[Numpy-discussion] structured arrays, recarrays, and record arrays

[Allan Haldane]

In light of my previous message I'd like to bring up 
https://github.com/numpy/numpy/issues/3581, as it is now clearer to me 
what is happening. In the example on that page the user creates a 
recarray and a record array (in my nomenclature) without realizing that 
they are slightly different types of beast. This is probably because the 
str() or repr() representations of these two objects are identical. To 
distinguish them you have to look at their dtype.type. Using the setup 
from my last message:

     >>> print repr(recarr)
     rec.array([(1, 'a'), (2, 'b')],
           dtype=[('foo', '<i8'), ('bar', 'S1')])
     >>> print repr(recordarr)
     rec.array([(1, 'a'), (2, 'b')],
           dtype=[('foo', '<i8'), ('bar', 'S1')])
     >>> print repr(recarr.dtype)
     dtype([('foo', '<i8'), ('bar', 'S1')])
     >>> print repr(recordarr.dtype)
     dtype([('foo', '<i8'), ('bar', 'S1')])
     >>> print recarr.dtype.type
     <type 'numpy.void'>
     >>> print recordarr.dtype.type
     <class 'numpy.core.records.record'>

Based on this, it occurs to me that the repr of a dtype should list 
dtype.type if it is not numpy.void. This might be nice to see:

 >>> print repr(recarr.dtype)
dtype([('foo', '<i8'), ('bar', 'S1')])
 >>> print repr(recordarr.dtype)
dtype((numpy.core.records.record, [('foo', '<i8'), ('bar', 'S1')]))

I could easily implement this by redefining __repr__ for the 
numpy.core.records.record class, but this does not solve the problem for 
any other cases of overridden base_dtype. So perhaps modifications 
should be made to the original repr function of dtype (in the functions 
arraydescr_struct_str and arraydescr_struct_repr in 
numpy/core/src/multiarray/descriptor.c). However, also note that the doc 
http://docs.scipy.org/doc/numpy/reference/arrays.dtypes.html says that 
creating dtypes using the form dtype((base_dtype, new_dtype)) is 
discouraged (near the bottom).

Another possibility is to discourage recarrays, and only document record 
arrays (or vv). However, many people's code already depends on both of 
these types.

Is any of this at all reasonable? It would require a change to dtype str 
and repr, which could affect a lot of things.

Cheers,
Allan

On 01/18/2015 11:36 PM, Allan Haldane wrote:
> Hello all,
>
> Documentation of recarrays is poor and I'd like to improve it. In order
> to do this I've been looking at core/records.py, and I would appreciate
> some feedback on my plan.
>
> Let me start by describing what I see. In the docs there is some
> confusion about 'structured arrays' vs 'record arrays' vs 'recarrays' -
> the docs use them often interchangeably. They also refer to structured
> dtypes alternately as 'struct data types', 'record data types' or simply
> 'records' (eg, see the reference/arrays.dtypes and
> reference/arrays.indexing doc pages).
>
> But by my reading of the code there are really three (or four) distinct
> types of arrays with structure. Here's a possible nomenclature:
>   * "Structured arrays" are simply ndarrays with structured dtypes. That
>     is, the data type is subdivided into fields of different type.
>   * "recarrays" are a subclass of ndarrays that allow access to the
>     fields by attribute.
>   * "Record arrays" are recarrays where the elements have additionally
>     been converted to 'numpy.core.records.record' type such that each
>     data element is an object with field attributes.
>   * (it is also possible to create arrays with dtype.dtype of
>     numpy.core.records.record, but which are not recarrays. However I
>     have never seen this done.)
>
> Here's code demonstrating the creation of the different types of array
> (in order: structured array, recarray, ???, record array).
>
>      >>> arr = np.array([(1,'a'), (2,'b')],
>                         dtype=[('foo', int), ('bar', 'S1')])
>      >>> recarr = arr.view(type=np.recarray)
>      >>> noname = arr.view(dtype=dtype(np.record, arr.dtype))
>      >>> recordarr = arr.view(dtype=dtype((np.record, arr.dtype)),
>                               type=np.recarray)
>
>      >>> type(arr), arr.dtype.type
>          (numpy.ndarray, numpy.void)
>      >>> type(recarr), recarr.dtype.type
>          (numpy.core.records.recarray, numpy.void)
>      >>> type(recordarr), recordarr.dtype.type
>          (numpy.core.records.recarray, numpy.core.records.record)
>
> Note that the functions numpy.rec.array, numpy.rec.fromrecords,
> numpy.rec.fromarrays, and np.recarray.__new__ create record arrays.
> However, in the docs you can see examples of the creation of recarrays,
> eg in the recarray and ndarray.view doctrings and in
> http://www.scipy.org/Cookbook/Recarray. The files
> numpy/lib/recfunctions.py and numpy/lib/npyio.py (and possibly masked
> arrays, but I haven't looked yet) make extensive use of recarrays (but
> not record arrays).
>
> The main functional difference between recarrays and record arrays is
> field access on individual elements:
>
>      >>> recordarr[0].foo
>      1
>      >>> recarr[0].foo
>      Traceback (most recent call last):
>        File "<stdin>", line 1, in <module>
>      AttributeError: 'numpy.void' object has no attribute 'foo'
>
> Also, note that recarrays have a small performance penalty relative to
> structured arrays, and record arrays have another one relative to
> recarrays because of the additional python logic.
>
> So my first goal in updating the docs is to use the right terms in the
> right place. In almost all cases, references to 'records' (eg 'record
> types') should be replaced with 'structured' (eg 'structured types'),
> with the exception of docs that deal specifically with record arrays.
> It's my guess that in the distant past structured datatypes were
> intended to always be of type numpy.core.records.record (thus the
> description in reference/arrays.dtypes) but that
> numpy.core.records.record became generally obsolete without updates to
> the docs. doc/records.rst.txt seems to document the transition.
>
> I've made a preliminary pass of the docs, which you can see here
> https://github.com/ahaldane/numpy/commit/d87633b228dabee2ddfe75d1ee9e41ba7039e715
>
> Mostly I renamed 'record type' to 'structured type', and added a very
> rough draft to numpy/doc/structured_arrays.py.
>
> I would love to hear from those more knowledgeable than myself on
> whether this works!
>
> Cheers,
> Allan