add script to convert notebooks to markdown

pandoc/convert_ipynb_to_blog.py

@@ -0,0 +1,55 @@
+import sys
+from pathlib import Path
+import subprocess
+from datetime import date
+
+sitepath = Path("~/git/tomhodson.github.com/").expanduser()
+defaults_file = sitepath / "pandoc/pandoc_notebook_to_markdown.yaml"
+postspath = sitepath / "_posts"
+
+filepath = Path(sys.argv[1]).resolve()
+title = sys.argv[2]
+# title = input(f"Give this blogpost a title (default to filename):")
+if not title: title = filepath.stem.replace("_", " ")
+print(f'Title: "{title}"')
+
+filename = title.lower().replace(" ", "_")
+print(f'Filename: "{title}"')
+
+# subprocess.run(["jupyter",  "nbconvert", "--to=markdown", filepath, "--output", f"{filename}.md"])
+
+print(f'Running pandoc to generate {title}.md')
+subprocess.run(["pandoc",  "-d", defaults_file, f"--extract-media=./{filename}", filepath, "--output", f"{filename}.md"])
+
+with open(f"{filename}.md", "r") as f:
+    content = f.read()
+
+print(f'Replacing image links')
+content = content.replace(f"./{filename}/", f"/assets/blog/{filename}/")
+
+# remove lines that begin with with ":::"
+print(f'Removing lines that begin with :::')
+lines = content.split("\n")
+lines = [line for line in lines if not line.startswith(":::")]
+content = "\n".join(lines)
+
+metadata = f"""---
+title:  {title}
+layout: post
+image: 
+---
+"""
+
+print(f'Adding metadata')
+print(f'Writing out to {filename}.md')
+with open(f"{filename}.md", "w") as f:
+    f.write(metadata + content)
+
+print(f'Copying over')
+todays_date = date.today().isoformat()
+subprocess.run(["mv", f"{filename}.md", postspath / f"{todays_date}-{filename}.md"])
+subprocess.run(["mv", f"./{filename}", sitepath / f"assets/blog/{filename}"])
+
+
+
+

pandoc/pandoc_notebook_to_markdown.yaml

@@ -0,0 +1,3 @@
+from: ipynb
+to: markdown
+wrap: none