Merge pull request #11 from Senzing/issue-1.dockter.1

Issue 1.dockter.1

rootfs/app/parse_senzing_database_url.py

@@ -1,26 +1,145 @@
 #! /usr/bin/env python
 
-import os
 import json
-from urlparse import urlparse
-
-senzing_database_url = os.environ.get('SENZING_DATABASE_URL')
-parsed = urlparse(senzing_database_url)
-schema = parsed.path.strip('/')
-
-result = {
-    'scheme': parsed.scheme,
-    'netloc': parsed.netloc,
-    'path': parsed.path,
-    'params': parsed.params,
-    'query': parsed.query,
-    'fragment': parsed.fragment,
-    'username': parsed.username,
-    'password': parsed.password,
-    'hostname': parsed.hostname,
-    'port': parsed.port,
-    'schema': schema,
-}
-
-result_json = json.dumps(result)
-print(result_json)
+import os
+import string
+import sys
+from urlparse import urlparse, urlunparse
+
+# Lists from https://www.ietf.org/rfc/rfc1738.txt
+
+safe_character_list = ['$', '-', '_', '.', '+', '!', '*', '(', ')', ',', '"' ] + list(string.ascii_letters)
+unsafe_character_list = [ '"', '<', '>', '#', '%', '{', '}', '|', '\\', '^', '~', '[', ']', '`']
+reserved_character_list = [ ';', ',', '/', '?', ':', '@', '=', '&']
+
+
+def translate(map, astring):
+    new_string = str(astring)
+    for key, value in map.items():
+        new_string = new_string.replace(key, value)
+    return new_string
+
+
+def get_unsafe_characters(astring):
+    result = []
+    for unsafe_character in unsafe_character_list:
+        if unsafe_character in astring:
+            result.append(unsafe_character)
+    return result
+
+
+def get_safe_characters(astring):
+    result = []
+    for safe_character in safe_character_list:
+        if safe_character not in astring:
+            result.append(safe_character)
+    return result
+
+# -----------------------------------------------------------------------------
+# Main
+# -----------------------------------------------------------------------------
+
+
+if __name__ == "__main__":
+
+    result = {
+        'scheme': "",
+        'netloc': "",
+        'path': "",
+        'params': "",
+        'query': "",
+        'fragment': "",
+        'username': "",
+        'password': "",
+        'hostname': "",
+        'port': "",
+        'schema': "",
+    }
+
+    # Get the value of SENZING_DATABASE_URL environment variable.
+
+    original_senzing_database_url = os.environ.get('SENZING_DATABASE_URL')
+    senzing_database_url = original_senzing_database_url
+
+    # Verify "senzing_database_url" value.
+
+    if not senzing_database_url:
+        result.update({
+            "error": "SENZING_DATABASE_URL has no value.",
+        })
+        result_json = json.dumps(result)
+        print(result_json)
+        sys.exit(1)
+
+    # Create lists of safe and unsafe characters.
+
+    unsafe_characters = get_unsafe_characters(senzing_database_url)
+    safe_characters = get_safe_characters(senzing_database_url)
+
+    # Detect an error condition where there are not enough safe characters.
+
+    if len(unsafe_characters) > len(safe_characters):
+        result.update({
+            "error": "There are not enough safe characters to do the translation.",
+            "unsafeCharacters": unsafe_characters,
+            "safeCharacters": safe_characters,
+        })
+        result_json = json.dumps(result)
+        print(result_json)
+        sys.exit(1)
+
+    # Perform translation.
+    # This makes a map of safe character mapping to unsafe characters.
+    # "senzing_database_url" is modified to have only safe characters.
+
+    translation_map = {}
+    safe_characters_index = 0
+    for unsafe_character in unsafe_characters:
+        safe_character = safe_characters[safe_characters_index]
+        safe_characters_index += 1
+        translation_map[safe_character] = unsafe_character
+        senzing_database_url = senzing_database_url.replace(unsafe_character, safe_character)
+
+    # Parse "translated" URL.
+
+    parsed = urlparse(senzing_database_url)
+    schema = parsed.path.strip('/')
+
+    # Construct result.
+
+    result = {
+        'scheme': translate(translation_map, parsed.scheme),
+        'netloc': translate(translation_map, parsed.netloc),
+        'path': translate(translation_map, parsed.path),
+        'params': translate(translation_map, parsed.params),
+        'query': translate(translation_map, parsed.query),
+        'fragment': translate(translation_map, parsed.fragment),
+        'username': translate(translation_map, parsed.username),
+        'password': translate(translation_map, parsed.password),
+        'hostname': translate(translation_map, parsed.hostname),
+        'port': translate(translation_map, parsed.port),
+        'schema': translate(translation_map, schema),
+    }
+
+    # For safety, compare original URL with reconstructed URL.
+
+    url_parts = [
+        result.get('scheme'),
+        result.get('netloc'),
+        result.get('path'),
+        result.get('params'),
+        result.get('query'),
+        result.get('fragment'),
+    ]
+    test_senzing_database_url = urlunparse(url_parts)
+    if test_senzing_database_url != original_senzing_database_url:
+        result.update({
+            "error": "Original and new URLs do not match.",
+            "originalUrl": original_senzing_database_url,
+            "newUrl": test_senzing_database_url,
+        })
+
+    # Print successful results.
+
+    result_json = json.dumps(result)
+    print(result_json)