JSON encoding engine instrumentation

.circleci/config.yml

@@ -103,7 +103,7 @@ jobs:
  name: Build html figures
  command: |
  . venv/bin/activate
- pip install pandas statsmodels --quiet
+ pip install pandas statsmodels orjson --quiet
  python test/percy/plotly-express.py
  - run:
  name: Run percy snapshots

packages/python/plotly/plotly/basedatatypes.py

@@ -916,6 +916,12 @@ def update(self, dict1=None, overwrite=False, **kwargs):
  BaseFigure._perform_update(self[k], v)
  else:
  self[k] = v
+
+ # # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(self)
+
  return self
 
  def pop(self, key, *args):

packages/python/plotly/plotly/express/_core.py

@@ -2092,6 +2092,12 @@ def make_figure(args, constructor, trace_patch=None, layout_patch=None):
 
  configure_axes(args, constructor, fig, orders)
  configure_animation_controls(args, constructor, fig)
+
+ # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(fig)
+
  return fig
 
 

packages/python/plotly/plotly/io/_json.py

@@ -57,7 +57,78 @@ def coerce_to_strict(const):
  return const
 
 
+def time_engine(engine, plotly_object):
+ import time
+ # time in seconds
+ t_total = 0
+ n_total = 0
+
+ # Call function for at least total of 2 seconds and at least 10 times
+ n_min = 10
+ t_min = 1
+
+ while t_total < t_min or n_total < n_min:
+ t0 = time.perf_counter()
+ _to_json_plotly(plotly_object, engine=engine)
+ t1 = time.perf_counter()
+ n_total += 1
+ t_total += (t1 - t0)
+
+ # return time in ms
+ return 1000 * t_total / n_total
+
+
 def to_json_plotly(plotly_object, pretty=False, engine=None):
+ if engine is not None:
+ return _to_json_plotly(plotly_object, pretty=pretty , engine=engine)
+
+ # instrucment _to_json_plotly by running it with all 3 engines and comparing results
+ # before returnin
+
+ import timeit
+ from IPython import get_ipython
+ ipython = get_ipython()
+ orjson = get_module("orjson", should_load=True)
+ results = {}
+ timing = {}
+ result_str = None
+ for engine in ["json", "orjson"]:
+ if orjson is None and engine == "orjson":
+ continue
+
+ result_str = _to_json_plotly(plotly_object, pretty=pretty, engine=engine)
+ results[engine] = from_json_plotly(result_str, engine=engine)
+ timing[engine] = time_engine(engine, plotly_object)
+
+ # Check matches
+ if "orjson" in results:
+ if results["json"] != results["orjson"]:
+ raise ValueError(
+ """
+ {json}
+
+ {orjson}""".format(
+ json=results["json"], orjson=results["orjson"]
+ )
+ )
+
+ # write timing
+ import uuid
+ import pickle
+ import os
+ uid = str(uuid.uuid4())
+ with open("json_timing.csv".format(engine), "at") as f:
+ f.write("{}, {}, {}, {}\n".format(
+ timing["json"], timing["orjson"], len(result_str), uid)
+ )
+ os.makedirs("json_object", exist_ok=True)
+ with open("json_object/{uid}.pkl".format(uid=uid), "wb") as f:
+ pickle.dump(plotly_object, f)
+
+ return result_str
+
+
+def _to_json_plotly(plotly_object, pretty=False, engine=None):
  """
  Convert a plotly/Dash object to a JSON string representation
 
@@ -466,7 +537,7 @@ def clean_to_json_compatible(obj, **kwargs):
  elif obj.dtype.kind == "U":
  return obj.tolist()
  elif obj.dtype.kind == "O":
- # Treat object array as a lists, continue processing
+ # Treat object array as plain list, allow recursive processing below
  obj = obj.tolist()
  elif isinstance(obj, np.datetime64):
  return str(obj)

packages/python/plotly/tox.ini

@@ -75,7 +75,7 @@ deps=
  optional: matplotlib==2.2.3
  optional: scikit-image==0.14.4
  optional: kaleido
-optional: orjson==3.4.6;python_version>"3.5"
+ orjson==3.4.6;python_version>"3.5"
 
 ; CORE ENVIRONMENTS
 [testenv:py27-core]

