build: fix clean rule

Describe how to get RSS feeds for YouTube channels

blog

@@ -1 +1 @@
-Subproject commit 30dd1e7e01c6348324a3fef001b3a2fabfc05788
+Subproject commit 5a1de185c30b8a1ef3400fe161fcc8f12fbacaa0

makefile

@@ -15,6 +15,9 @@ BLOG_BUILD_DIR	= $(BLOG_BASE_DIR)out/html/
 BLOG_OUT_DIR	= $(OUT_DIR)/blog
 BLOG_INDEX		= $(BLOG_BUILD_DIR)/index.html
 
+GARDEN_BASE_DIR 	= $(SRC_DIR)/garden/
+GARDEN_BUILD_DIR 	= $(GARDEN_BASE_DIR)html
+
 PAGES 			= $(shell find $(ROOT_DIR) -wholename "$(ROOT_DIR)*.html")
 STYLES			= $(shell find $(ROOT_DIR) -wholename "$(ROOT_DIR)*.css")
 IMAGES			= $(shell find $(IMG_DIR) -wholename "$(IMG_DIR)/*.png")
@@ -26,7 +29,8 @@ HTML_TARGETS 	= $(PAGES:$(ROOT_DIR)/%.html=$(OUT_DIR)/%.html)
 CSS_TARGETS 	= $(STYLES:$(ROOT_DIR)/%.css=$(OUT_DIR)/%.css)
 PNG_TARGETS		= $(IMG_DIR)/%.png=$(OUT_DIR)/%.png
 
-all: html blog | $(OUT_DIR)
+all: html blog garden | $(OUT_DIR)
+	cp -r src/garden/html $(OUT_DIR)/garden
 
 html: $(HTML_TARGETS) $(CSS_TARGETS) | $(OUT_DIR)
 	cp $(IMG_DIR)/*.png $(OUT_DIR)/
@@ -62,8 +66,12 @@ blog: $(HTML_INCLUDES) $(CSS_TARGETS)
 		cp $(CSS_TARGETS) `dirname $$page` ; \
 	done
 
+garden:
+	make --directory $(GARDEN_BASE_DIR) site
+
 clean:
+	make --directory $(GARDEN_BASE_DIR) clean
 	make --directory $(BLOG_BASE_DIR) clean
 	rm -rf $(OUT_DIR)
 
-.PHONY: blog
+.PHONY: blog garden

src/garden/.gitignore

@@ -0,0 +1 @@
+*.html

src/garden/Makefile

@@ -0,0 +1,22 @@
+py = feed.py books.py
+md = rss.md book-collecting.md gardens.md
+html = $(md:%.md=%.html)
+
+site: Makefile $(md) $(py)
+	mkdir html
+	python md2html.py $(md)
+	cp $(html) $(py) Makefile html
+
+clean-html:
+	rm -r html
+
+.PHONY: clean-html
+
+feed:
+	python feed.py `pwd`
+
+rss: feed
+
+clean: clean-html
+
+.PHONY: feed clean

src/garden/book-collecting.md

@@ -0,0 +1,106 @@
+how do you define a book collection?
+my book collection is the set of all books.
+
+i prefer physical books to e-readers.
+unfortunately i have quite a few these days.
+
+i want to read them all eventually!
+i also tend to live in quite small places
+and i want to be able to move city easily!
+
+so here's my system for organising my physical book collection.
+
+i want to:
+*   read books i already have
+*   read as many different books as possible
+*   minimise physical storage requirements
+*   keep track of books i've read
+*   gather books i don't already have
+
+constraints
+*   i don't know for sure what book i will want to read next
+
+for every book in the world
+*   i either have or have not read it
+*   i have access to it or i don't
+
+so i sort my book collection with 4 categories
+
+*-------------------*-----------------------*
+|                   |                       |
+|       unread      |          read         |
+|        have       |          have         |
+|                   |                       |
+|   37º2 le matin   |   L'Homme des Jeux    |
+|                   |                       |
+*-------------------*-----------------------*
+|                   |                       |
+|       unread      |         read          |
+|       haven't     |        haven't        |
+|                   |                       |
+|    Das Kapital    |      Frankisstein     |
+|                   |                       |
+*-------------------*-----------------------*
+
+i can then begin to optimise my collection.
+
+*   i do not have this book, and i have read it.
+*   i have this book, but i have not read it.
+*   i have this book, and i have read it.
+*   i do not have this book, but i have not read it.
+
+the books i am most interested in having nearby are unread ones, as i would like to read as many different books as possible.
+
+books i have already read i don't need nearby anymore.
+i might pass it on, or store it somewhere with less of a premium on space.
+i could also attempt to track where it is!
+
+based on my requirements and my categories, i create four lists for the books
+
+*   ready
+*   all done
+*   read and gone
+*   hunted
+
+that looks like a decent start to the system, so i suppose now i'll start collecting!
+
+so i'll use markdown lists in the format
+
+```
+* [x] author - title    # reading
+* [ ] author - title    # nearby
+```
+when collecting music i use artist - year - name
+however, publication year is an extra step that will slow data entry, so won't use this to start with - i have a lot of books
+
+i realised i had a gpt-4 sub and that it could look at pictures now so i gave it a go
+i fed it some photos and some formatting preferences and i got out perfect markdown lists
+
+```
+- [ ] Doctorow, Cory - Walkaway
+- [ ] Ferreira, Pedro G. - The Perfect Theory
+- [ ] Hadfield, Chris - An Astronaut's Guide to Life on Earth
+- [ ] Heinlein, Robert A. - Beyond This Horizon
+```
+
+books are lovely are great to look at, but the mishmash of fonts and presentation are a nightmare for indexing.
+now we have some good and lovely metadata :)
+this is an imperfect method, as the only way i can check it is still by combing through the physical books manually
+but it does let me target my combing after identifying problems in the index
+and in the meantime gives us a bunch of data to play with
+
+markdown lists also allow me to mark some items in a list
+this is looks flexible, so i think in my 'ready' list i will mark which book(s) i am currently reading
+
+```
+- [ ] Doctorow, Cory - Walkaway
+- [ ] Ferreira, Pedro G. - The Perfect Theory
+- [ ] Hadfield, Chris - An Astronaut's Guide to Life on Earth
+- [ ] Heinlein, Robert A. - Beyond This Horizon
+```
+
+the other lists i will leave unmarked for now, until i think of something to do with them.
+
+as for doing things with them, i wrote a [python script](#) which processes the data in the now-populated all-done and ready lists to yield some interesting (?) and fun (?) results?
+
+[example output](#)

