Established aclanthology.org as canonical site (acl-org#1328)

* changed canonical to aclanthology.org
* straightened out mirror and main uploads
* removed mirror target since it will just be redirects
* updated host target in some support scripts
* Adapt .htaccess to new canonical host
  - rewrite from www.aclanthology.org → aclanthology.org
  - http → https
  - make anthology-files a symlink so we don't have to modify its location in the .htaccess file
* moved thumbnail script

Co-authored-by: Arne Köhn <[email protected]>

.github/workflows/publish.yml

@@ -1,14 +1,14 @@
-# Publishes the build to our main site at aclweb.org/anthology
+# Publishes the build to our current mirror, aclanthology.org
 
-name: publish
+name: publish-aclanthology
 
 on:
   push:
     branches:
       - master
 
 jobs:
-  publish:
+  publish-aclanthology:
     runs-on: ubuntu-20.04
     steps:
     - name: install hugo
@@ -27,12 +27,9 @@ jobs:
     - uses: actions/checkout@v1
     - name: build
       env:
-        ANTHOLOGY_PREFIX: https://www.aclweb.org/anthology
+        ANTHOLOGY_PREFIX: https://aclanthology.org
       run: |
         make ANTHOLOGY_PREFIX=${ANTHOLOGY_PREFIX} check site
     - name: publish
-      env:
-        PUBLISH_TARGET: 50.87.169.12:anthology-static
-        ANTHOLOGY_PREFIX: https://www.aclweb.org/anthology
       run: |
         make ANTHOLOGY_PREFIX=${ANTHOLOGY_PREFIX} upload

Makefile

@@ -29,14 +29,14 @@ SHELL := /bin/bash
 # If you want to host the anthology on your own, set ANTHOLOGY_PREFIX
 # in your call to make to your prefix, e.g.
 #
-#     ANTHOLOGY_PREFIX="https://aclanthology.org" make
+#     ANTHOLOGY_PREFIX="https://www.aclweb.org/anthology" make
 #
 # PLEASE NOTE that the prefix cannot contain any '#' character, or a Perl regex
 # below will fail.
 # The following line ensures that it is exported as an environment variable
 # for all sub-processes:
 
-export ANTHOLOGY_PREFIX ?= https://www.aclweb.org/anthology
+export ANTHOLOGY_PREFIX ?= https://aclanthology.org
 
 SLASHATEND:=$(shell echo ${ANTHOLOGY_PREFIX} | grep -q '/$$'; echo $$?)
 
@@ -58,7 +58,7 @@ endif
 # Root location for PDF and attachment hierarchy.
 # This is the directory where you have to put all the papers and attachments.
 # Easiest if the server can just serve them from /anthology-files.
-ANTHOLOGYFILES ?= /anthology-files
+ANTHOLOGYFILES ?= /var/www/anthology-files
 
 HUGO_ENV ?= production
 
@@ -245,8 +245,8 @@ build/.hugo: build/.static build/.pages build/.bibtex build/.mods build/.endnote
 	         --cleanDestinationDir \
 	         --minify
 	@cd build/website/$(ANTHOLOGYDIR) \
-	    && perl -i -pe 's|ANTHOLOGYDIR|$(ANTHOLOGYDIR)|g' .htaccess \
-	    && perl -i -pe 's|ANTHOLOGYFILES|$(ANTHOLOGYFILES)|g' .htaccess
+	    && ln -s $(ANTHOLOGYFILES) anthology-files \
+	    && perl -i -pe 's|ANTHOLOGYDIR|$(ANTHOLOGYDIR)|g' .htaccess
 	@touch build/.hugo
 
 .PHONY: mirror
@@ -306,20 +306,14 @@ serve:
 	 @echo "INFO     Starting a server at http://localhost:8000/"
 	 @cd build/website && python3 -m http.server 8000
 