.circleci/config.yml

@@ -103,7 +103,7 @@ jobs:
  name: Build html figures
  command: |
  . venv/bin/activate
- pip install pandas statsmodels --quiet
+ pip install pandas statsmodels orjson --quiet
  python test/percy/plotly-express.py
  - run:
  name: Run percy snapshots

packages/python/plotly/plotly/basedatatypes.py

@@ -916,6 +916,12 @@ def update(self, dict1=None, overwrite=False, **kwargs):
  BaseFigure._perform_update(self[k], v)
  else:
  self[k] = v
+
+ # # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(self)
+
  return self
 
  def pop(self, key, *args):

packages/python/plotly/plotly/express/_core.py

@@ -2092,6 +2092,12 @@ def make_figure(args, constructor, trace_patch=None, layout_patch=None):
 
  configure_axes(args, constructor, fig, orders)
  configure_animation_controls(args, constructor, fig)
+
+ # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(fig)
+
  return fig
 
 

packages/python/plotly/plotly/io/_json.py

@@ -57,7 +57,78 @@ def coerce_to_strict(const):
  return const
 
 
+def time_engine(engine, plotly_object):
+ import time
+ # time in seconds
+ t_total = 0
+ n_total = 0
+
+ # Call function for at least total of 2 seconds and at least 10 times
+ n_min = 10
+ t_min = 1
+
+ while t_total < t_min or n_total < n_min:
+ t0 = time.perf_counter()
+ _to_json_plotly(plotly_object, engine=engine)
+ t1 = time.perf_counter()
+ n_total += 1
+ t_total += (t1 - t0)
+
+ # return time in ms
+ return 1000 * t_total / n_total
+
+
 def to_json_plotly(plotly_object, pretty=False, engine=None):
+ if engine is not None:
+ return _to_json_plotly(plotly_object, pretty=pretty , engine=engine)
+
+ # instrucment _to_json_plotly by running it with all 3 engines and comparing results
+ # before returnin
+
+ import timeit
+ from IPython import get_ipython
+ ipython = get_ipython()
+ orjson = get_module("orjson", should_load=True)
+ results = {}
+ timing = {}
+ result_str = None
+ for engine in ["json", "orjson"]:
+ if orjson is None and engine == "orjson":
+ continue
+
+ result_str = _to_json_plotly(plotly_object, pretty=pretty, engine=engine)
+ results[engine] = from_json_plotly(result_str, engine=engine)
+ timing[engine] = time_engine(engine, plotly_object)
+
+ # Check matches
+ if "orjson" in results:
+ if results["json"] != results["orjson"]:
+ raise ValueError(
+ """
+ {json}
+
+ {orjson}""".format(
+ json=results["json"], orjson=results["orjson"]
+ )
+ )
+
+ # write timing
+ import uuid
+ import pickle
+ import os
+ uid = str(uuid.uuid4())
+ with open("json_timing.csv".format(engine), "at") as f:
+ f.write("{}, {}, {}, {}\n".format(
+ timing["json"], timing["orjson"], len(result_str), uid)
+ )
+ os.makedirs("json_object", exist_ok=True)
+ with open("json_object/{uid}.pkl".format(uid=uid), "wb") as f:
+ pickle.dump(plotly_object, f)
+
+ return result_str
+
+
+def _to_json_plotly(plotly_object, pretty=False, engine=None):
  """
  Convert a plotly/Dash object to a JSON string representation
 
@@ -466,7 +537,7 @@ def clean_to_json_compatible(obj, **kwargs):
  elif obj.dtype.kind == "U":
  return obj.tolist()
  elif obj.dtype.kind == "O":
- # Treat object array as a lists, continue processing
+ # Treat object array as plain list, allow recursive processing below
  obj = obj.tolist()
  elif isinstance(obj, np.datetime64):
  return str(obj)

packages/python/plotly/tox.ini

@@ -75,7 +75,7 @@ deps=
  optional: matplotlib==2.2.3
  optional: scikit-image==0.14.4
  optional: kaleido
-optional: orjson==3.4.6;python_version>"3.5"
+ orjson==3.4.6;python_version>"3.5"
 
 ; CORE ENVIRONMENTS
 [testenv:py27-core]

