WIPakislyuk-xq

5 files changed, 30 insertions, 7 deletions

diff --git a/README.rst b/README.rst
index 827e624..28995f7 100644
--- a/README.rst
+++ b/README.rst
@@ -1,3 +1,6 @@
+FIXME: use https://github.com/martinblech/xmltodict
+
+
 yq: Command-line YAML processor - jq wrapper for YAML documents
 ===============================================================
 
diff --git a/scripts/xq b/scripts/xq
new file mode 100755
index 0000000..3435723
--- /dev/null
+++ b/scripts/xq
@@ -0,0 +1,5 @@
+#!/usr/bin/env python
+
+from yq import main
+
+main(input_format="xml")
diff --git a/scripts/yq b/scripts/yq
index b6672ba..054acf3 100755
--- a/scripts/yq
+++ b/scripts/yq
@@ -2,4 +2,4 @@
 
 from yq import main
 
-main()
+main(input_format="yaml")
diff --git a/setup.py b/setup.py
index 8f44242..5fe8156 100755
--- a/setup.py
+++ b/setup.py
@@ -12,11 +12,13 @@ setup(
     license="Apache Software License",
     author="Andrey Kislyuk",
     author_email="kislyuk@gmail.com",
-    description="Command-line YAML processor - jq wrapper for YAML documents",
+    description="Command-line YAML and XML processor - jq wrapper for YAML and XML documents",
     long_description=open("README.rst").read(),
     install_requires=[
         "setuptools",
-        "PyYAML >= 3.11"
+        "PyYAML >= 3.11",
+        "xmljson >= 0.1.9",
+        "defusedxml >= 0.5.0"
     ],
     tests_require=tests_require,
     extras_require={"test": tests_require},
diff --git a/yq/__init__.py b/yq/__init__.py
index af765d2..bfea92a 100755
--- a/yq/__init__.py
+++ b/yq/__init__.py
@@ -10,7 +10,9 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 import os, sys, argparse, subprocess, json
 from collections import OrderedDict
 from datetime import datetime, date, time
+
 import yaml
+
 from .version import __version__
 
 class Parser(argparse.ArgumentParser):
@@ -54,6 +56,9 @@ parser = Parser(description=__doc__, formatter_class=argparse.RawTextHelpFormatt
 parser.add_argument("--yaml-output", "--yml-output", "-y", help="Transcode jq JSON output back into YAML and emit it",
                     action="store_true")
 parser.add_argument("--width", "-w", type=int, help="When using --yaml-output, specify string wrap width")
+parser.add_argument("--xml-converter", nargs="?", choices=["abdera", "badgerfish", "cobra", "gdata", "parker", "yahoo"],
+                    default="badgerfish")
+parser.add_argument("--xml-output", "-x", action="store_true")
 parser.add_argument("--version", action="version", version="%(prog)s {version}".format(version=__version__))
 
 # jq arguments that consume positionals must be listed here to avoid our parser mistaking them for our positionals
@@ -65,7 +70,7 @@ for arg in jq_arg_spec:
 parser.add_argument("jq_filter")
 parser.add_argument("files", nargs="*", type=argparse.FileType())
 
-def main(args=None):
+def main(args=None, input_format="yaml"):
     args, jq_args = parser.parse_known_args(args=args)
     for arg in jq_arg_spec:
         values = getattr(args, arg, None)
@@ -102,9 +107,17 @@ def main(args=None):
             yaml.dump_all(decode_docs(jq_out, json_decoder), stream=sys.stdout, Dumper=OrderedDumper, width=args.width,
                           allow_unicode=True, default_flow_style=False)
         else:
-            for input_stream in input_streams:
-                for doc in yaml.load_all(input_stream, Loader=OrderedLoader):
-                    json.dump(doc, jq.stdin, cls=JSONDateTimeEncoder)
+            if input_format == "yaml":
+                for input_stream in input_streams:
+                    for doc in yaml.load_all(input_stream, Loader=OrderedLoader):
+                        json.dump(doc, jq.stdin, cls=JSONDateTimeEncoder)
+                        jq.stdin.write("\n")
+            elif input_format == "xml":
+                from defusedxml.ElementTree import parse
+                import xmljson
+                for input_stream in input_streams:
+                    data = getattr(xmljson, args.xml_converter).data(parse(input_stream).getroot())
+                    json.dump(data, jq.stdin)
                     jq.stdin.write("\n")
             jq.stdin.close()
             jq.wait()