From af9f275716113fa7d6a1e9b75be77ea5438a558d Mon Sep 17 00:00:00 2001
From: Taylor Turner <taylorfturner@gmail.com>
Date: Wed, 5 Jun 2024 11:59:52 -0500
Subject: [PATCH] pre-commit fix (#1122)

---
 .pre-commit-config.yaml                  |  2 +-
 dataprofiler/__init__.py                 | 16 ++++++++++
 dataprofiler/tests/test_data_profiler.py | 40 ++++++++++++++++++++++++
 requirements.txt                         |  2 +-
 4 files changed, 58 insertions(+), 2 deletions(-)

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index f36c52663..7baeb59ec 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -55,7 +55,7 @@ repos:
             pyarrow>=1.0.1,
             chardet>=3.0.4,
             fastavro>=1.0.0.post1,
-            cramjam>=2.7.0,
+            python-snappy>=0.5.4,
             charset-normalizer>=1.3.6,
             psutil>=4.0.0,
             scipy>=1.4.1,
diff --git a/dataprofiler/__init__.py b/dataprofiler/__init__.py
index 5f218bd85..2e89d3e2b 100644
--- a/dataprofiler/__init__.py
+++ b/dataprofiler/__init__.py
@@ -20,6 +20,22 @@
 from .validators.base_validators import Validator
 from .version import __version__
 
+try:
+    import snappy
+except ImportError:
+    import warnings
+
+    warnings.warn(
+        "Snappy must be installed to use parquet/avro datasets."
+        "\n\n"
+        "For macOS use Homebrew:\n"
+        "\t`brew install snappy`"
+        "\n\n"
+        "For linux use apt-get:\n`"
+        "\tsudo apt-get -y install libsnappy-dev`\n",
+        ImportWarning,
+    )
+
 
 def set_seed(seed=None):
     # also check it's an integer
diff --git a/dataprofiler/tests/test_data_profiler.py b/dataprofiler/tests/test_data_profiler.py
index 9ebdfa039..ef7664cea 100644
--- a/dataprofiler/tests/test_data_profiler.py
+++ b/dataprofiler/tests/test_data_profiler.py
@@ -56,6 +56,46 @@ def test_data_profiling(self):
             self.assertIsNotNone(profile.profile)
             self.assertIsNotNone(profile.report())
 
+    def test_no_snappy(self):
+        import importlib
+        import sys
+        import types
+
+        orig_import = __import__
+        # necessary for any wrapper around the library to test if snappy caught
+        # as an issue
+
+        def reload_data_profiler():
+            """Recursively reload modules."""
+            sys_modules = sys.modules.copy()
+            for module_name, module in sys_modules.items():
+                # Only reload top level of the dataprofiler
+                if "dataprofiler" in module_name and len(module_name.split(".")) < 3:
+                    if isinstance(module, types.ModuleType):
+                        importlib.reload(module)
+
+        def import_mock(name, *args, **kwargs):
+            if name == "snappy":
+                raise ImportError("test")
+            return orig_import(name, *args, **kwargs)
+
+        with mock.patch("builtins.__import__", side_effect=import_mock):
+            with self.assertWarns(ImportWarning) as w:
+                import dataprofiler
+
+                reload_data_profiler()
+
+        self.assertEqual(
+            str(w.warning),
+            "Snappy must be installed to use parquet/avro datasets."
+            "\n\n"
+            "For macOS use Homebrew:\n"
+            "\t`brew install snappy`"
+            "\n\n"
+            "For linux use apt-get:\n`"
+            "\tsudo apt-get -y install libsnappy-dev`\n",
+        )
+
     def test_no_tensorflow(self):
         import sys
 
diff --git a/requirements.txt b/requirements.txt
index 405f808b3..a45dc34ae 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -7,7 +7,7 @@ pytz>=2020.1
 pyarrow>=1.0.1
 chardet>=3.0.4
 fastavro>=1.0.0.post1
-cramjam>=2.7.0
+python-snappy>=0.5.4
 charset-normalizer>=1.3.6
 psutil>=4.0.0
 scipy>=1.10.0