Testing notebooks + fixes

popmon/hist/histogram.py

@@ -211,7 +211,7 @@ def __repr__(self):
         return f"HistogramContainer(dtype={self.npdtype}, n_dims={self.n_dim})"
 
     def __str__(self):
-        return str(self)
+        return repr(self)
 
     def _edit_name(self, axis_name, xname, yname, convert_time_index, short_keys):
         if convert_time_index and self.is_ts:

popmon/notebooks/popmon_tutorial_advanced.ipynb

@@ -4,7 +4,6 @@
    "cell_type": "code",
    "execution_count": null,
    "metadata": {
-    "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
     },
@@ -26,10 +25,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "%%capture\n",
     "# install popmon (if not installed yet)\n",
     "import sys\n",
     "\n",
-    "!{sys.executable} -m pip install popmon"
+    "!\"{sys.executable}\" -m pip install popmon"
    ]
   },
   {
@@ -145,11 +145,13 @@
    "outputs": [],
    "source": [
     "# download histogrammar jar files if not already installed, used for histogramming of spark dataframe\n",
-    "from pyspark.sql import SparkSession\n",
+    "try:\n",
+    "    from pyspark.sql import SparkSession\n",
     "\n",
-    "spark = SparkSession.builder.config(\n",
-    "    \"spark.jars.packages\", \"org.diana-hep:histogrammar-sparksql_2.11:1.0.4\"\n",
-    ").getOrCreate()"
+    "    pyspark_installed = True\n",
+    "except ImportError:\n",
+    "    print(\"pyspark needs to be installed for this example\")\n",
+    "    pyspark_installed = False"
    ]
   },
   {
@@ -158,18 +160,19 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "sdf = spark.createDataFrame(df)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "sdf.pm_stability_report(\n",
-    "    time_axis=\"DATE\", time_width=\"1w\", time_offset=\"2015-07-02\", extended_report=False\n",
-    ")"
+    "if pyspark_installed:\n",
+    "    spark = SparkSession.builder.config(\n",
+    "        \"spark.jars.packages\", \"org.diana-hep:histogrammar-sparksql_2.11:1.0.4\"\n",
+    "    ).getOrCreate()\n",
+    "\n",
+    "    sdf = spark.createDataFrame(df)\n",
+    "\n",
+    "    sdf.pm_stability_report(\n",
+    "        time_axis=\"DATE\",\n",
+    "        time_width=\"1w\",\n",
+    "        time_offset=\"2015-07-02\",\n",
+    "        extended_report=False,\n",
+    "    )"
    ]
   },
   {
@@ -287,7 +290,7 @@
    "outputs": [],
    "source": [
     "split_hist = split_hists.query(\"date == '2015-07-05 12:00:00'\")\n",
-    "split_hist.histogram[0].hist.plot.matplotlib();"
+    "split_hist.histogram[0].hist.plot.matplotlib()"
    ]
   },
   {
@@ -303,7 +306,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "split_hist.histogram_ref[0].hist.plot.matplotlib();"
+    "split_hist.histogram_ref[0].hist.plot.matplotlib()"
    ]
   },
   {
@@ -320,11 +323,14 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "import pickle\n",
+    "# As HTML report\n",
+    "report.to_file(\"report.html\")\n",
+    "\n",
+    "# Alternatively, as serialized Python object\n",
+    "# import pickle\n",
     "\n",
-    "with open(\"report.pkl\", \"wb\") as f:\n",
-    "    pickle.dump(report, f)\n",
-    "report.to_file(\"report.html\")"
+    "# with open(\"report.pkl\", \"wb\") as f:\n",
+    "#     pickle.dump(report, f)"
    ]
   },
   {
@@ -350,7 +356,7 @@
     "    report_filepath=None,\n",
     "    store_key=\"html_report\",\n",
     "    sections_key=\"report_sections\",\n",
-    ")\n",
+    ")"
    ]
   },
   {
@@ -473,18 +479,18 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.7"
+   "version": "3.8.6"
   },
   "nteract": {
    "version": "0.15.0"
   },
   "pycharm": {
    "stem_cell": {
     "cell_type": "raw",
-    "source": [],
     "metadata": {
      "collapsed": false
-    }
+    },
+    "source": []
    }
   }
  },

