Initial Commit

LICENSE

@@ -0,0 +1,7 @@
+Copyright © 2020 Colgate-Palmolive
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

README.md

@@ -0,0 +1,5 @@
+# Data Engineering Container Tools
+
+This is a package containing tools for data engineering containers. Use
+[Github-flavored Markdown](https://guides.github.com/features/mastering-markdown/)
+to write more content.

build/lib/dataEng-container-tools/__init__.py

@@ -0,0 +1,4 @@
+from .safe_stdout import setup_stdout
+from .cla import default_gcs_secret_location
+
+setup_stdout(default_gcs_secret_location)

build/lib/dataEng-container-tools/cla.py

@@ -0,0 +1,171 @@
+import argparse
+import json
+import sys
+from enum import Enum
+from .safe_stdout import setup_stdout
+
+default_gcs_secret_location = '/vault/secrets/gcp-sa-storage.json'
+
+class custom_command_line_argument:
+    """Class for creating custom command line arguments.
+       Takes in all arguments needed for 'command_line_arguments'
+       to construct an additional custom argument."""
+    name = None
+    action = None
+    nargs = None
+    const = None
+    default = None
+    data_type = None
+    choices = None
+    required = None
+    help_message = None
+    metavar = None
+    dest = None
+    def __init__(self, name, action = None, nargs = None, const = None, default = None,
+                data_type = None, choices = None, required = None, help_message = None,
+                metavar = None, dest = None):
+        self.name = name
+        self.action = action
+        self.nargs = nargs
+        self.const = const
+        self.default = default
+        self.data_type = data_type
+        self.choices = choices
+        self.required = required
+        self.help_message = help_message
+        self.metavar = metavar
+        self.dest = dest
+
+    def __str__(self):
+        return ("name: " + self.name + ", " +
+        "action: " + self.action + ", " +
+        "nargs: " + self.nargs + ", " +
+        "const: " + self.const + ", " +
+        "default: " + self.default + ", " +
+        "data_type: " + self.data_type + ", " +
+        "choices: " + self.choices + ", " +
+        "required: " + self.required + ", " +
+        "help_message: " + self.help_message + ", " +
+        "metavar: " + self.metavar + ", " +
+        "dest: " + self.dest)
+
+class command_line_argument_type(Enum):
+    """Enum type for dictating whether fields in 'command_line_arguments' class are OPTIONAL or REQUIRED."""
+    OPTIONAL = False
+    REQUIRED = True
+
+class command_line_arguments:
+    """Simplified process for creating command line arguments. Allows for custom CLAs."""
+    __args = None
+    __input_files = None
+    __output_files = None
+    __secret_location = None
+    __default_file_type = None
+    __custom_inputs = None
+    __description = None
+    __input_dtypes = None
+    __default_secret_location = default_gcs_secret_location
+    def __init__(self, input_files = None, output_files = None, secret_location = None,
+                default_file_type = None, custom_inputs = None, description = None,
+                input_dtypes = None, parser = None):
+        self.__input_files = input_files
+        self.__output_files = output_files
+        self.__secret_location = secret_location
+        self.__default_file_type = default_file_type
+        self.__custom_inputs = custom_inputs
+        self.__description = description
+        self.__input_dtypes = input_dtypes
+        parser = parser if parser else argparse.ArgumentParser(description=description)
+        if input_files:
+            parser.add_argument("--input_bucket_names", type=str, required=input_files.value,
+                                nargs = '+', help="GCS Buckets to read from.")
+
+            parser.add_argument("--input_paths", type=str, required=input_files.value,
+                                nargs = '+', help="GCS folders in bucket to read file from.")
+
+            parser.add_argument("--input_filenames", type=str, required=input_files.value,
+                                nargs = '+', help="Filenames to read file from.")
+            if input_dtypes:
+                parser.add_argument("--input_dtypes", type=json.loads, required=input_dtypes.value,
+                                nargs ='+', help = "JSON dictionaries of (column: type) pairs to cast columns to")
+
+        if output_files:
+            parser.add_argument("--output_bucket_names", type=str, required=output_files.value,
+                                nargs = '+', help="GCS Bucket to write to.")
+
+            parser.add_argument("--output_paths", type=str, required=output_files.value,
+                                nargs = '+', help="GCS folder in bucket to write file to.")
+
+            parser.add_argument("--output_filenames", type=str, required=output_files.value,
+                                nargs = '+', help="Filename to write file to.")
+        if secret_location:
+            parser.add_argument("--gcs_secret_location", type = str, required=secret_location.value,
+                                default = self.__default_secret_location,
+                                help = "Location of GCS secret injected by Vault. Default: '" + self.__default_secret_location + "'.")
+        if default_file_type:
+            parser.add_argument("--default_file_type", type = str,required=default_file_type.value,
+                                choices = ["parquet", "csv", "pkl", "json"], default = "parquet",
+                                help = "How to handle input/output files if no file extension found. Choice of 'parquet', 'csv', 'pkl', and 'json'. Default 'parquet'.")
+
+        if custom_inputs:
+            for item in custom_inputs:
+                parser.add_argument(name = "--"+item.name, action = item.action, nargs = item.nargs,
+                                        const = item.const, default = item.default, type = item.data_type,
+                                        choices = item.choices, required = item.required,
+                                        help = item.help_message, metavar = item.metavar,
+                                        dest = item.dest)
+        self.__args = parser.parse_args()
+        print("CLA Input:", self)
+        self.check_args()
+        if self.__secret_location and (self.__args.gcs_secret_location != self.__default_secret_location):
+            setup_stdout(self.__args.gcs_secret_location)
+
+    def __str__(self):
+        return self.__args.__str__()
+
+    def get_arguments(self):
+        return self.__args
+
+    def get_input_dtypes(self):
+        if not self.__input_dtypes:
+            return None
+        return self.__args.input_dtypes
+
+    def get_input_uris(self):
+        if not self.__input_files:
+            return []
+        constant_bucket = False
+        bucket_name = ''
+        output = []
+        if len(self.__args.input_bucket_names) == 1:
+            constant_bucket = True
+            bucket_name = self.__args.input_bucket_names[0]
+        for pos, filename in enumerate(self.__args.input_filenames):
+            if not constant_bucket:
+                bucket_name = self.__args.input_bucket_names[pos]
+            output.append("gs://"+bucket_name+"/"+self.__args.input_paths[pos]+"/"+filename)
+        return output
+
+    def get_output_uris(self):
+        if not self.__output_files:
+            return []
+        constant_bucket = False
+        bucket_name = ''
+        output = []
+        if len(self.__args.output_bucket_names) == 1:
+            constant_bucket = True
+            bucket_name = self.__args.output_bucket_names[0]
+        for pos, filename in enumerate(self.__args.output_filenames):
+            if not constant_bucket:
+                bucket_name = self.__args.output_bucket_names[pos]
+            output.append("gs://"+bucket_name+"/"+self.__args.output_paths[pos]+"/"+filename)
+        return output
+
+    def get_secret_location(self):
+        if not self.__secret_location:
+            return None
+        return self.__args.gcs_secret_location
+
+    def check_args(self):
+        #TODO: Implement this
+        return

