If set will make all numpy arrays read-only

clear(exclude=None)¶

Remove elements in the cache.

Parameters:: exclude (list) – List of keys in cache to not clear.

delete(key)¶: Remove a key from the cache.

id_set()¶: Set the current ID to the value of the ID function.

update(items)¶: Update the cache with a set of key, value pairs without checking id_function.

verify()¶: Verify that the cached values are still for the same value of id_function and delete all stored items if the value of id_function has changed.

class trimesh.caching.DataStore¶

Bases: Mapping

A class to store multiple numpy arrays and track them all for changes.

Operates like a dict that only stores numpy.ndarray

__init__()¶

clear()¶: Remove all data from the DataStore.

is_empty()¶

Is the current DataStore empty or not.

Returns:: empty – False if there are items in the DataStore
Return type:: bool

property mutable¶

Is data allowed to be altered or not.

Returns:: is_mutable – Can data be altered in the DataStore
Return type:: bool

pop(key)¶

update(values)¶

class trimesh.caching.DiskCache(path, expire_days=30)¶

Bases: object

Store results of expensive operations on disk with an option to expire the results. This is used to cache the multi-gigabyte test corpuses in tests/corpus.py

__init__(path, expire_days=30)¶

Create a cache on disk for storing expensive results.

Parameters:

path (str) – A writeable location on the current file path.
expire_days (int or float) – How old should results be considered expired.

get(key, fetch)¶

Get a key from the cache or run a calculation.

Parameters:

key (str) – Key to reference item with
fetch (function) – If key isn’t stored and recent run this function and store its result on disk.

class trimesh.caching.TrackedArray¶

Bases: ndarray

Subclass of numpy.ndarray that provides hash methods to track changes.

General method is to aggressively set ‘modified’ flags on operations which might (but don’t necessarily) alter the array, ideally we sometimes compute hashes when we don’t need to, but we don’t return wrong hashes ever.

We store boolean modified flag for each hash type to make checks fast even for queries of different hashes.

__hash__ : int

Runs the fastest available hash in this order:: xxh3_64, xxh_64, blake2b, sha256

byteswap(inplace=False)¶

Swap the bytes of the array elements

Toggle between low-endian and big-endian data representation by returning a byteswapped array, optionally swapped in-place. Arrays of byte-strings are not swapped. The real and imaginary parts of a complex number are swapped individually.

Parameters:: inplace (bool, optional) – If True, swap bytes in-place, default is False.
Returns:: out – The byteswapped array. If inplace is True, this is a view to self.
Return type:: ndarray

Examples

>>> import numpy as np
>>> A = np.array([1, 256, 8755], dtype=np.int16)
>>> list(map(hex, A))
['0x1', '0x100', '0x2233']
>>> A.byteswap(inplace=True)
array([  256,     1, 13090], dtype=int16)
>>> list(map(hex, A))
['0x100', '0x1', '0x3322']

Arrays of byte-strings are not swapped

>>> A = np.array([b'ceg', b'fac'])
>>> A.byteswap()
array([b'ceg', b'fac'], dtype='|S3')

A.view(A.dtype.newbyteorder()).byteswap() produces an array with the same values but different representation in memory

>>> A = np.array([1, 2, 3])
>>> A.view(np.uint8)
array([1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 3, 0, 0, 0, 0, 0,
       0, 0], dtype=uint8)
>>> A.view(A.dtype.newbyteorder()).byteswap(inplace=True)
array([1, 2, 3])
>>> A.view(np.uint8)
array([0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0,
       0, 3], dtype=uint8)

fill(value)¶

Fill the array with a scalar value.

Parameters:: value (scalar) – All elements of a will be assigned this value.

Examples

>>> import numpy as np
>>> a = np.array([1, 2])
>>> a.fill(0)
>>> a
array([0, 0])
>>> a = np.empty(2)
>>> a.fill(1)
>>> a
array([1.,  1.])

Fill expects a scalar value and always behaves the same as assigning to a single array element. The following is a rare example where this distinction is important:

>>> a = np.array([None, None], dtype=object)
>>> a[0] = np.array(3)
>>> a
array([array(3), None], dtype=object)
>>> a.fill(np.array(3))
>>> a
array([array(3), array(3)], dtype=object)

Where other forms of assignments will unpack the array being assigned:

>>> a[...] = np.array(3)
>>> a
array([3, 3], dtype=object)

itemset(*args, **kwargs)¶

property mutable¶

partition(kth, axis=-1, kind='introselect', order=None)¶

Partially sorts the elements in the array in such a way that the value of the element in k-th position is in the position it would be in a sorted array. In the output array, all elements smaller than the k-th element are located to the left of this element and all equal or greater are located to its right. The ordering of the elements in the two partitions on the either side of the k-th element in the output array is undefined.