-# this target does not use ANTHOLOGYDIR because the official website
-# only works if ANTHOLOGYDIR == anthology.
+# Main site: aclanthology.org. Requires ANTHOLOGYDIR to be unset.
 .PHONY: upload
 upload:
-	@if [ $(ANTHOLOGYDIR) != "anthology" ]; then \
-            echo "WARNING: Can't upload because ANTHOLOGYDIR was set to '${ANTHOLOGYDIR}' instead of 'anthology'"; \
+	@if [[ $(ANTHOLOGYDIR) != "" ]]; then \
+            echo "WARNING: Can't upload because ANTHOLOGYDIR was set to '${ANTHOLOGYDIR}' instead of being empty"; \
             exit 1; \
         fi
-	@echo "INFO     Running rsync for main site..."
-	@rsync -aze "ssh -o StrictHostKeyChecking=accept-new" --delete build/website/anthology/ [email protected]:anthology-static
-
-.PHONY: upload-mirror
-upload-mirror:
-	@echo "INFO     Running rsync for aclanthology.org mirror..."
+	@echo "INFO     Running rsync for main site (aclanthology.org)..."
 	@rsync -aze "ssh -o StrictHostKeyChecking=accept-new" build/website/ [email protected]:/var/www/aclanthology.org
 
 # Push a preview to the mirror

bin/anthology/data.py

@@ -25,7 +25,7 @@
 
 # this is the canonical URL.  In contrast to all other
 # URL templates, it always links to the official anthology.
-CANONICAL_URL_TEMPLATE = "https://www.aclweb.org/anthology/{}"
+CANONICAL_URL_TEMPLATE = "https://aclanthology.org/{}"
 
 # the prefix is used in different programs and we need to set it everywhere
 # We use a environment variable to set this and not have to forward the value
@@ -34,7 +34,7 @@
 try:
     ANTHOLOGY_PREFIX = os.environ["ANTHOLOGY_PREFIX"]
 except:
-    ANTHOLOGY_PREFIX = "https://www.aclweb.org/anthology"
+    ANTHOLOGY_PREFIX = "https://aclanthology.org"
 
 ATTACHMENT_PREFIX = ANTHOLOGY_PREFIX + "/attachments"
 ATTACHMENT_TEMPLATE = ATTACHMENT_PREFIX + "/{}"

bin/create_mirror.py

@@ -22,7 +22,7 @@
 fetch everything.
 
 Options:
-  --source=SRC               where to fetch the files from [default: https://www.aclweb.org/anthology]
+  --source=SRC               where to fetch the files from [default: https://aclanthology.org]
   --to=DIR                 Directory to write files to [default: {scriptdir}/../build/anthology-files]
   --only-papers            Do not mirror attachments, only papers.
   --debug                  Output debug-level log messages.

bin/create_sitemapindex.sh

@@ -22,7 +22,7 @@ echo '<sitemapindex xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">'
 
 for f in "$@"; do
 	echo '  <sitemap>'
-    echo '    <loc>https://www.aclweb.org/anthology/'$f'</loc>'
+    echo '    <loc>https://aclanthology.org/'$f'</loc>'
 	echo '  </sitemap>'
 done
 echo '</sitemapindex>'

bin/upload_files.py

@@ -43,14 +43,11 @@
 from anthology.utils import deconstruct_anthology_id
 from typing import List
 
-# Name for the SSH alias in ~/.ssh/config.
-SSH_CONFIG_TARGET = "aclweb"
-
 # The root directory for files, currently containing pdf/ and attachments/
-ACLWEB_FILE_ROOT = "/home3/aclwebor/anthology-files"
+ACLWEB_FILE_ROOT = "/home/anthologizer/anthology-files"
 
-# The ssh shortcut or full hostname
-ACLWEB_HOST = "aclweb"
+# The ssh shortcut (in ~/.ssh/config) or full hostname
+ACLWEB_HOST = "anth"
 
 
 def get_dest_path(filepath: str):

hugo/static/.htaccess

@@ -28,30 +28,22 @@ RewriteEngine On
 # The redirects are only applicable to the official aclweb.org anthology; all the
 # 301 redirects are guarded with a RewriteCond.
 