src/garden/books.py

@@ -0,0 +1,87 @@
+import sys
+import os
+import re
+
+def print_usage():
+    print(f"usage: python {sys.argv[0]} DIR")
+    print(f"")
+    print(f"\tDIR\tdirectory containing markdown lists in files.")
+
+
+if len(sys.argv) != 2:
+    print_usage()
+    exit(1)
+
+base_path = os.path.abspath(sys.argv[1])
+ready_list_name = "ready.md"
+done_list_name = "all-done.md"
+
+def get_path(list_name : str) -> str:
+    return os.path.join(base_path, list_name)
+
+
+def get_matches(list_name : str) -> list[re.Match]:
+    # Matches a markdown list item
+    entry_pattern = re.compile(r"^[*-] \[([ *x])\] (.+) - (.+)")
+
+    matches = []
+    with open(get_path(list_name)) as f:
+        matches = [entry_pattern.match(l) for l in f.readlines()]
+    return [m for m in matches if m is not None]
+
+
+class Book:
+    def __init__(self, match : re.Pattern):
+        self.mark = match.group(1) != " "
+        self.author = match.group(2)
+        self.title = match.group(3)
+
+    def is_metadata_complete(self):
+        if not self.title or not self.author:
+            return False
+
+        if self.title == "???" or self.author == "???":
+            return False
+
+        return True
+
+    @staticmethod
+    def get_list(list_name : str, filter_partial_metadata = True) -> []:
+        books = [Book(m) for m in get_matches(list_name)] 
+
+        if filter_partial_metadata:
+            books = [b for b in books if b.is_metadata_complete()]
+
+        return books
+
+
+def print_section(title : str, books : list[str]):
+    print(f"# {title} ({len(books)})\n")
+
+    longest_title = max([len(b.title) for b in books])
+    title_column_width = longest_title + 2
+
+    for book in books:
+        row = [book.title, book.author]
+        format_str = "- {: <" + str(title_column_width) + "} {: <20}"
+        print(format_str.format(*row))
+    print()
+
+def print_in_progress():
+    books = [b for b in Book.get_list(ready_list_name, False) if b.mark]
+    print_section("in progress", books)
+
+def print_completed():
+    books = Book.get_list(done_list_name)
+    print_section("up for borrowing", books)
+
+def print_partial_metadata():
+    books = Book.get_list(ready_list_name, False)
+    books += Book.get_list(done_list_name, False)
+    books = [b for b in books if not b.is_metadata_complete()]
+
+    print_section("metadata incomplete", books)
+
+print_in_progress()
+print_completed()
+print_partial_metadata()

