From 4d7625227b6f231b79b0f7213a3b8deb955463ed Mon Sep 17 00:00:00 2001
From: Rohan Kumar <seirdy@seirdy.one>
Date: Sat, 9 Jan 2021 20:48:42 -0800
Subject: [PATCH] IndieWeb: gather WebWentions with static-webmention

Update the Makefile to download the old version of the site, run
static-webmentions, and collect the WebMentions to send in a json file
saved as a build artifact.

Don't send these automatically; just save them for now. Until I work out
a solution to save sent WebMentions and avoid sending duplicates, I'll
keep the sending of WebMentions manual.

For some reason this caused webhint's axe/aria test to error out with a
"Protocol error" so I disabled it. Axe tests are covered by Lighthouse
anyway.
---
 .build.yml   |  3 +++
 .gitignore   | 10 ++++++----
 .hintrc      |  1 +
 .rsyncignore |  6 +++---
 Makefile     | 17 +++++++++++++----
 README.md    |  3 +++
 config.toml  | 22 ++++++++++++++++++++++
 7 files changed, 51 insertions(+), 11 deletions(-)

diff --git a/.build.yml b/.build.yml
index ad34393..f03feca 100644
--- a/.build.yml
+++ b/.build.yml
@@ -6,6 +6,7 @@ packages:
   - brotli  # for max compression w/ brotli_static, decompressing binaries
   - git  # for Hugo's gitInfo
   - make
+  - jq
   - npm  # for testing with lighthouse and webhint
   - chromium  # for testing with lighthouse and webhint
 environment:
@@ -33,5 +34,7 @@ tasks:
       make DOMAIN=staging.seirdy.one ZOPFLI_ITERATIONS=50 test-staging
       tar czf ~/lighthouse-reports.tar.gz lighthouse-reports
       make clean deploy
+      cp mentions.json ~/mentions.json
 artifacts:
   - lighthouse-reports.tar.gz
+  - mentions.json
diff --git a/.gitignore b/.gitignore
index d07a591..100c2be 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,9 +1,11 @@
-public/
-public_*/
-hint-report/
-.hintrc-*
 *.report.html
 *.report.json
+.hintrc-*
 .lighthouseci/*
+hint-report/
 lighthouse-reports/*
+mentions.json
 node_modules/*
+old/
+public/
+public_*/
diff --git a/.hintrc b/.hintrc
index 1380d3e..3866600 100644
--- a/.hintrc
+++ b/.hintrc
@@ -14,6 +14,7 @@
 	],
 	"hints": {
 		"apple-touch-icons": "off",
+		"axe/aria": "off",
 		"axe/other": "error",
 		"compat-api/css": [
 			"error",
diff --git a/.rsyncignore b/.rsyncignore
index 8b3d00d..fdbabc8 100644
--- a/.rsyncignore
+++ b/.rsyncignore
@@ -1,6 +1,6 @@
+.well-known/
+_*
 misc/
 music.txt
 music.txt.gz
-.well-known
-_*
-webmentions
+webmentions/
diff --git a/Makefile b/Makefile
index a2c7e0a..98dfa37 100644
--- a/Makefile
+++ b/Makefile
@@ -28,7 +28,7 @@ hugo: clean
 
 .PHONY: clean
 clean:
-	rm -rf $(OUTPUT_DIR) .lighthouseci lighthouse-reports
+	rm -rf $(OUTPUT_DIR) .lighthouseci lighthouse-reports mentions.json
 
 .PHONY: lint-css
 lint-css:
@@ -72,14 +72,23 @@ ifndef NO_STATIC
 		| xargs brotli -q 11
 endif
 