Added in version 1.8.0.

Parameters:

kth (int or sequence of ints) –
Element index to partition by. The kth element value will be in its final sorted position and all smaller elements will be moved before it and all equal or greater elements behind it. The order of all elements in the partitions is undefined. If provided with a sequence of kth it will partition all elements indexed by kth of them into their sorted position at once.

Deprecated since version 1.22.0: Passing booleans as index is deprecated.
axis (int, optional) – Axis along which to sort. Default is -1, which means sort along the last axis.
kind ({'introselect'}, optional) – Selection algorithm. Default is ‘introselect’.
order (str or list of str, optional) – When a is an array with fields defined, this argument specifies which fields to compare first, second, etc. A single field can be specified as a string, and not all fields need to be specified, but unspecified fields will still be used, in the order in which they come up in the dtype, to break ties.

See also

numpy.put: equivalent function

setflags(write=None, align=None, uic=None)¶

Set array flags WRITEABLE, ALIGNED, WRITEBACKIFCOPY, respectively.

These Boolean-valued flags affect how numpy interprets the memory area used by a (see Notes below). The ALIGNED flag can only be set to True if the data is actually aligned according to the type. The WRITEBACKIFCOPY flag can never be set to True. The flag WRITEABLE can only be set to True if the array owns its own memory, or the ultimate owner of the memory exposes a writeable buffer interface, or is a string. (The exception for string is made so that unpickling can be done without copying memory.)

Parameters:

write (bool, optional) – Describes whether or not a can be written to.
align (bool, optional) – Describes whether or not a is aligned properly for its type.
uic (bool, optional) – Describes whether or not a is a copy of another “base” array.

Notes

Array flags provide information about how the memory area used for the array is to be interpreted. There are 7 Boolean flags in use, only three of which can be changed by the user: WRITEBACKIFCOPY, WRITEABLE, and ALIGNED.

WRITEABLE (W) the data area can be written to;

ALIGNED (A) the data and strides are aligned appropriately for the hardware (as determined by the compiler);

WRITEBACKIFCOPY (X) this array is a copy of some other array (referenced by .base). When the C-API function PyArray_ResolveWritebackIfCopy is called, the base array will be updated with the contents of this array.

All flags can be accessed using the single (upper case) letter as well as the full name.

Examples

>>> import numpy as np
>>> y = np.array([[3, 1, 7],
...               [2, 0, 0],
...               [8, 5, 9]])
>>> y
array([[3, 1, 7],
       [2, 0, 0],
       [8, 5, 9]])
>>> y.flags
  C_CONTIGUOUS : True
  F_CONTIGUOUS : False
  OWNDATA : True
  WRITEABLE : True
  ALIGNED : True
  WRITEBACKIFCOPY : False
>>> y.setflags(write=0, align=0)
>>> y.flags
  C_CONTIGUOUS : True
  F_CONTIGUOUS : False
  OWNDATA : True
  WRITEABLE : False
  ALIGNED : False
  WRITEBACKIFCOPY : False
>>> y.setflags(uic=1)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ValueError: cannot set WRITEBACKIFCOPY flag to True

sort(axis=-1, kind=None, order=None)¶

Sort an array in-place. Refer to numpy.sort for full documentation.

Parameters:

axis (int, optional) – Axis along which to sort. Default is -1, which means sort along the last axis.
kind ({'quicksort', 'mergesort', 'heapsort', 'stable'}, optional) –
Sorting algorithm. The default is ‘quicksort’. Note that both ‘stable’ and ‘mergesort’ use timsort under the covers and, in general, the actual implementation will vary with datatype. The ‘mergesort’ option is retained for backwards compatibility.

Changed in version 1.15.0: The ‘stable’ option was added.
order (str or list of str, optional) – When a is an array with fields defined, this argument specifies which fields to compare first, second, etc. A single field can be specified as a string, and not all fields need be specified, but unspecified fields will still be used, in the order in which they come up in the dtype, to break ties.