.circleci/config.yml

@@ -103,7 +103,7 @@ jobs:
  name: Build html figures
  command: |
  . venv/bin/activate
- pip install pandas statsmodels --quiet
+ pip install pandas statsmodels orjson --quiet
  python test/percy/plotly-express.py
  - run:
  name: Run percy snapshots

packages/python/plotly/plotly/basedatatypes.py

@@ -916,6 +916,12 @@ def update(self, dict1=None, overwrite=False, **kwargs):
  BaseFigure._perform_update(self[k], v)
  else:
  self[k] = v
+
+ # # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(self)
+
  return self
 
  def pop(self, key, *args):

packages/python/plotly/plotly/express/_core.py

@@ -2092,6 +2092,12 @@ def make_figure(args, constructor, trace_patch=None, layout_patch=None):
 
  configure_axes(args, constructor, fig, orders)
  configure_animation_controls(args, constructor, fig)
+
+ # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(fig)
+
  return fig
 
 

packages/python/plotly/plotly/io/_json.py

@@ -57,7 +57,78 @@ def coerce_to_strict(const):
  return const
 
 
+def time_engine(engine, plotly_object):
+ import time
+ # time in seconds
+ t_total = 0
+ n_total = 0
+
+ # Call function for at least total of 2 seconds and at least 10 times
+ n_min = 10
+ t_min = 1
+
+ while t_total < t_min or n_total < n_min:
+ t0 = time.perf_counter()
+ _to_json_plotly(plotly_object, engine=engine)
+ t1 = time.perf_counter()
+ n_total += 1
+ t_total += (t1 - t0)
+
+ # return time in ms
+ return 1000 * t_total / n_total
+
+
 def to_json_plotly(plotly_object, pretty=False, engine=None):
+ if engine is not None:
+ return _to_json_plotly(plotly_object, pretty=pretty , engine=engine)
+
+ # instrucment _to_json_plotly by running it with all 3 engines and comparing results
+ # before returnin
+
+ import timeit
+ from IPython import get_ipython
+ ipython = get_ipython()
+ orjson = get_module("orjson", should_load=True)
+ results = {}
+ timing = {}
+ result_str = None
+ for engine in ["json", "orjson"]:
+ if orjson is None and engine == "orjson":
+ continue
+
+ result_str = _to_json_plotly(plotly_object, pretty=pretty, engine=engine)
+ results[engine] = from_json_plotly(result_str, engine=engine)
+ timing[engine] = time_engine(engine, plotly_object)
+
+ # Check matches
+ if "orjson" in results:
+ if results["json"] != results["orjson"]:
+ raise ValueError(
+ """
+ {json}
+
+ {orjson}""".format(
+ json=results["json"], orjson=results["orjson"]
+ )
+ )
+
+ # write timing
+ import uuid
+ import pickle
+ import os
+ uid = str(uuid.uuid4())
+ with open("json_timing.csv".format(engine), "at") as f:
+ f.write("{}, {}, {}, {}\n".format(
+ timing["json"], timing["orjson"], len(result_str), uid)
+ )
+ os.makedirs("json_object", exist_ok=True)
+ with open("json_object/{uid}.pkl".format(uid=uid), "wb") as f:
+ pickle.dump(plotly_object, f)
+
+ return result_str
+
+
+def _to_json_plotly(plotly_object, pretty=False, engine=None):
  """
  Convert a plotly/Dash object to a JSON string representation
 
@@ -466,7 +537,7 @@ def clean_to_json_compatible(obj, **kwargs):
  elif obj.dtype.kind == "U":
  return obj.tolist()
  elif obj.dtype.kind == "O":
- # Treat object array as a lists, continue processing
+ # Treat object array as plain list, allow recursive processing below
  obj = obj.tolist()
  elif isinstance(obj, np.datetime64):
  return str(obj)

packages/python/plotly/tox.ini

@@ -75,7 +75,7 @@ deps=
  optional: matplotlib==2.2.3
  optional: scikit-image==0.14.4
  optional: kaleido
-optional: orjson==3.4.6;python_version>"3.5"
+ orjson==3.4.6;python_version>"3.5"
 
 ; CORE ENVIRONMENTS
 [testenv:py27-core]