src/garden/feed.py

@@ -0,0 +1,63 @@
+#!/usr/bin/env python3
+
+import markdown
+import pathlib
+import sys
+import re
+import glob
+import os
+
+def print_usage():
+    print("\nusage: python feed.py ROOT\n")
+    print("\n")
+    print("\t\ROOT\tbase folder")
+
+# check args for at most one file paths
+if len(sys.argv) > 2:
+    print_usage()
+    sys.exit(1)
+
+base_folder = sys.argv[1] if len(sys.argv) == 2 else os.getcwd()
+print(base_folder)
+
+def get_paths() -> [str]:
+    return [x for x in glob.glob(f"{base_folder}/*.md")]
+
+def get_text(path):
+    with open(path) as f:
+        return f.read()
+
+#def to_html(md : str) -> str:
+#    return markdown.markdown(md, extensions=["fenced_code"])
+
+def get_title(md):
+    m = re.compile(r"^# (.+)\n").match(md)
+    if m is not None:
+        return m.groups(1)[0]
+
+    # truncated first line of file for auto-title
+    return md.splitlines()[0][0:30]
+
+def get_entry(path):
+    return get_title(get_text(path))
+
+def get_entries() -> [str]:
+    entries = [get_entry(p) for p in get_paths()]
+    return "\n\n".join(entries)
+
+def get_header() -> str:
+    return """<?xml version="1.0" encoding="utf-8" ?>
+<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0">
+<channel>
+    <title></title>
+    <link>https://ktyl.dev/garden</link>
+    <description></description>
+    <atom:link href="https://ktyl.dev/garden/feed.xml" rel="self" type="application/rss+xml"/>
+"""
+
+def get_footer() -> str:
+    return "\n</channel></rss>"
+
+print(get_header())
+print(get_entries())
+print(get_footer())

src/garden/feed.xml

@@ -0,0 +1,15 @@
+
+python feed.py `pwd`
+<?xml version="1.0" encoding="utf-8" ?>
+<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0">
+<channel>
+    <title></title>
+    <link>https://ktyl.dev/garden</link>
+    <description></description>
+    <atom:link href="https://ktyl.dev/garden/feed.xml" rel="self" type="application/rss+xml"/>
+
+Digital Gardens
+
+RSS
+
+</channel></rss>

src/garden/gardens.md

@@ -0,0 +1,5 @@
+# Digital Gardens
+
+* Obsidian
+* Notion
+* Git

src/garden/md2html.py

@@ -0,0 +1,32 @@
+#!/usr/bin/env python
+
+import sys
+import markdown
+
+def print_usage():
+    print(f"usage: python {sys.argv[0]} PATHS")
+    print("")
+    print("\t\PATHS\tpaths of input markdown files")
+
+if len(sys.argv) < 2:
+    print_usage()
+    sys.exit(1)
+
+paths = sys.argv[1:]
+
+bad_paths = [p for p in paths if not p.endswith(".md")]
+if len(bad_paths) != 0:
+    for p in bad_paths:
+        print(f"Not a markdown file: {p}")
+
+    exit(1)
+
+def write_html(src : str):
+    with open(src) as md:
+        dest = src.replace(".md", ".html")
+        with open(dest, "w") as html:
+            print(f"{src} -> {dest}")
+            html.write(markdown.markdown(md.read()))
+
+for p in paths:
+    write_html(p)

src/garden/rss.md

@@ -0,0 +1,35 @@
+# RSS
+
+Really Simple Syndication (RSS) is an ancient file format for communicating updates over the Web, and my personal favourite.
+
+* [It's Time for an RSS Revival](https://www.wired.com/story/rss-readers-feedly-inoreader-old-reader/)
+
+### Stubs:
+
+* my curated feeds
+* reader applications
+
+## Technical Simplicity
+
+From a technical perspective, like [Gemini](gemini://gemini.circumlunar.space) [\(HTTP\)](https://gemini.circumlunar.space/) it's grokkable with a couple good reads of the specification.
+I wrote [a simple generator](https://ktyl.dev/blog/2022/6/3/rss.md)] for my blog posts, and am now working on [a simpler one](./feed.py) for the garden.
+To start with, I'd like to generate it as simply as possible, from just a directory structure, but I can already see I'd like to do more with it.
+
+I work predominantly in [Git](https://git-scm.com/) repositories, which, like a directory, or a garden, is tree-based.
+My blog's feed has a chronological hierarchy, but at present it's based on manually creating and naming folders.
+It's also very much a _published_ format, rather than a living one.
+This is silly!
+Instead, I'd like to structure my feed around a Git repository, as I think it'd be a much better descriptor of activity.
+
+## Managing YouTube Subscriptions
+
+There are many reasons to avoid the YouTube homepage, such as recommended videos or the accursed Shorts.
+YouTube publishes channel-specific RSS feeds, making it totally possible to circumvent its subscription system entirely - including having to make an account - and keep track of channels one enjoys with RSS feeds instead.
+Get a channel's ID:
+
+1. Go to the channel's page
+2. Go to the About section
+3. Share icon under Stats block
+4. Copy channel ID
+
+The channel's feed is available at `https://www.youtube.com/feeds/videos.xml?channel_id=CHANNEL_ID_HERE`.

src/inc_html/nav.html

@@ -2,4 +2,5 @@
     <li><a href="/about.html">about</a></li>
     <li><a href="/blog.html">blog</a></li>
     <li><a href="/gallery.html">gallery</a></li>
+    <li><a href="/garden/index.html">garden</a></li>
 </ul>