Add RSS generation

Remove old deprecated CLI converter in __init__

3 files changed, 68 insertions, 28 deletions

diff --git a/gemini-to-web/src/gemini_to_web/__init__.py b/gemini-to-web/src/gemini_to_web/__init__.py
index 0b2b70e..827031f 100644
--- a/gemini-to-web/src/gemini_to_web/__init__.py
+++ b/gemini-to-web/src/gemini_to_web/__init__.py
@@ -1,25 +1,4 @@
-import argparse
-import pathlib
-import shutil
+import re
 
-import htmlgenerator
 
-from gemini_to_web import html
-
-def converter():
-    parser = argparse.ArgumentParser()
-    parser.add_argument("source", type=pathlib.Path)
-    parser.add_argument("target", type=pathlib.Path)
-    args = parser.parse_args()
-
-    shutil.copytree(args.source, args.target)
-    for gmi in args.target.glob("**/*.gmi"):
-        html_path = gmi.with_suffix(".html")
-        html_path.write_text(
-            html.pretty(
-                htmlgenerator.render(
-                    html.to_html(gmi.read_text()),
-                    {}
-                )
-            )
-        )
+ENTRY_ELEMENT_TITLE = re.compile(r"^(\d{4}-\d{2}-\d{2}) (.*)$")
diff --git a/gemini-to-web/src/gemini_to_web/indexer.py b/gemini-to-web/src/gemini_to_web/indexer.py
index 3d463e3..a6ceee0 100644
--- a/gemini-to-web/src/gemini_to_web/indexer.py
+++ b/gemini-to-web/src/gemini_to_web/indexer.py
@@ -1,15 +1,12 @@
 import argparse
 import pathlib
-import re
 import sys
 
+import gemini_to_web
 from gemini_to_web import parser
 from gemini_to_web import html
 
 
-ENTRY_ELEMENT_TITLE = re.compile(r"^(\d{4}-\d{2}-\d{2}) (.*)$")
-
-
 def cli_indexer():
     argument_parser = argparse.ArgumentParser()
     argument_parser.add_argument("base_path", type=pathlib.Path)
@@ -25,9 +22,10 @@ def cli_indexer():
         parsed = parser.parse(path.read_text())
         parsed = list(parsed)
         first_header_title = html.first_header_title_extractor(parsed)
-        match = ENTRY_ELEMENT_TITLE.match(first_header_title)
+        match = gemini_to_web.ENTRY_ELEMENT_TITLE.match(first_header_title)
         if match:
             posts.append((match.group(1), path.relative_to(args.base_path), match.group(2)))
     posts = sorted(posts, reverse=True)
     for date, path, title in posts:
+        path = str(path).removesuffix(".gmi")
         print(f"=> {path} {date} {title}")
diff --git a/gemini-to-web/src/gemini_to_web/rss.py b/gemini-to-web/src/gemini_to_web/rss.py
new file mode 100644
index 0000000..a03a4aa
--- /dev/null
+++ b/gemini-to-web/src/gemini_to_web/rss.py
@@ -0,0 +1,63 @@
+import argparse
+import datetime
+import pathlib
+import sys
+
+from feedgen import feed
+import htmlgenerator
+
+import gemini_to_web
+from gemini_to_web import html
+from gemini_to_web import parser
+
+
+def cli_to_rss():
+    argument_parser = argparse.ArgumentParser()
+    argument_parser.add_argument("--title", default="")
+    argument_parser.add_argument("--subtitle", default=None)
+    argument_parser.add_argument("--base-url", default="")
+    argument_parser.add_argument("base_dir", type=pathlib.Path)
+    args = argument_parser.parse_args()
+
+    fg = feed.FeedGenerator()
+    fg.title(args.title)
+    fg.subtitle(args.subtitle)
+    fg.link(href=args.base_url, rel="self")
+
+    input_ = sys.stdin.read()
+    parsed = parser.parse(input_)
+    parsed = list(parsed)
+
+    entries = []
+    for element in parsed:
+        match element:
+            case parser.LinkLine(url, link_name):
+                match = gemini_to_web.ENTRY_ELEMENT_TITLE.match(link_name)
+                if match:
+                    entries.append((match.group(1), url, match.group(2)))
+            case _:
+                pass
+
+    entries = sorted(entries, reverse=True)
+    entries = entries[0:10]
+
+    for (date, url, title) in entries:
+        feed_entry = fg.add_entry()
+        feed_entry.link(href=args.base_url + "/" + url)
+        feed_entry.published(
+            datetime.datetime.combine(
+                datetime.date.fromisoformat(date),
+                datetime.datetime.min.time(),
+                tzinfo=datetime.UTC,
+            )
+        )
+        feed_entry.title(title)
+        parsed = parser.parse((args.base_dir / pathlib.Path(url).with_suffix(".gmi")).read_text())
+        parsed = list(parsed)
+        parsed = parsed[1:]
+        content = html.to_html(parsed)
+        rendered = htmlgenerator.render(content[1], {})
+        rendered = html.pretty(rendered)
+        feed_entry.content(rendered.encode("utf8"), type="html")
+
+    print(fg.rss_str(pretty=True).decode("utf8"), end="")