.circleci/config.yml

@@ -103,7 +103,7 @@ jobs:
  name: Build html figures
  command: |
  . venv/bin/activate
- pip install pandas statsmodels --quiet
+ pip install pandas statsmodels orjson --quiet
  python test/percy/plotly-express.py
  - run:
  name: Run percy snapshots

packages/python/plotly/plotly/basedatatypes.py

@@ -916,6 +916,12 @@ def update(self, dict1=None, overwrite=False, **kwargs):
  BaseFigure._perform_update(self[k], v)
  else:
  self[k] = v
+
+ # # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(self)
+
  return self
 
  def pop(self, key, *args):

packages/python/plotly/plotly/express/_core.py

@@ -2092,6 +2092,12 @@ def make_figure(args, constructor, trace_patch=None, layout_patch=None):
 
  configure_axes(args, constructor, fig, orders)
  configure_animation_controls(args, constructor, fig)
+
+ # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(fig)
+
  return fig
 
 

packages/python/plotly/plotly/io/_json.py

@@ -57,7 +57,78 @@ def coerce_to_strict(const):
  return const
 
 
+def time_engine(engine, plotly_object):
+ import time
+ # time in seconds
+ t_total = 0
+ n_total = 0
+
+ # Call function for at least total of 2 seconds and at least 10 times
+ n_min = 10
+ t_min = 1
+
+ while t_total < t_min or n_total < n_min:
+ t0 = time.perf_counter()
+ _to_json_plotly(plotly_object, engine=engine)
+ t1 = time.perf_counter()
+ n_total += 1
+ t_total += (t1 - t0)
+
+ # return time in ms
+ return 1000 * t_total / n_total
+
+
 def to_json_plotly(plotly_object, pretty=False, engine=None):
+ if engine is not None:
+ return _to_json_plotly(plotly_object, pretty=pretty , engine=engine)
+
+ # instrucment _to_json_plotly by running it with all 3 engines and comparing results
+ # before returnin
+
+ import timeit
+ from IPython import get_ipython
+ ipython = get_ipython()
+ orjson = get_module("orjson", should_load=True)
+ results = {}
+ timing = {}
+ result_str = None
+ for engine in ["json", "orjson"]:
+ if orjson is None and engine == "orjson":
+ continue
+
+ result_str = _to_json_plotly(plotly_object, pretty=pretty, engine=engine)
+ results[engine] = from_json_plotly(result_str, engine=engine)
+ timing[engine] = time_engine(engine, plotly_object)
+
+ # Check matches
+ if "orjson" in results:
+ if results["json"] != results["orjson"]:
+ raise ValueError(
+ """
+ {json}
+
+ {orjson}""".format(
+ json=results["json"], orjson=results["orjson"]
+ )
+ )
+
+ # write timing
+ import uuid
+ import pickle
+ import os
+ uid = str(uuid.uuid4())
+ with open("json_timing.csv".format(engine), "at") as f:
+ f.write("{}, {}, {}, {}\n".format(
+ timing["json"], timing["orjson"], len(result_str), uid)
+ )
+ os.makedirs("json_object", exist_ok=True)
+ with open("json_object/{uid}.pkl".format(uid=uid), "wb") as f:
+ pickle.dump(plotly_object, f)
+
+ return result_str
+
+
+def _to_json_plotly(plotly_object, pretty=False, engine=None):
  """
  Convert a plotly/Dash object to a JSON string representation
 
@@ -466,7 +537,7 @@ def clean_to_json_compatible(obj, **kwargs):
  elif obj.dtype.kind == "U":
  return obj.tolist()
  elif obj.dtype.kind == "O":
- # Treat object array as a lists, continue processing
+ # Treat object array as plain list, allow recursive processing below
  obj = obj.tolist()
  elif isinstance(obj, np.datetime64):
  return str(obj)

packages/python/plotly/tox.ini

@@ -75,7 +75,7 @@ deps=
  optional: matplotlib==2.2.3
  optional: scikit-image==0.14.4
  optional: kaleido
-optional: orjson==3.4.6;python_version>"3.5"
+ orjson==3.4.6;python_version>"3.5"
 
 ; CORE ENVIRONMENTS
 [testenv:py27-core]

