From c9291830d07f2cf3b02a08cd76b798796ea2997a Mon Sep 17 00:00:00 2001
From: Rohan Kumar <seirdy@seirdy.one>
Date: Tue, 7 Jun 2022 09:39:32 -0700
Subject: [PATCH] Automatically pull webring links during build

---
 .build.yml                              |  1 +
 .gitignore                              |  1 +
 Makefile                                |  5 +-
 csv/{webrings.csv => webrings.base.csv} |  1 -
 scripts/populate-webrings.sh            | 84 +++++++++++++++++++++++++
 scripts/xhtmlize-single-file.sh         |  1 +
 6 files changed, 91 insertions(+), 2 deletions(-)
 rename csv/{webrings.csv => webrings.base.csv} (93%)
 create mode 100755 scripts/populate-webrings.sh

diff --git a/.build.yml b/.build.yml
index 7426726..b3b59d2 100644
--- a/.build.yml
+++ b/.build.yml
@@ -4,6 +4,7 @@
 # hugo, brotli, ect (like gzip/zopfli), sd, and xmlfmt.
 image: alpine/edge
 packages:
+  - curl # for webring update script
   - rsync
   - git  # for Hugo's gitInfo
   - bmake
diff --git a/.gitignore b/.gitignore
index 8b23cde..7aaaa17 100644
--- a/.gitignore
+++ b/.gitignore
@@ -10,3 +10,4 @@ old/
 public/
 public_*/
 .hugo_build.lock
+csv/webrings.csv
diff --git a/Makefile b/Makefile
index 29d1f07..8a8de1b 100644
--- a/Makefile
+++ b/Makefile
@@ -19,8 +19,11 @@ ECT_LEVEL=9
 
 VNU ?= vnu
 
+csv/webrings.csv:
+	sh scripts/populate-webrings.sh
+
 .PHONY: hugo
-hugo: $(SRCFILES)
+hugo: csv/webrings.csv $(SRCFILES)
 	hugo -b $(HUGO_BASEURL) $(HUGO_FLAGS) -d $(OUTPUT_DIR)
 
 # .hintrc-local for linting local files
diff --git a/csv/webrings.csv b/csv/webrings.base.csv
similarity index 93%
rename from csv/webrings.csv
rename to csv/webrings.base.csv
index e1b0606..512b67d 100644
--- a/csv/webrings.csv
+++ b/csv/webrings.base.csv
@@ -6,7 +6,6 @@ no js webring,https://nojs.sugarfi.dev/prev.php?url=https://seirdy.one,https://n
 Fediring,https://fediring.net/previous?host=seirdy.one,https://fediring.net/,https://fediring.net/next?host=seirdy.one
 Loop (JS),https://graycot.com/webring/loop-redirect.html?action=prev,https://graycot.com/webring/,https://graycot.com/webring/loop-redirect.html?action=next
 YesterWeb,https://webring.yesterweb.org/noJS/index.php?d=prev&url=https://seirdy.one/,https://yesterweb.org/webring/,https://webring.yesterweb.org/noJS/index.php?d=next&url=https://seirdy.one/
-GEORGE,https://aroace.space,https://george.gh0.pw,http://truttle1.xyz/truttlecities/~Itzzy/MainStreet/index.html
 Retronaut,https://webring.dinhe.net/prev/https://seirdy.one,https://webring.dinhe.net/,https://webring.dinhe.net/next/https://seirdy.one
 Hotline,https://hotlinewebring.club/seirdy/previous,https://hotlinewebring.club,https://hotlinewebring.club/seirdy/next
 Bucket (JS),https://webring.bucketfish.me/redirect.html?to=prev&name=seirdy,https://webring.bucketfish.me/,https://webring.bucketfish.me/redirect.html?to=next&name=seirdy
diff --git a/scripts/populate-webrings.sh b/scripts/populate-webrings.sh
new file mode 100755
index 0000000..5d6a7eb
--- /dev/null
+++ b/scripts/populate-webrings.sh
@@ -0,0 +1,84 @@
+#!/bin/sh
+
+# Get webring links to append to the CSV before building
+
+# Some webrings require an iframe or scripts, but I want a plain static
+# first-party HTML+CSS page. This script fetches webring links by
+# parsing the iframes and appends appropriate values to my webrings.csv file.
+
+set -e -u
+
+
+# the name of this program
+progname="$(basename "$0")"
+dirname="$(dirname "$0")"
+
+webrings_src="$dirname/../csv/webrings.base.csv"
+webrings_dest="$dirname/../csv/webrings.csv"
+
+help_text="Usage: $progname [OPTION...]
+
+Update webrings.csv with new prev/next links, to avoid iframes/scripts
+
+Options:
+  -h            Print this help and exit
+  -d            Dry run; just print results, don't update webrings.csv
+"
+
+usage() {
+	printf '%s' "$help_text"
+}
+
+# when the user passess bad args, send a msg to stderr and exit
+# usage: bad_option <option> <reason>
+bad_option() {
+	echo "$progname: option $1: $2" >&2
+	usage >&2
+	exit 1
+}
+
+dry_run='0'
+
+while getopts "hd" flags; do
+	case $flags in
+		h)
+			usage
+			exit 0
+			;;
+		d)
+			dry_run='1'
+			shift
+			;;
+		*)
+			bad_option "$flags" 'invalid option'
+			exit 1
+			;;
+	esac
+done
+
+# values for the GEORGE webring
+george() {
+	echo GEORGE
+	curl -s 'https://george.gh0.pw/embed.cgi?seirdy' | htmlq -a href 'main p a'
+}
+
+values_to_csv() {
+	tr '\n' ',' && echo
+}
+
+print_csv_values() {
+	printf %s "$(george)" | values_to_csv
+}
+
+if [ -f "$webrings_dest" ]; then
+	echo "webrings file already generated"
+	exit 0
+fi
+
+if [ "$dry_run" = '1' ]; then
+	print_csv_values
+else
+	print_csv_values | cat "$webrings_src" - >"$webrings_dest"
+fi
+
+# vi:ft=sh
diff --git a/scripts/xhtmlize-single-file.sh b/scripts/xhtmlize-single-file.sh
index 75495af..456991b 100644
--- a/scripts/xhtmlize-single-file.sh
+++ b/scripts/xhtmlize-single-file.sh
@@ -29,6 +29,7 @@ sed 7d "$html_file" | xmllint --format --encode UTF-8 --noent - | sd '^\t' '' >"
 {
 	head -n7 "$tmp_file"
 	cat tmp.css
+	# shellcheck disable=SC2016 # these are regex statements, not shell expressions
 	tail -n +8 "$tmp_file" \
 		| sd '<pre(?: tabindex="0")?>\n\t*<code ' '<pre tabindex="0"><code ' \
 		| sd '(?:\n)?</code>\n(?:[\t\s]*)?</pre>' '</code></pre>' \