Merge pull request #2 from awslabs/feature/move-templates

Exclude __pycache__ when using init

pyproject.toml

@@ -11,6 +11,7 @@ python = "^3.9"
 click = "^7.1.2"
 boto3 = "^1.26.6"
 pytest-cov = "^4.0.0"
+pyyaml = "^6.0"
 
 [tool.poetry.group.dev.dependencies]
 pytest = "7.2.2"

src/emr_cli/packaging/python_project.py

@@ -1,13 +1,13 @@
 import os
-import shutil
 import subprocess
 import sys
 from pathlib import Path
+from shutil import copy
 
 import boto3
 
 from emr_cli.deployments.emr_serverless import DeploymentPackage
-from emr_cli.utils import console_log, parse_bucket_uri
+from emr_cli.utils import console_log, copy_template, parse_bucket_uri
 
 
 class PythonProject(DeploymentPackage):
@@ -19,7 +19,7 @@ def initialize(self, target_dir: str = os.getcwd()):
         - Creates a Dockerfile
         """
         console_log(f"Initializing project in {target_dir}")
-        self._copy_template(target_dir)
+        copy_template("pyspark", target_dir)
         console_log("Project initialized.")
 
     def copy_single_file(self, relative_file_path: str, target_dir: str = os.getcwd()):
@@ -30,13 +30,7 @@ def copy_single_file(self, relative_file_path: str, target_dir: str = os.getcwd(
             Path(__file__).parent.parent / "templates" / "pyspark" / relative_file_path
         )
         target_path = Path(target_dir)
-        shutil.copy(template_path, target_path)
-
-    def _copy_template(self, target_dir: str):
-        source = os.path.abspath(
-            os.path.join(os.path.dirname(__file__), "..", "templates", "pyspark")
-        )
-        shutil.copytree(source, target_dir, dirs_exist_ok=True)
+        copy(template_path, target_path)
 
     def build(self):
         """

src/emr_cli/utils/__init__.py

@@ -1,5 +1,6 @@
 import os
-import shutil
+from pathlib import Path
+from shutil import copytree, ignore_patterns
 from typing import List
 from urllib.parse import urlparse
 
@@ -31,7 +32,10 @@ def mkdir(path: str):
 
 
 def copy_template(source: str, target_dir: str):
-    source = os.path.abspath(
-        os.path.join(os.path.dirname(__file__), "..", "templates", source)
+    """
+    Copies the entire `source` directory to `target_dir`.
+    """
+    source = os.path.abspath(Path(__file__).parent.parent / "templates" / source)
+    copytree(
+        source, target_dir, dirs_exist_ok=True, ignore=ignore_patterns("__pycache__")
     )
-    shutil.copytree(source, target_dir, dirs_exist_ok=True)

tests/test_init.py

@@ -0,0 +1,21 @@
+from emr_cli.packaging.python_poetry_project import PythonPoetryProject
+from emr_cli.packaging.python_project import PythonProject
+
+
+class TestInit:
+    def test_default_init(self, tmp_path):
+        p = PythonProject()
+
+        p.initialize(tmp_path)
+        assert (tmp_path / "pyproject.toml").exists()
+        assert (tmp_path / "entrypoint.py").exists()
+        assert (tmp_path / "jobs" / "extreme_weather.py").exists()
+        assert not (tmp_path / "README.md").exists()
+
+    def test_poetry_init(self, tmp_path):
+        p = PythonPoetryProject()
+
+        p.initialize(tmp_path)
+        assert (tmp_path / "entrypoint.py").exists()
+        assert (tmp_path / "pyproject.toml").exists()
+        assert (tmp_path / "README.md").exists()