* add sanity check for unknown arguments

* add unlicense to .ipynb

Author: Bondoki

ParsingMetadataMD2JSON.ipynb

@@ -1,16 +1,47 @@
 {
  "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "8bbf8f85-3a07-4b4b-82fe-75f51e54caa1",
+   "metadata": {},
+   "source": [
+    "This is free and unencumbered software released into the public domain.\n",
+    "\n",
+    "Anyone is free to copy, modify, publish, use, compile, sell, or\n",
+    "distribute this software, either in source code form or as a compiled\n",
+    "binary, for any purpose, commercial or non-commercial, and by any\n",
+    "means.\n",
+    "\n",
+    "In jurisdictions that recognize copyright laws, the author or authors\n",
+    "of this software dedicate any and all copyright interest in the\n",
+    "software to the public domain. We make this dedication for the benefit\n",
+    "of the public at large and to the detriment of our heirs and\n",
+    "successors. We intend this dedication to be an overt act of\n",
+    "relinquishment in perpetuity of all present and future rights to this\n",
+    "software under copyright law.\n",
+    "\n",
+    "THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND,\n",
+    "EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF\n",
+    "MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.\n",
+    "IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR\n",
+    "OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,\n",
+    "ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR\n",
+    "OTHER DEALINGS IN THE SOFTWARE.\n",
+    "\n",
+    "For more information, please refer to <https://unlicense.org>"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "id": "8737b6b4-a572-46b2-bd22-b70cfccb501d",
    "metadata": {},
    "outputs": [],
    "source": [
-    "from pathlib import Path\n",
-    "import re\n",
-    "import json\n",
-    "import sys"
+    "from pathlib import Path  # Importing Path class from pathlib for file path manipulations\n",
+    "import re    # Importing re for regular expression operations\n",
+    "import json  # Importing json for handling JSON data\n",
+    "import sys   # Importing sys for command line argument handling"
    ]
   },
   {
@@ -19,43 +50,47 @@
    "id": "8373a7f1-49fe-4279-8a35-fadb413acde0",
    "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "def show_help():\n",
+    "    \"\"\"Display the help message for the script.\"\"\"\n",
+    "    print(\"Usage:\")\n",
+    "    print(\"./ParsingMetadataMD2JSON [options] <input file>...\")\n",
+    "    print(\"Options:\")\n",
+    "    print(\"--help            Show this screen.\")\n",
+    "    print()"
+   ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "id": "ee24ff0d-07b7-4b21-8f38-211783deb980",
    "metadata": {},
    "outputs": [],
    "source": [
-    "\"\"\"\n",
-    "Usage:\n",
-    "    ./ParsingMetadataMD2JSON [options] <input file>...\n",
-    "Options:\n",
-    "    --help            Show this screen.\n",
-    "\"\"\"\n",
-    "\n",
     "def main() -> int:\n",
-    "    # Reads in a MarkDown M_datafile.md parsing every section\n",
-    "    # into a separate JSON M_datafile.json file\n",
+    "    \"\"\"\n",
+    "    Reads in a MarkDown M_datafile.md parsing every section\n",
+    "    into a separate JSON M_datafile.json file\n",
+    "    \"\"\"\n",
     "\n",
     "    # Get command line arguments for file\n",
     "    args = sys.argv[1:]\n",
     "    \n",
+    "    # Check for help option\n",
+    "    if '--help' in args:\n",
+    "        show_help()  # Call the help function\n",
+    "        return 0  # Exit after showing help\n",
+    "    \n",
     "    # Get the file to convert\n",
     "    try:\n",
     "        md_path = Path(args[0])\n",
     "        if not md_path.exists():\n",
     "            raise FileNotFoundError(f\"{md_path} does not exist.\")\n",
     "    except IndexError:\n",
-    "        print(\"Usage:\")\n",
-    "        print(\"./ParsingMetadataMD2JSON [options] <input file>...\")\n",
-    "        print(\"Options:\")\n",
-    "        print(\"--help            Show this screen.\")\n",
-    "        print()\n",
-    "        #raise IndexError(f\"Expected file name as first argument\")\n",
+    "        # Handle the case where no file is provided\n",
     "        print(f\"Expected file name as first argument...EXITING...\")\n",
-    "        sys.exit(1) # something went wrong\n",
+    "        show_help()  # Show help message\n",
+    "        sys.exit(1)  # Exit the program with an error code as something went wrong\n",
     "    \n",
     "    with open(str(md_path), \"r\", encoding=\"utf-8\") as f:\n",
     "        contents = f.read()#.replace(\"\\r\", \"\") # Remove Windows style line breaks\n",
@@ -68,14 +103,15 @@
     "    with open(Path(args[0][:-len(\".md\")] + \".json\"), 'w', encoding='utf-8') as f:\n",
     "        json.dump(dictionaryJSON, f, ensure_ascii=False, indent=4)\n",
     "    \n",
+    "    # Print success message indicating the output file\n",
     "    print(\"SUCCESS: \" + str(Path(args[0]))+\" parsed to \" + str(Path(args[0][:-len(\".md\")] + \".json\")) )\n",
     "\n",
     "    return 0 # success\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "id": "650d3f63-758a-49b6-9663-338dcc47f653",
    "metadata": {},
    "outputs": [],
@@ -96,7 +132,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "id": "f9f69886-9c77-4017-90bb-6f9e9ee82d42",
    "metadata": {},
    "outputs": [],
@@ -106,22 +142,22 @@
     "    # change, amend, or replace to your needs\n",
     "\n",
     "    # Initialize variables to store the JSON data\n",
-    "    M_title = {}\n",
-    "    M_creators = [] # can be a list\n",
-    "    M_publisher = {}\n",
-    "    M_contributors = [] # can be a list\n",
-    "    M_description = {}\n",
-    "    M_subjects = [] # a list of keywords\n",
-    "    M_date = {}\n",
-    "    M_language = [] # a list of languages\n",
-    "    M_formats = [] # a list of file formats\n",
-    "    M_type = {}\n",
-    "    M_coverage = {}\n",
-    "    M_identifier = {}\n",
-    "    M_methods = []\n",
-    "    M_sources = []\n",
-    "    M_relations = []\n",
-    "    M_rights = []\n",
+    "    M_title = {}        # the title of entry\n",
+    "    M_creators = []     # can be a list of creators\n",
+    "    M_publisher = {}    # the publisher\n",
+    "    M_contributors = [] # can be a list of contributors\n",
+    "    M_description = {}  # the description of entry\n",
+    "    M_subjects = []     # a list of keywords\n",
+    "    M_date = {}         # the last-changed date\n",
+    "    M_language = []     # a list of languages\n",
+    "    M_formats = []      # a list of file formats\n",
+    "    M_type = {}         # the type of entry\n",
+    "    M_coverage = {}     # the period of time\n",
+    "    M_identifier = {}   # unique identifier (PID)\n",
+    "    M_methods = []      # a list of methods for creating the entry\n",
+    "    M_sources = []      # a list of sources of the entry\n",
+    "    M_relations = []    # a list of relations about the entry\n",
+    "    M_rights = []       # a refernce of licence statements\n",
     "    \n",
     "    \n",
     "    for index, entry in enumerate(contentdictionary):\n",
@@ -270,18 +306,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": null,
    "id": "32674c9d-4060-4fcb-bfe0-97b922d64227",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "SUCCESS: M_Dataset_README_Example.md parsed to M_Dataset_README_Example.json\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "%%python3 ParsingMetadataMD2JSON.py ./M_Dataset_README_Example.md\n",
     "# cell magic to fake command line execution\n",