+# save webmentions to a file, don't send yet
+mentions.json: hugo
+	# gather old version of the site
+	# rsync $(RSYNCFLAGS) --exclude '*.gz' --exclude '*.br' --exclude '*.png' --exclude-from .rsyncignore $(WWW_RSYNC_DEST)/ old
+	static-webmentions find
+	mv mentions.json mentions.json.unfiltered
+	# filter the webmentions a bit; jq offers more flexibility than config.toml
+	jq '[ .[] | select(.Dest|test("https://(git.sr.ht/~seirdy/seirdy.one/log/master|seirdy.one|web.archive.org|matrix.to)") | not) ]' <mentions.json.unfiltered >mentions.json
+	rm mentions.json.unfiltered
 
 .PHONY: deploy-html
-deploy-html: build
+deploy-html: build mentions.json
 	rsync $(RSYNCFLAGS) --exclude 'gemini' --exclude '*.gmi' --exclude-from .rsyncignore $(OUTPUT_DIR)/ $(WWW_RSYNC_DEST) --delete
 
 .PHONY: deploy-gemini
 deploy-gemini: hugo
-	rsync $(RSYNCFLAGS) --exclude '*.html' --exclude '*.xml' --exclude '*.gz' --exclude-from .rsyncignore $(OUTPUT_DIR)/gemini/ $(OUTPUT_DIR)/about $(OUTPUT_DIR)/posts $(OUTPUT_DIR)/publickey.* $(GEMINI_RSYNC_DEST)/ --delete
+	rsync $(RSYNCFLAGS) --exclude '*.html' --exclude '*.xml' --exclude '*.gz' --exclude '*.br' --exclude-from .rsyncignore $(OUTPUT_DIR)/gemini/ $(OUTPUT_DIR)/about $(OUTPUT_DIR)/posts $(OUTPUT_DIR)/publickey.* $(GEMINI_RSYNC_DEST)/ --delete
 	rsync $(RSYNCFLAGS) $(OUTPUT_DIR)/posts/gemini.xml $(GEMINI_RSYNC_DEST)/feed.xml
 
 .PHONY: deploy
@@ -87,7 +96,7 @@ deploy: deploy-html deploy-gemini
 
 ## stuff for the staging server
 .PHONY: test-staging
-test-staging: deploy-html
+test-staging:
 	yq e '.ci .collect .url | .[]' .lighthouserc.yml | xargs npx hint -f codeframe
 	npx lhci autorun
 
diff --git a/README.md b/README.md
index 28da9e8..b7379d1 100644
--- a/README.md
+++ b/README.md
@@ -42,4 +42,7 @@ To test in CI, after deploying to the staging environment:
 - webhint CLI
 - [lighthouse-ci](https://github.com/GoogleChrome/lighthouse-ci)
 
+CI also runs [static-webmention](https://github.com/nekr0z/static-webmentions) to
+gather a list of WebMentions for me to send and review manually.
+
 See the `Makefile` for details. The CI saves lighthouse reports as a build artifact.
diff --git a/config.toml b/config.toml
index 07474af..dfac2b1 100644
--- a/config.toml
+++ b/config.toml
@@ -78,6 +78,28 @@ path = "gemini/"
 [outputs]
 section = ["HTML", "RSS", "GEMRSS"]
 
+# https://github.com/nekr0z/static-webmentions
+[webmentions]
+newDir = "public"
+oldDir = "old"
+webmentionsFile = "mentions.json"
+excludeSources = [
+  "/tags/*",                    # only trailing * are supported at the moment
+  "/posts/*",                     # this only excludes /posts/index.html, not /posts/somepost/
+  "/",                          # same as "/index.html"
+]
+
+# addresses that we don't want to send webmentions to
+# other schemes and stuff that's I link to too often
+excludeDestinations = [
+  "mailto:",
+	"gemini:",
+	"https://web.archive.org",
+	"https://lists.sr.ht/~seirdy/seirdy.one-comments",
+	"https://useplaintext.email/",
+	"https://seirdy.one",
+]
+
 [server]
 [[server.headers]]
 for = "/**.{css,png,webp,webm}"