From fa7721c31d47e6876728c19773c54cf5ab28da82 Mon Sep 17 00:00:00 2001
From: Kelly Brazil <kellyjonbrazil@gmail.com>
Date: Tue, 19 Jul 2022 07:16:28 -0700
Subject: [PATCH] add initial URL parser

---
 CHANGELOG         |  3 +-
 jc/lib.py         |  1 +
 jc/parsers/url.py | 97 +++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 100 insertions(+), 1 deletion(-)
 create mode 100644 jc/parsers/url.py

diff --git a/CHANGELOG b/CHANGELOG
index 08a39992..a0369407 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -1,8 +1,9 @@
 jc changelog
 
 20220705 v1.20.3
-- Add pager functionality to help (parser documentation only)
+- Add URL string parser
 - Add m3u/m3u8 file parser
+- Add pager functionality to help (parser documentation only)
 - Minor parser performance optimizations
 
 20220705 v1.20.2
diff --git a/jc/lib.py b/jc/lib.py
index f554a824..dd8767cc 100644
--- a/jc/lib.py
+++ b/jc/lib.py
@@ -105,6 +105,7 @@ parsers = [
     'update-alt-q',
     'upower',
     'uptime',
+    'url',
     'vmstat',
     'vmstat-s',
     'w',
diff --git a/jc/parsers/url.py b/jc/parsers/url.py
new file mode 100644
index 00000000..3e3c1181
--- /dev/null
+++ b/jc/parsers/url.py
@@ -0,0 +1,97 @@
+"""jc - JSON Convert URL parser
+
+Usage (cli):
+
+    $ echo "http://example.com/test/path?q1=foo&q2=bar#frag" | jc --url
+
+Usage (module):
+
+    import jc
+    result = jc.parse('url', url_string)
+
+Schema:
+
+    [
+      {
+        "url":     string,
+        "bar":     boolean,
+        "baz":     integer
+      }
+    ]
+
+Examples:
+
+    $ echo "http://example.com/test/path?q1=foo&q2=bar#frag" | jc --url -p
+    []
+
+    $ FTP example, etc.
+    []
+"""
+from urllib.parse import urlparse
+from typing import List, Dict
+import jc.utils
+
+
+class info():
+    """Provides parser metadata (version, author, etc.)"""
+    version = '1.0'
+    description = 'URL parser'
+    author = 'Kelly Brazil'
+    author_email = 'kellyjonbrazil@gmail.com'
+    compatible = ['linux', 'darwin', 'cygwin', 'win32', 'aix', 'freebsd']
+
+
+__version__ = info.version
+
+
+def _process(proc_data: Dict) -> Dict:
+    """
+    Final processing to conform to the schema.
+
+    Parameters:
+
+        proc_data:   (Dictionary) raw structured data to process
+
+    Returns:
+
+        Dictionary. Structured to conform to the schema.
+    """
+    return proc_data
+
+
+def parse(
+    data: str,
+    raw: bool = False,
+    quiet: bool = False
+) -> Dict:
+    """
+    Main text parsing function
+
+    Parameters:
+
+        data:        (string)  text data to parse
+        raw:         (boolean) unprocessed output if True
+        quiet:       (boolean) suppress warning messages if True
+
+    Returns:
+
+        Dictionary. Raw or processed structured data.
+    """
+    jc.utils.compatibility(__name__, info.compatible, quiet)
+    jc.utils.input_type_check(data)
+
+    raw_output: Dict = {}
+
+    if jc.utils.has_data(data):
+        parts = urlparse(data)
+
+        raw_output = {
+            'scheme': parts.scheme,
+            'netloc': parts.netloc,
+            'path': parts.path,
+            'params': parts.params,
+            'query': parts.query,
+            'fragment': parts.fragment
+        }
+
+    return raw_output if raw else _process(raw_output)