Add simple CLI utility by @tribals

README.rst

@@ -186,10 +186,45 @@ The `Parker`_ convention absorbs the root element by default.
 Installation
 ------------
 
-This is a pure-Python package built for Python 2.6+ and Python 3.0+. To set up::
+This is a pure-Python package built for Python 2.7+ and Python 3.0+. To set up::
 
     pip install xmljson
 
+
+Simple CLI utility
+------------------
+
+After installation, you can benefit from using this package as simple CLI utility. By now only XML to JSON conversion supported. Example::
+
+    $ python -m xmljson -h
+    usage: xmljson [-h] [-o OUT_FILE]
+                [-d {abdera,badgerfish,cobra,gdata,parker,xmldata,yahoo}]
+                [in_file]
+
+    positional arguments:
+    in_file               defaults to stdin
+
+    optional arguments:
+    -h, --help            show this help message and exit
+    -o OUT_FILE, --out_file OUT_FILE
+                            defaults to stdout
+    -d {abdera,badgerfish,...}, --dialect {...}
+                            defaults to parker
+
+    $ python -m xmljson -d parker tests/mydata.xml
+    {
+      "foo": "spam",
+      "bar": 42
+    }
+
+This is a typical UNIX filter program: it reads file (or ``stdin``), processes it in some way (convert XML to JSON in this case), then prints it to ``stdout`` (or file). Example with pipe::
+
+    $ some-xml-producer | python -m xmljson | some-json-processor
+
+There is also ``pip``'s ``console_script`` entry-point, you can call this utility as ``xml2json``::
+
+    $ xml2json -d abdera mydata.xml
+
 Roadmap
 -------
 

setup.py

@@ -1,17 +1,17 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
-
+import io
 
 try:
     from setuptools import setup
 except ImportError:
     from distutils.core import setup
 import xmljson
 
-with open('README.rst') as readme_file:
+with io.open('README.rst', encoding='utf-8') as readme_file:
     readme = readme_file.read()
 
-with open('HISTORY.rst') as history_file:
+with io.open('HISTORY.rst', encoding='utf-8') as history_file:
     history = history_file.read().replace('.. :changelog:', '')
 
 setup(
@@ -46,4 +46,9 @@
     ],
     test_suite='tests',
     tests_require=['lxml'],
+    entry_points={
+        'console_scripts': [
+            'xml2json = xmljson.__main__:main'
+        ]
+    }
 )

tests/test_xmljson.py

@@ -21,23 +21,53 @@
 import lxml.etree
 import xml.etree.cElementTree
 import xmljson
+from xmljson.__main__ import main, parse, closing
 
 _folder = os.path.dirname(os.path.abspath(__file__))
 
 # For Python 3, decode byte strings as UTF-8
 if sys.version_info[0] == 3:
     def decode(s):
         return s.decode('utf-8')
+
+    def openwrite(path):
+        return io.open(path, 'w', encoding='utf-8')
+
 elif sys.version_info[0] == 2:
     def decode(s):
         return s
 
+    def openwrite(path):
+        return io.open(path, 'wb')
+
 
 def read(path):
     with io.open(os.path.join(_folder, path), 'r', encoding='utf-8') as handle:
         return handle.read()
 
 
+class TestCLI(unittest.TestCase):
+    tmp = os.path.join(_folder, 'delete-output.json')
+
+    def test_cli(self):
+        dialects = [xmljson.Abdera(), xmljson.BadgerFish(), xmljson.Cobra(),
+                    xmljson.GData(), xmljson.Parker(), xmljson.Yahoo()]
+
+        for dialect in dialects:
+            for path in ['abdera-1.xml', 'abdera-2.xml', 'abdera-3.xml', 'abdera-4.xml']:
+                in_file = io.open(os.path.join(_folder, path), encoding='utf-8')
+                out_file = openwrite(self.tmp)
+                main(in_file, out_file, dialect)
+                in_file = io.open(os.path.join(_folder, path), encoding='utf-8')
+                out_file = io.open(self.tmp, encoding='utf-8')
+                with closing(in_file), closing(out_file):
+                    self.assertEqual(json.load(out_file), dialect.data(parse(in_file).getroot()))
+
+    def tearDown(self):
+        if os.path.exists(self.tmp):
+            os.remove(self.tmp)
+
+
 class TestXmlJson(unittest.TestCase):
 
     def setUp(self):

xmljson/__main__.py

@@ -0,0 +1,43 @@
+import sys
+import json
+import argparse
+from contextlib import closing
+import xmljson
+
+try:
+    from lxml.etree import parse
+except ImportError:
+    from xml.etree.cElementTree import parse
+
+dialects = {
+    key.lower(): val for key, val in sorted(vars(xmljson).items())
+    if isinstance(val, type) and issubclass(val, xmljson.XMLData)
+}
+
+
+def parse_args(args=None, in_file=sys.stdin, out_file=sys.stdout):
+    parser = argparse.ArgumentParser(prog='xmljson')
+    parser.add_argument('in_file', type=argparse.FileType(), nargs='?', default=in_file,
+                        help='defaults to stdin')
+    parser.add_argument('-o', '--out_file', type=argparse.FileType('w'), default=out_file,
+                        help='defaults to stdout')
+    parser.add_argument('-d', '--dialect', choices=list(dialects.keys()), default='parker',
+                        type=str.lower, help='defaults to parker')
+    args = parser.parse_args() if args is None else parser.parse_args(args)
+
+    if args.dialect not in dialects:
+        raise TypeError('Unknown dialect: %s' % args.dialect)
+    else:
+        dialect = dialects[args.dialect]()
+
+    return args.in_file, args.out_file, dialect
+
+
+def main(*test_args):
+    in_file, out_file, dialect = parse_args() if not test_args else test_args
+    with closing(in_file) as in_file, closing(out_file) as out_file:
+        json.dump(dialect.data(parse(in_file).getroot()), out_file, indent=2)
+
+
+if __name__ == '__main__':
+    main()