-# Redirect non-canonical URL bases to https://aclweb.org/anthology/
-RewriteCond %{HTTP_HOST} ^anthology\.aclweb\.org$ [OR]
-RewriteCond %{HTTP_HOST} ^www\.anthology\.aclweb\.org$
-RewriteRule ^/?(.*) https://www.aclweb.org/anthology/$1 [R=301,L]
-
-# Now we know we are at the canonical URL.
-# Now make sure that we redirect from non-HTTPS to HTTPS.
-# **Note**: at this point, paths are relative to /anthology, that is, they *do not include* the /anthology/ text.
-RewriteCond %{HTTP_HOST} aclweb.org
 RewriteCond %{HTTPS} !=on [OR]
-RewriteCond %{HTTP_HOST} !^www\. [NC]
-RewriteRule ^(.*)$ https://www.aclweb.org/anthology/$1 [R=301,L]
+RewriteCond %{HTTP_HOST} www\.aclanthology\.org [NC]
+RewriteRule ^(.*)$ https://aclanthology.org/$1 [R=301,L]
+
 
 ## Nested file paths
 # Redirect old nested file paths (e.g., P/P17/P17-1069.pdf -> https://www.aclweb.org/anthology/P17-1069.pdf)
 # Note that since capture patterns can't be reused in the capture portion of the string, this is a leaky match
-# that will also redirect X/Z19/P17-1069.pdf -> ANTHOLOGYFILES/pdf/P/P17/P17-1069.pdf
-RewriteCond %{HTTP_HOST} ^www\.aclweb\.org$
-RewriteRule ^[A-Za-z]/[A-Za-z]\d{2}/([A-Za-z])(\d{2})\-(\d{4})(/|\.[a-z]+)?$ https://www.aclweb.org/anthology/$1$2-$3$4 [R=301,L]
+# that will also redirect X/Z19/P17-1069.pdf -> anthology-files/pdf/P/P17/P17-1069.pdf
+RewriteCond %{HTTP_HOST} ^aclanthology\.org$
+RewriteRule ^[A-Za-z]/[A-Za-z]\d{2}/([A-Za-z])(\d{2})\-(\d{4})(/|\.[a-z]+)?$ https://aclanthology.org/$1$2-$3$4 [R=301,L]
 
 # Redirect nested paper pages to the canonical location (e.g., papers/P/P19/P19-1001/ -> P19-1001)
 # This way there is only one page. Should maintain for backward compatibility for some time after August 2019.
-RewriteCond %{HTTP_HOST} ^www\.aclweb\.org$
-RewriteRule ^papers/[A-Za-z]/[A-Za-z]\d{2}/([A-Za-z])(\d{2})\-(\d{4})/?$ https://www.aclweb.org/anthology/$1$2-$3 [R=301,L]
+RewriteCond %{HTTP_HOST} ^aclanthology\.org$
+RewriteRule ^papers/[A-Za-z]/[A-Za-z]\d{2}/([A-Za-z])(\d{2})\-(\d{4})/?$ https://aclanthology.org/$1$2-$3 [R=301,L]
 
 #
 # INTERNAL RETRIEVALS
@@ -66,20 +58,20 @@ RewriteRule ^([A-Za-z]\d{2}\-\d{1,2})/?$ /ANTHOLOGYDIR/volumes/$1 [L,NC]
 RewriteRule ^(\d{4}\.[a-zA-Z\d]+\-[a-zA-Z\d]+)/?$ /ANTHOLOGYDIR/volumes/$1 [L,NC]
 