ParsingMetadataMD2JSON.py

@@ -28,11 +28,6 @@
 For more information, please refer to <https://unlicense.org>
 """
 
-from pathlib import Path
-import re
-import json
-import sys
-
 from pathlib import Path  # Importing Path class from pathlib for file path manipulations
 import re    # Importing re for regular expression operations
 import json  # Importing json for handling JSON data
@@ -42,7 +37,7 @@
 def show_help():
     """Display the help message for the script."""
     print("Usage:")
-    print("./ParsingMetadataMD2JSON [options] <input file>...")
+    print("./ParsingMetadataMD2JSON [options] <input file.md>...")
     print("Options:")
     print("--help            Show this screen.")
     print()
@@ -61,6 +56,13 @@ def main() -> int:
         show_help()  # Call the help function
         return 0  # Exit after showing help
 
+    # Check for unknown arguments
+    for arg in args:
+        if not arg.startswith('--help') and not arg.endswith('.md'):
+            print(f"Error: Unknown argument '{arg}'.")
+            show_help()
+            sys.exit(1)
+    
     # Get the file to convert
     try:
         md_path = Path(args[0])
@@ -72,6 +74,7 @@ def main() -> int:
         show_help()  # Show help message
         sys.exit(1)  # Exit the program with an error code as something went wrong
 
+    # read the file to convert and parse
     with open(str(md_path), "r", encoding="utf-8") as f:
         contents = f.read()#.replace("\r", "") # Remove Windows style line breaks