.circleci/config.yml

@@ -103,7 +103,7 @@ jobs:
  name: Build html figures
  command: |
  . venv/bin/activate
- pip install pandas statsmodels --quiet
+ pip install pandas statsmodels orjson --quiet
  python test/percy/plotly-express.py
  - run:
  name: Run percy snapshots

packages/python/plotly/plotly/basedatatypes.py

@@ -916,6 +916,12 @@ def update(self, dict1=None, overwrite=False, **kwargs):
  BaseFigure._perform_update(self[k], v)
  else:
  self[k] = v
+
+ # # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(self)
+
  return self
 
  def pop(self, key, *args):

packages/python/plotly/plotly/express/_core.py

@@ -2092,6 +2092,12 @@ def make_figure(args, constructor, trace_patch=None, layout_patch=None):
 
  configure_axes(args, constructor, fig, orders)
  configure_animation_controls(args, constructor, fig)
+
+ # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(fig)
+
  return fig
 
 

packages/python/plotly/plotly/io/_json.py

@@ -57,7 +57,78 @@ def coerce_to_strict(const):
  return const
 
 
+def time_engine(engine, plotly_object):
+ import time
+ # time in seconds
+ t_total = 0
+ n_total = 0
+
+ # Call function for at least total of 2 seconds and at least 10 times
+ n_min = 10
+ t_min = 1
+
+ while t_total < t_min or n_total < n_min:
+ t0 = time.perf_counter()
+ _to_json_plotly(plotly_object, engine=engine)
+ t1 = time.perf_counter()
+ n_total += 1
+ t_total += (t1 - t0)
+
+ # return time in ms
+ return 1000 * t_total / n_total
+
+
 def to_json_plotly(plotly_object, pretty=False, engine=None):
+ if engine is not None:
+ return _to_json_plotly(plotly_object, pretty=pretty , engine=engine)
+
+ # instrucment _to_json_plotly by running it with all 3 engines and comparing results
+ # before returnin
+
+ import timeit
+ from IPython import get_ipython
+ ipython = get_ipython()
+ orjson = get_module("orjson", should_load=True)
+ results = {}
+ timing = {}
+ result_str = None
+ for engine in ["json", "orjson"]:
+ if orjson is None and engine == "orjson":
+ continue
+
+ result_str = _to_json_plotly(plotly_object, pretty=pretty, engine=engine)
+ results[engine] = from_json_plotly(result_str, engine=engine)
+ timing[engine] = time_engine(engine, plotly_object)
+
+ # Check matches
+ if "orjson" in results:
+ if results["json"] != results["orjson"]:
+ raise ValueError(
+ """
+ {json}
+
+ {orjson}""".format(
+ json=results["json"], orjson=results["orjson"]
+ )
+ )
+
+ # write timing
+ import uuid
+ import pickle
+ import os
+ uid = str(uuid.uuid4())
+ with open("json_timing.csv".format(engine), "at") as f:
+ f.write("{}, {}, {}, {}\n".format(
+ timing["json"], timing["orjson"], len(result_str), uid)
+ )
+ os.makedirs("json_object", exist_ok=True)
+ with open("json_object/{uid}.pkl".format(uid=uid), "wb") as f:
+ pickle.dump(plotly_object, f)
+
+ return result_str
+
+
+def _to_json_plotly(plotly_object, pretty=False, engine=None):
  """
  Convert a plotly/Dash object to a JSON string representation
 
@@ -466,7 +537,7 @@ def clean_to_json_compatible(obj, **kwargs):
  elif obj.dtype.kind == "U":
  return obj.tolist()
  elif obj.dtype.kind == "O":
- # Treat object array as a lists, continue processing
+ # Treat object array as plain list, allow recursive processing below
  obj = obj.tolist()
  elif isinstance(obj, np.datetime64):
  return str(obj)

packages/python/plotly/tox.ini

@@ -75,7 +75,7 @@ deps=
  optional: matplotlib==2.2.3
  optional: scikit-image==0.14.4
  optional: kaleido
-optional: orjson==3.4.6;python_version>"3.5"
+ orjson==3.4.6;python_version>"3.5"
 
 ; CORE ENVIRONMENTS
 [testenv:py27-core]

