Add serialization support for Client and Dataset

[billschereriii]

C, C++, Fortran get to_string() methods
C++ also gets ostream& operator<<() methods
Python gets str() methods

Sample output (Client):

>>> from smartredis import Client
>>> c = Client("127.0.0.1:6381", cluster=True, logger_name="testcase"); print(c)
Client (testcase):
Clustered Redis connection:
  Redis shards at:
    127.0.0.1:6383
    127.0.0.1:6381
    127.0.0.1:6382
  Protocol: TCP
  Command parameters:
    Retry attempts: 101
    Retry interval (ms): 1000
    Attempt timeout (ms): 100
  Connection parameters:
    Retry attempts: 101
    Retry interval (ms): 1000
    Attempt timeout (ms): 100
  Threadpool: 4 threads

Sample output (DataSet):

>>> from smartredis import Dataset
>>> import numpy as np
>>> d = Dataset("This is a test")
>>> data = np.uint8([2,4,8])
>>> d.add_tensor("u8_tensor", data)
>>> data = np.double([2.0,4.1,8.3, 5.6])
>>> d.add_tensor("double_tensor", data)
>>> d.add_meta_scalar("float2_scalar", np.float(3.1415926535))
>>> d.add_meta_scalar("float_scalar", np.double(3.1415926535))
>>> d.add_meta_string("metastring", "metavalue")
>>> print(d)
DataSet (This is a test):
Tensors:
  double_tensor:
    type: double
    dimensions: [4]
    elements: 4
  u8_tensor:
    type: 8 bit unsigned integer
    dimensions: [3]
    elements: 3
Metadata:
  metastring:
    type: string
  float2_scalar:
    type: double
  float_scalar:
    type: double

[codecov-commenter]

Codecov Report

Merging #293 (b5c2a93) into develop (a90f051) will decrease coverage by 0.65%.
The diff coverage is 61.87%.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #293      +/-   ##
===========================================
- Coverage    77.13%   76.48%   -0.65%     
===========================================
  Files           66       66              
  Lines         4151     4266     +115     
===========================================
+ Hits          3202     3263      +61     
- Misses         949     1003      +54     

Impacted Files	Coverage Δ
include/metadata.h	100.00% <ø> (ø)
include/metadatafield.h	100.00% <ø> (ø)
include/redisserver.h	0.00% <ø> (ø)
include/scalarfield.h	100.00% <ø> (ø)
include/srobject.h	11.11% <ø> (ø)
include/stringfield.h	100.00% <ø> (ø)
include/tensorpack.h	100.00% <ø> (ø)
src/cpp/utility.cpp	55.05% <10.52%> (-33.18%)	⬇️
src/cpp/dataset.cpp	73.71% <61.53%> (-5.15%)	⬇️
src/cpp/redis.cpp	47.90% <80.00%> (+0.48%)	⬆️
... and 14 more

[MattToast]

Looked through the *.py/pybind/Py*.cpp files in detail and have a couple of small notes, but otherwise everything looks good. Skimmed through the reset and didn't see any immediate red flags.

Overall looks great!! This should be very helpful for users attempting to debug their simulations!!

[ashao]

Looks great and again, a great quality-of-life improvement. Feel free to ignore the comment about passing by value. However, I would request that you change the write unit to something something other than STDERR. Those are two-line fixes and don't need to re-review. Thanks @billschereriii!

[MattToast]

LGTM!!

[mellis13]

Just one minor question/suggestion. Otherwise, great addition to the library and LGTM!