popmon/notebooks/popmon_tutorial_basic.ipynb

@@ -4,7 +4,6 @@
    "cell_type": "code",
    "execution_count": null,
    "metadata": {
-    "collapsed": false,
     "jupyter": {
      "outputs_hidden": false
     },
@@ -36,10 +35,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "%%capture\n",
     "# install popmon (if not installed yet)\n",
     "import sys\n",
     "\n",
-    "!{sys.executable} -m pip install popmon"
+    "!\"{sys.executable}\" -m pip install popmon"
    ]
   },
   {

popmon/notebooks/popmon_tutorial_incremental_data.ipynb

@@ -28,10 +28,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "%%capture\n",
     "# install popmon (if not installed yet)\n",
     "import sys\n",
     "\n",
-    "!{sys.executable} -m pip install popmon"
+    "!\"{sys.executable}\" -m pip install popmon"
    ]
   },
   {

requirements-test.txt

@@ -1,7 +1,8 @@
 flake8>=3.7.8
 pytest>=4.0.2
-nbconvert>=5.3.1
+pytest-notebook>=0.6.1
+nbconvert~=5.6.1
 jupyter_client>=5.2.3
 ipykernel>=5.1.3
 black>=19.10b0
-isort>=5.0.7
+isort>=5.0.7

tests/popmon/notebooks/test_notebooks.py

@@ -1,55 +1,30 @@
-import os
-import unittest
-
-import nbformat
 import pytest
-from jupyter_client.kernelspec import KernelSpecManager, NoSuchKernel
-from nbconvert.preprocessors import ExecutePreprocessor
-from nbconvert.preprocessors.execute import CellExecutionError
-
-from popmon import resources
-
-kernel_name = "python3"
-
-# check if jupyter python3 kernel can be opened. if kernel not found, skip unit tests below.
-try:
-    km = KernelSpecManager()
-    km.get_kernel_spec(kernel_name)
-    kernel_found = True
-except NoSuchKernel:
-    kernel_found = False
+from pytest_notebook.nb_regression import NBRegressionFixture
 
+from popmon.resources import notebook
 
-class NotebookTest(unittest.TestCase):
-    """Unit test notebook"""
 
-    def run_notebook(self, notebook):
-        """ Test notebook """
+@pytest.fixture(scope="module")
+def nb_tester():
+    """Test notebooks using pytest-notebook"""
+    nb_regression = NBRegressionFixture(
+        diff_ignore=(
+            "/metadata/language_info",
+            "/cells/*/execution_count",
+            "/cells/*/outputs/*",
+        ),
+        exec_timeout=1800,
+    )
+    return nb_regression
 
-        # load notebook
-        with open(notebook) as f:
-            nb = nbformat.read(f, as_version=4)
 
-        # execute notebook
-        ep = ExecutePreprocessor(timeout=600, kernel_name=kernel_name)
-        try:
-            ep.preprocess(nb, {})
-            status = True
-        except CellExecutionError:
-            # store if failed
-            status = False
-            executed_notebook = os.getcwd() + "/" + notebook.split("/")[-1]
-            with open(executed_notebook, mode="wt") as f:
-                nbformat.write(nb, f)
+def test_notebook_basic(nb_tester):
+    nb_tester.check(notebook("popmon_tutorial_basic.ipynb"))
 
-        # check status
-        self.assertTrue(status, "Notebook execution failed (%s)" % notebook)
 
+def test_notebook_advanced(nb_tester):
+    nb_tester.check(notebook("popmon_tutorial_advanced.ipynb"))
 
-@pytest.mark.filterwarnings("ignore:Session._key_changed is deprecated")
-@pytest.mark.skipif(not kernel_found, reason=f"{kernel_name} kernel not found.")
-class PipelineNotebookTest(NotebookTest):
-    """Unit test notebook"""
 
-    def test_basic_tutorial(self):
-        self.run_notebook(resources.notebook("popmon_tutorial_basic.ipynb"))
+def test_notebook_incremental_data(nb_tester):
+    nb_tester.check(notebook("popmon_tutorial_incremental_data.ipynb"))