JSON and CSV readers are ready

Author: 0xIbra

.gitignore

@@ -0,0 +1,4 @@
+.mypy*
+data/
+__pycache__/
+*test.py

README.md

@@ -0,0 +1,10 @@
+Data Reader
+===========
+> Python package that **lazily** reads data from large **JSON**, **XML** and **CSV** files.
+
+Why not use the already existing libs you ask ?  
+Well, for me, it was crucial to have a small package size that can be deployed to a **Serverless** environment as this was the main goal.  
+
+Installation
+------------
+    pip install data-reader

data_reader/readers/csv_reader.py

@@ -0,0 +1,53 @@
+import os
+
+
+class CSVReader:
+
+    def __init__(self, file_path, arguments={'separator': ','}):
+        self.__file_path = file_path
+        self.__arguments = arguments
+
+        self.__initialize()
+
+    def __initialize(self):
+        """ Prepares and validates argument. """
+
+        if 'separator' not in self.__arguments:
+            raise Exception('"separator" argument is required to parse CSV.')
+
+        self.__separator = self.__arguments['separator']
+        self.__skip_header = True
+        self.__enclosing = '"'
+
+        if 'skip_header' in self.__arguments:
+            self.__skip_header = self.__arguments['skip_header']
+        if 'enclosing' in self.__arguments:
+            self.__enclosing = self.__arguments['enclosing']
+
+    def iterate(self):
+        """ allows iteration over parsed data objects with a python generator. """
+
+        with open(self.__file_path) as f:
+            for index, line in enumerate(f):
+                if index == 0 and self.__skip_header:
+                    continue
+
+                obj = self.__parseLine(line)
+                yield obj
+
+    def __parseLine(self, line):
+        """ Parses a line into python dictionary object """
+
+        split = line.split(self.__separator)
+        formatted = []
+
+        if self.__enclosing is not None:
+            for entry in split:
+                if entry.startswith(self.__enclosing) and entry.endswith(self.__enclosing):
+                    entry = entry[1:]
+                    entry = entry[:-1]
+                    formatted.append(entry)
+
+        del split
+
+        return formatted

data_reader/readers/json_reader.py

@@ -0,0 +1,22 @@
+import ijson
+import os
+
+
+class JSONReader:
+
+    def __init__(self, file_path, arguments={}):
+        self.__file_path = file_path
+        self.__arguments = arguments
+
+        self.__item_node = 'item'
+        if 'item_node' in arguments:
+            self.__item_node = arguments['item_node']
+
+    def iterate(self):
+        """ Iterates over JSON file stream. """
+
+        with open(self.__file_path, 'rb') as fh:
+            parsed_json = ijson.items(fh, self.__item_node)
+
+            for object in parsed_json:
+                yield object

setup.py

@@ -0,0 +1,29 @@
+import setuptoolsSo i had no choice, this package is
+
+
+
+long_desc = ''
+with open('README.md', 'r') as f:
+    long_desc = f.read()
+
+setuptools.setup(
+    name='data-reader',
+    version='0.0.1',
+    author='Ibragim Abubakarov',
+    author_email='ibragim.ai95@gmail.com',
+    maintainer='Ibragim Abubakarov',
+    maintainer_email='ibragim.ai95@gmail.com',
+    description='Python package that allows easy importation of data from large files like CSV, JSON and XML.',
+    long_description=long_desc,
+    long_description_content_type='text/markdown',
+    url='https://github.com/polkovnik-z/data-reader',
+    packages=[
+        'data_reader'
+    ],
+    install_requires=['pandas', 'imperium', 'ijson'],
+    classifiers=[
+        'Programming Language :: Python :: 3',
+        'Operating System :: OS Independent',
+        'Intended Audience :: Developers'
+    ]
+)