.circleci/config.yml

@@ -103,7 +103,7 @@ jobs:
  name: Build html figures
  command: |
  . venv/bin/activate
- pip install pandas statsmodels --quiet
+ pip install pandas statsmodels orjson --quiet
  python test/percy/plotly-express.py
  - run:
  name: Run percy snapshots

packages/python/plotly/plotly/basedatatypes.py

@@ -916,6 +916,12 @@ def update(self, dict1=None, overwrite=False, **kwargs):
  BaseFigure._perform_update(self[k], v)
  else:
  self[k] = v
+
+ # # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(self)
+
  return self
 
  def pop(self, key, *args):

packages/python/plotly/plotly/express/_core.py

@@ -2092,6 +2092,12 @@ def make_figure(args, constructor, trace_patch=None, layout_patch=None):
 
  configure_axes(args, constructor, fig, orders)
  configure_animation_controls(args, constructor, fig)
+
+ # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(fig)
+
  return fig
 
 

packages/python/plotly/plotly/io/_json.py

@@ -57,7 +57,78 @@ def coerce_to_strict(const):
  return const
 
 
+def time_engine(engine, plotly_object):
+ import time
+ # time in seconds
+ t_total = 0
+ n_total = 0
+
+ # Call function for at least total of 2 seconds and at least 10 times
+ n_min = 10
+ t_min = 1
+
+ while t_total < t_min or n_total < n_min:
+ t0 = time.perf_counter()
+ _to_json_plotly(plotly_object, engine=engine)
+ t1 = time.perf_counter()
+ n_total += 1
+ t_total += (t1 - t0)
+
+ # return time in ms
+ return 1000 * t_total / n_total
+
+
 def to_json_plotly(plotly_object, pretty=False, engine=None):
+ if engine is not None:
+ return _to_json_plotly(plotly_object, pretty=pretty , engine=engine)
+
+ # instrucment _to_json_plotly by running it with all 3 engines and comparing results
+ # before returnin
+
+ import timeit
+ from IPython import get_ipython
+ ipython = get_ipython()
+ orjson = get_module("orjson", should_load=True)
+ results = {}
+ timing = {}
+ result_str = None
+ for engine in ["json", "orjson"]:
+ if orjson is None and engine == "orjson":
+ continue
+
+ result_str = _to_json_plotly(plotly_object, pretty=pretty, engine=engine)
+ results[engine] = from_json_plotly(result_str, engine=engine)
+ timing[engine] = time_engine(engine, plotly_object)
+
+ # Check matches
+ if "orjson" in results:
+ if results["json"] != results["orjson"]:
+ raise ValueError(
+ """
+ {json}
+
+ {orjson}""".format(
+ json=results["json"], orjson=results["orjson"]
+ )
+ )
+
+ # write timing
+ import uuid
+ import pickle
+ import os
+ uid = str(uuid.uuid4())
+ with open("json_timing.csv".format(engine), "at") as f:
+ f.write("{}, {}, {}, {}\n".format(
+ timing["json"], timing["orjson"], len(result_str), uid)
+ )
+ os.makedirs("json_object", exist_ok=True)
+ with open("json_object/{uid}.pkl".format(uid=uid), "wb") as f:
+ pickle.dump(plotly_object, f)
+
+ return result_str
+
+
+def _to_json_plotly(plotly_object, pretty=False, engine=None):
  """
  Convert a plotly/Dash object to a JSON string representation
 
@@ -466,7 +537,7 @@ def clean_to_json_compatible(obj, **kwargs):
  elif obj.dtype.kind == "U":
  return obj.tolist()
  elif obj.dtype.kind == "O":
- # Treat object array as a lists, continue processing
+ # Treat object array as plain list, allow recursive processing below
  obj = obj.tolist()
  elif isinstance(obj, np.datetime64):
  return str(obj)

packages/python/plotly/tox.ini

@@ -75,7 +75,7 @@ deps=
  optional: matplotlib==2.2.3
  optional: scikit-image==0.14.4
  optional: kaleido
-optional: orjson==3.4.6;python_version>"3.5"
+ orjson==3.4.6;python_version>"3.5"
 
 ; CORE ENVIRONMENTS
 [testenv:py27-core]