build/lib/dataEng-container-tools/gcs.py

@@ -0,0 +1,120 @@
+import json
+import pandas as pd
+import io
+from google.cloud import storage
+import pickle
+
+class gcs_file_io:
+    gcs_client = None
+    gcs_secret_location = None
+
+    def __init__(self, gcs_secret_location):
+        self.gcs_secret_location = gcs_secret_location
+        with open(gcs_secret_location,'r') as f:
+            gcs_sa = json.load(f)
+        with open('gcs-sa.json', 'w') as json_file:
+            json.dump(gcs_sa, json_file)
+        self.gcs_client = storage.Client.from_service_account_json('gcs-sa.json')
+
+    def __get_parts(self, gcs_uri):
+        if gcs_uri.startswith('gs://'):
+            uri = gcs_uri[5:]
+        bucket = uri[:uri.find("/")]
+        file_path = uri[uri.find("/")+1:]
+        return bucket, file_path
+
+    def download_file_to_object(self, gcs_uri, default_file_type = None, dtype = None):
+        bucket_name, file_path = self.__get_parts(gcs_uri)
+        bucket = self.gcs_client.bucket(bucket_name)
+        binary_object = bucket.get_blob(file_path).download_as_string()
+        file_like_object = io.BytesIO(binary_object)
+        hasEnding = file_path.endswith('.parquet') or file_path.endswith('.csv') or file_path.endswith('.pkl')
+        if file_path.endswith('.parquet') or ((not hasEnding) and (default_file_type == 'parquet')):
+            return pd.read_parquet(file_like_object, dtype = dtype) if dtype else pd.read_parquet(file_like_object)
+        if file_path.endswith('.csv') or ((not hasEnding) and (default_file_type == 'csv')):
+            return pd.read_csv(file_like_object, dtype = dtype) if dtype else pd.read_csv(file_like_object)
+        if file_path.endswith('.pkl') or ((not hasEnding) and (default_file_type == 'pkl')):
+            return pd.read_pickle(file_like_object) if dtype else pd.read_pickle(file_like_object)#, dtype = dtype
+        if file_path.endswith('.json') or ((not hasEnding) and (default_file_type == 'json')):
+            return json.load(file_like_object)
+        return file_like_object
+
+    def download_files_to_objects(self, gcs_uris, default_file_type = None, dtypes = None):
+        dtypes_status = 0
+        if dtypes:
+            dtypes_status +=1
+            if len(dtypes)>1:
+                dtypes_status +=1
+        return_objects = []
+        for pos, gcs_uri in enumerate(gcs_uris):
+            if dtypes_status == 0:
+                return_objects.append(self.download_file_to_object(gcs_uri, default_file_type= default_file_type))
+            elif dtypes_status == 1:
+                return_objects.append(self.download_file_to_object(gcs_uri,
+                                                                    default_file_type= default_file_type,
+                                                                    dtype = dtypes[0]))
+            else:
+                return_objects.append(self.download_file_to_object(gcs_uri,
+                                                                    default_file_type= default_file_type,
+                                                                    dtype = dtypes[pos]))
+        return return_objects
+
+    def download_file_to_disk(self, gcs_uri, local_location = None):
+        bucket_name, file_path = self.__get_parts(gcs_uri)
+        print("Bucket:", bucket_name, "File:", file_path)
+        print(self.gcs_client)
+        bucket = self.gcs_client.bucket(bucket_name)
+        print("1")
+        local_location = local_location if local_location else file_path
+        print("2")
+        bucket.get_blob(file_path).download_to_filename(local_location)
+        print("3")
+        return local_location
+
+    def download_files_to_disk(self, gcs_uris, local_locations = None):
+        return_locations = []
+        for pos, gcs_uri in enumerate(gcs_uris):
+            return_locations.append(self.download_file_to_disk(gcs_uri = gcs_uri, local_location = local_locations[pos]))