 # Volume URLs (e.g., P17-1.pdf loads P/P17/P17-1.pdf, 2020.acl-main loads acl/2020.acl-main.pdf)
-RewriteRule ^([A-Za-z])(\d{2})\-(\d{1,2})\.pdf$ ANTHOLOGYFILES/pdf/$1/$1$2/$1$2-$3.pdf [L,NC]
-RewriteRule ^(\d{4})\.([a-zA-Z\d]+)\-([a-zA-Z\d]+)\.pdf$ ANTHOLOGYFILES/pdf/$2/$1.$2-$3.pdf [L,NC]
+RewriteRule ^([A-Za-z])(\d{2})\-(\d{1,2})\.pdf$ anthology-files/pdf/$1/$1$2/$1$2-$3.pdf [L,NC]
+RewriteRule ^(\d{4})\.([a-zA-Z\d]+)\-([a-zA-Z\d]+)\.pdf$ anthology-files/pdf/$2/$1.$2-$3.pdf [L,NC]
 
 # PDF link, revisions, and errata (P17-1069[v2].pdf loads P/P17/P17-1069[v2].pdf --- with "v2" optional)
 # TODO: decide on a new format for revisions and errata
-RewriteRule ^([A-Za-z])(\d{2})\-(\d{4})([ve]\d+)?\.pdf$ ANTHOLOGYFILES/pdf/$1/$1$2/$1$2-$3$4.pdf [L,NC]
-RewriteRule ^(\d{4})\.([a-zA-Z\d]+)\-([a-zA-Z\d]+)\.(\d+([ve]\d+)?)\.pdf$ ANTHOLOGYFILES/pdf/$2/$1.$2-$3.$4.pdf [L,NC]
+RewriteRule ^([A-Za-z])(\d{2})\-(\d{4})([ve]\d+)?\.pdf$ anthology-files/pdf/$1/$1$2/$1$2-$3$4.pdf [L,NC]
+RewriteRule ^(\d{4})\.([a-zA-Z\d]+)\-([a-zA-Z\d]+)\.(\d+([ve]\d+)?)\.pdf$ anthology-files/pdf/$2/$1.$2-$3.$4.pdf [L,NC]
 
-# Attachments (e.g., P17-1069.Poster.pdf loads ANTHOLOGYFILES/attachments/P/P17/P17-1069.Poster.pdf)
-RewriteRule ^attachments/([A-Za-z])(\d{2})\-(\d{4})(\..*)?$ ANTHOLOGYFILES/attachments/$1/$1$2/$1$2-$3$4 [L,NC]
-RewriteRule ^attachments/(\d{4})\.([a-zA-Z\d]+)\-([a-zA-Z\d]+\.\d+)\.(.*)$ ANTHOLOGYFILES/attachments/$2/$1.$2-$3.$4 [L,NC]
+# Attachments (e.g., P17-1069.Poster.pdf loads anthology-files/attachments/P/P17/P17-1069.Poster.pdf)
+RewriteRule ^attachments/([A-Za-z])(\d{2})\-(\d{4})(\..*)?$ anthology-files/attachments/$1/$1$2/$1$2-$3$4 [L,NC]
+RewriteRule ^attachments/(\d{4})\.([a-zA-Z\d]+)\-([a-zA-Z\d]+\.\d+)\.(.*)$ anthology-files/attachments/$2/$1.$2-$3.$4 [L,NC]
 
-# Thumbnails (e.g., /thumb/P17-1069.jpg loads ANTHOLOGYFILES/thumb/P17-1069.jpg)
-RewriteRule ^thumb/(.*)$ ANTHOLOGYFILES/thumb/$1 [L,NC]
+# Thumbnails (e.g., /thumb/P17-1069.jpg loads anthology-files/thumb/P17-1069.jpg)
+RewriteRule ^thumb/(.*)$ anthology-files/thumb/$1 [L,NC]
 
 # Author pages (e.g., people/arya-d-mccarthy loads people/a/arya-d-mccarthy)
 RewriteRule ^people/([a-z])([\-a-z0-9]*)$ people/$1/$1$2/ [L,NC]