.circleci/config.yml

@@ -103,7 +103,7 @@ jobs:
  name: Build html figures
  command: |
  . venv/bin/activate
- pip install pandas statsmodels --quiet
+ pip install pandas statsmodels orjson --quiet
  python test/percy/plotly-express.py
  - run:
  name: Run percy snapshots

packages/python/plotly/plotly/basedatatypes.py

@@ -916,6 +916,12 @@ def update(self, dict1=None, overwrite=False, **kwargs):
  BaseFigure._perform_update(self[k], v)
  else:
  self[k] = v
+
+ # # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(self)
+
  return self
 
  def pop(self, key, *args):

packages/python/plotly/plotly/express/_core.py

@@ -2092,6 +2092,12 @@ def make_figure(args, constructor, trace_patch=None, layout_patch=None):
 
  configure_axes(args, constructor, fig, orders)
  configure_animation_controls(args, constructor, fig)
+
+ # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(fig)
+
  return fig
 
 

packages/python/plotly/plotly/io/_json.py

@@ -57,7 +57,78 @@ def coerce_to_strict(const):
  return const
 
 
+def time_engine(engine, plotly_object):
+ import time
+ # time in seconds
+ t_total = 0
+ n_total = 0
+
+ # Call function for at least total of 2 seconds and at least 10 times
+ n_min = 10
+ t_min = 1
+
+ while t_total < t_min or n_total < n_min:
+ t0 = time.perf_counter()
+ _to_json_plotly(plotly_object, engine=engine)
+ t1 = time.perf_counter()
+ n_total += 1
+ t_total += (t1 - t0)
+
+ # return time in ms
+ return 1000 * t_total / n_total
+
+
 def to_json_plotly(plotly_object, pretty=False, engine=None):
+ if engine is not None:
+ return _to_json_plotly(plotly_object, pretty=pretty , engine=engine)
+
+ # instrucment _to_json_plotly by running it with all 3 engines and comparing results
+ # before returnin
+
+ import timeit
+ from IPython import get_ipython
+ ipython = get_ipython()
+ orjson = get_module("orjson", should_load=True)
+ results = {}
+ timing = {}
+ result_str = None
+ for engine in ["json", "orjson"]:
+ if orjson is None and engine == "orjson":
+ continue
+
+ result_str = _to_json_plotly(plotly_object, pretty=pretty, engine=engine)
+ results[engine] = from_json_plotly(result_str, engine=engine)
+ timing[engine] = time_engine(engine, plotly_object)
+
+ # Check matches
+ if "orjson" in results:
+ if results["json"] != results["orjson"]:
+ raise ValueError(
+ """
+ {json}
+
+ {orjson}""".format(
+ json=results["json"], orjson=results["orjson"]
+ )
+ )
+
+ # write timing
+ import uuid
+ import pickle
+ import os
+ uid = str(uuid.uuid4())
+ with open("json_timing.csv".format(engine), "at") as f:
+ f.write("{}, {}, {}, {}\n".format(
+ timing["json"], timing["orjson"], len(result_str), uid)
+ )
+ os.makedirs("json_object", exist_ok=True)
+ with open("json_object/{uid}.pkl".format(uid=uid), "wb") as f:
+ pickle.dump(plotly_object, f)
+
+ return result_str
+
+
+def _to_json_plotly(plotly_object, pretty=False, engine=None):
  """
  Convert a plotly/Dash object to a JSON string representation
 
@@ -466,7 +537,7 @@ def clean_to_json_compatible(obj, **kwargs):
  elif obj.dtype.kind == "U":
  return obj.tolist()
  elif obj.dtype.kind == "O":
- # Treat object array as a lists, continue processing
+ # Treat object array as plain list, allow recursive processing below
  obj = obj.tolist()
  elif isinstance(obj, np.datetime64):
  return str(obj)

packages/python/plotly/tox.ini

@@ -75,7 +75,7 @@ deps=
  optional: matplotlib==2.2.3
  optional: scikit-image==0.14.4
  optional: kaleido
-optional: orjson==3.4.6;python_version>"3.5"
+ orjson==3.4.6;python_version>"3.5"
 
 ; CORE ENVIRONMENTS
 [testenv:py27-core]