+        return return_locations
+
+    def upload_file_from_disk(self, gcs_uri, local_location):
+        bucket_name, file_path = self.__get_parts(gcs_uri)
+        bucket = self.gcs_client.bucket(bucket_name)
+        blob = bucket.blob(file_path)
+        return blob.upload_from_filename(local_location)
+
+    def upload_files_from_disk(self, gcs_uris, local_locations):
+        return_objects = []
+        for pos, gcs_uri in enumerate(gcs_uris):
+            return_objects.append(self.upload_file_from_disk(gcs_uri, local_locations[pos]))
+        return return_objects
+
+    def upload_file_from_object(self, gcs_uri, object_to_upload, default_file_type = None):
+        bucket_name, file_path = self.__get_parts(gcs_uri)
+        bucket = self.gcs_client.bucket(bucket_name)
+        blob = bucket.blob(file_path)
+        hasEnding = file_path.endswith('.parquet') or file_path.endswith('.csv') or file_path.endswith('.pkl')
+        if file_path.endswith('.parquet') or ((not hasEnding) and (default_file_type == 'parquet')):
+            fileObject = io.BytesIO()
+            object_to_upload.to_parquet(fileObject)
+            fileObject.seek(0)
+            return blob.upload_from_file(fileObject)
+        if file_path.endswith('.csv') or ((not hasEnding) and (default_file_type == 'csv')):
+            csv_string = object_to_upload.to_csv(encoding='utf-8')
+            return blob.upload_from_string(csv_string)
+        if file_path.endswith('.pkl') or ((not hasEnding) and (default_file_type == 'pkl')):
+            fileObject = io.BytesIO(pickle.dumps(object_to_upload))
+            fileObject.seek(0)
+            return blob.upload_from_file(fileObject)
+        if file_path.endswith('.json') or ((not hasEnding) and (default_file_type == 'json')):
+            json_string = json.dumps(fileObject)
+            return blob.upload_from_string(json_string)
+        fileObject = io.BytesIO(pickle.dumps(object_to_upload))
+        fileObject.seek(0)
+        return blob.upload_from_file(fileObject)
+
+    def upload_files_from_objects(self, gcs_uris, objects_to_upload, default_file_type = None):
+        return_objects = []
+        for pos, gcs_uri in enumerate(gcs_uris):
+            return_objects.append(self.upload_file_from_object(gcs_uri, objects_to_upload[pos], default_file_type = default_file_type))
+        return return_objects

build/lib/dataEng-container-tools/safe_stdout.py

@@ -0,0 +1,36 @@
+import sys
+import json
+
+class safe_stdout:
+    def __init__(self, bad_words):
+        self.__bad_words = bad_words
+        self.__bad_word_lengths = [len(bad_word) for bad_word in bad_words]
+        self.__old_stdout = sys.stdout
+
+    def write(self, message):
+        for location, bad_word in enumerate(self.__bad_words):
+            bad_word_length = self.__bad_word_lengths[location]
+            bad_word_location = message.find(bad_word)
+            while(bad_word_location != -1):
+                message = (message[0:bad_word_location] + '*'*bad_word_length +
+                           message[bad_word_location + bad_word_length:])
+                bad_word_location = message.find(bad_word)
+        self.__old_stdout.write(message)
+
+    def flush(self):
+        pass
+
+def setup_stdout(secret_location):
+        secret = {}
+        try:
+            secret = json.load(open(secret_location,'r'))
+        except OSError:
+            print("No secret file found")
+            return
+        bad_words = list(secret.values())
+        dumped_words = []
+        for word in bad_words:
+            dumped_words.append(json.dumps(word))
+        bad_words += dumped_words
+        #print("\nBad words:", bad_words)
+        sys.stdout = safe_stdout(bad_words)

dataEng-container-tools/__init__.py

@@ -0,0 +1,4 @@
+from .safe_stdout import setup_stdout
+from .cla import default_gcs_secret_location
+
+setup_stdout(default_gcs_secret_location)