change: ignore code cells with shell commands in v2 migration tool (#1801)

Author: laurenyu

doc/v2.rst

@@ -407,6 +407,12 @@ You can apply it to a set of files using a loop:
 Limitations
 ===========
 
+Jupyter Notebook Cells with Shell Commands
+------------------------------------------
+
+If your Jupyter notebook has a code cell with lines that start with either ``%%`` or ``!``, the tool ignores that cell.
+The other cells in the notebook are still updated.
+
 Aliased Imports
 ---------------
 

src/sagemaker/cli/compatibility/v2/files.py

@@ -121,12 +121,30 @@ def update(self):
         """
         nb_json = self._read_input_file()
         for cell in nb_json["cells"]:
-            if cell["cell_type"] == "code":
+            if cell["cell_type"] == "code" and not self._contains_shell_cmds(cell):
                 updated_source = self._update_code_from_cell(cell)
                 cell["source"] = updated_source
 
         self._write_output_file(nb_json)
 
+    def _contains_shell_cmds(self, cell):
+        """Checks if the cell's source uses either ``%%`` or ``!`` to execute shell commands.
+
+        Args:
+            cell (dict): A dictionary representation of a code cell from
+                a Jupyter notebook. For more info, see
+                https://ipython.org/ipython-doc/dev/notebook/nbformat.html#code-cells.
+
+        Returns:
+            bool: If the first line starts with ``%%`` or any line starts with ``!``.
+        """
+        source = cell["source"]
+
+        if source[0].startswith("%%"):
+            return True
+
+        return any(line.startswith("!") for line in source)
+
     def _update_code_from_cell(self, cell):
         """Updates the code from a code cell so that it is
         compatible with version 2.0 and later of the SageMaker Python SDK.

tests/unit/sagemaker/cli/compatibility/v2/test_file_updaters.py

@@ -95,6 +95,25 @@ def test_update(ast_transformer, pasta_parse, pasta_dump, json_dump):
           "execution_count": 1,
           "metadata": {},
           "outputs": [],
+          "source": [
+           "!echo ignore this"
+          ]
+         },
+         {
+          "cell_type": "code",
+          "execution_count": 2,
+          "metadata": {},
+          "outputs": [],
+          "source": [
+           "%%%%bash\\n",
+           "echo ignore this too"
+          ]
+         },
+         {
+          "cell_type": "code",
+          "execution_count": 3,
+          "metadata": {},
+          "outputs": [],
           "source": [
            "# code to be modified\\n",
            "%s"