.circleci/config.yml

@@ -103,7 +103,7 @@ jobs:
  name: Build html figures
  command: |
  . venv/bin/activate
- pip install pandas statsmodels --quiet
+ pip install pandas statsmodels orjson --quiet
  python test/percy/plotly-express.py
  - run:
  name: Run percy snapshots

packages/python/plotly/plotly/basedatatypes.py

@@ -916,6 +916,12 @@ def update(self, dict1=None, overwrite=False, **kwargs):
  BaseFigure._perform_update(self[k], v)
  else:
  self[k] = v
+
+ # # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(self)
+
  return self
 
  def pop(self, key, *args):

packages/python/plotly/plotly/express/_core.py

@@ -2092,6 +2092,12 @@ def make_figure(args, constructor, trace_patch=None, layout_patch=None):
 
  configure_axes(args, constructor, fig, orders)
  configure_animation_controls(args, constructor, fig)
+
+ # instrumentation
+ # from plotly.io._json import to_json_plotly
+ #
+ # to_json_plotly(fig)
+
  return fig
 
 

packages/python/plotly/plotly/io/_json.py

@@ -57,7 +57,78 @@ def coerce_to_strict(const):
  return const
 
 
+def time_engine(engine, plotly_object):
+ import time
+ # time in seconds
+ t_total = 0
+ n_total = 0
+
+ # Call function for at least total of 2 seconds and at least 10 times
+ n_min = 10
+ t_min = 1
+
+ while t_total < t_min or n_total < n_min:
+ t0 = time.perf_counter()
+ _to_json_plotly(plotly_object, engine=engine)
+ t1 = time.perf_counter()
+ n_total += 1
+ t_total += (t1 - t0)
+
+ # return time in ms
+ return 1000 * t_total / n_total
+
+
 def to_json_plotly(plotly_object, pretty=False, engine=None):
+ if engine is not None:
+ return _to_json_plotly(plotly_object, pretty=pretty , engine=engine)
+
+ # instrucment _to_json_plotly by running it with all 3 engines and comparing results
+ # before returnin
+
+ import timeit
+ from IPython import get_ipython
+ ipython = get_ipython()
+ orjson = get_module("orjson", should_load=True)
+ results = {}
+ timing = {}
+ result_str = None
+ for engine in ["json", "orjson"]:
+ if orjson is None and engine == "orjson":
+ continue
+
+ result_str = _to_json_plotly(plotly_object, pretty=pretty, engine=engine)
+ results[engine] = from_json_plotly(result_str, engine=engine)
+ timing[engine] = time_engine(engine, plotly_object)
+
+ # Check matches
+ if "orjson" in results:
+ if results["json"] != results["orjson"]:
+ raise ValueError(
+ """
+ {json}
+
+ {orjson}""".format(
+ json=results["json"], orjson=results["orjson"]
+ )
+ )
+
+ # write timing
+ import uuid
+ import pickle
+ import os
+ uid = str(uuid.uuid4())
+ with open("json_timing.csv".format(engine), "at") as f:
+ f.write("{}, {}, {}, {}\n".format(
+ timing["json"], timing["orjson"], len(result_str), uid)
+ )
+ os.makedirs("json_object", exist_ok=True)
+ with open("json_object/{uid}.pkl".format(uid=uid), "wb") as f:
+ pickle.dump(plotly_object, f)
+
+ return result_str
+
+
+def _to_json_plotly(plotly_object, pretty=False, engine=None):
  """
  Convert a plotly/Dash object to a JSON string representation
 
@@ -466,7 +537,7 @@ def clean_to_json_compatible(obj, **kwargs):
  elif obj.dtype.kind == "U":
  return obj.tolist()
  elif obj.dtype.kind == "O":
- # Treat object array as a lists, continue processing
+ # Treat object array as plain list, allow recursive processing below
  obj = obj.tolist()
  elif isinstance(obj, np.datetime64):
  return str(obj)

packages/python/plotly/tox.ini

@@ -75,7 +75,7 @@ deps=
  optional: matplotlib==2.2.3
  optional: scikit-image==0.14.4
  optional: kaleido
-optional: orjson==3.4.6;python_version>"3.5"
+ orjson==3.4.6;python_version>"3.5"
 
 ; CORE ENVIRONMENTS
 [testenv:py27-core]