install pdf.js and fits through builder images

Dockerfile

@@ -5,12 +5,13 @@
 ##
 FROM ruby:2.7.8-slim-bullseye as spot-base
 
-RUN apt update && \
-    apt install -y --no-install-recommends ca-certificates curl gnupg && \
+RUN apt-get clean && \
+    apt-get update && \
+    apt-get install -y --no-install-recommends ca-certificates curl gnupg && \
     mkdir -p /etc/apt/keyrings && \
     curl -fsSL https://deb.nodesource.com/gpgkey/nodesource-repo.gpg.key | gpg --dearmor -o /etc/apt/keyrings/nodesource.gpg && \
     echo "deb [signed-by=/etc/apt/keyrings/nodesource.gpg] https://deb.nodesource.com/node_20.x nodistro main" | tee /etc/apt/sources.list.d/nodesource.list && \
-    apt update && apt install -y --no-install-recommends \
+    apt-get update -y && apt-get install -y --no-install-recommends \
         awscli \
         build-essential \
         coreutils \
@@ -53,6 +54,7 @@ CMD ["bundle", "exec", "rails", "server", "-b", "ssl://0.0.0.0:443?key=/spot/tmp
 
 HEALTHCHECK CMD curl -skf https://localhost/healthcheck/default || exit 1
 
+
 ##
 #  Target: spot-asset-builder
 #  !! This is a builder image, do not use !!
@@ -63,6 +65,18 @@ COPY . /spot
 
 RUN SECRET_KEY_BASE="$(bin/rake secret)" FEDORA_URL="http://fakehost:8080/rest" bundle exec rake assets:precompile
 
+##
+# TARGET: pdfjs-installer
+# !! This is a builder image. Not for general use !!
+##
+FROM alpine:3 as pdfjs-installer
+ARG PDFJS_VERSION="4.0.379"
+ENV PDFJS_VERSION="${PDFJS_VERSION}"
+ADD https://github.com/mozilla/pdf.js/releases/download/v${PDFJS_VERSION}/pdfjs-${PDFJS_VERSION}-legacy-dist.zip /tmp/pdfjs.zip
+RUN unzip -d /tmp/pdfjs /tmp/pdfjs.zip
+COPY config/pdfjs/web/viewer.html /tmp/pdfjs/web
+COPY config/pdfjs/web/lafayette-viewer-enhancements.js /tmp/pdfjs/web
+
 
 ##
 # TARGET: spot-web-development
@@ -75,12 +89,14 @@ RUN bundle config unset with &&\
     bundle config set with "development:test" && \
     bundle install --jobs "$(nproc)"
 COPY . /spot/
+COPY --from=pdfjs-installer /tmp/pdfjs /spot/public/pdf
 
 ENV RAILS_ENV=development
 
 ENTRYPOINT ["/spot/bin/spot-dev-entrypoint.sh"]
 CMD ["bundle", "exec", "rails", "server", "-b", "ssl://0.0.0.0:443?key=/spot/tmp/ssl/application.key&cert=/spot/tmp/ssl/application.crt"]
 
+
 ##
 # TARGET: spot-web-production
 # Precompiles assets for production
@@ -90,6 +106,23 @@ ENV RAILS_ENV=production
 COPY . /spot
 COPY --from=spot-asset-builder /spot/public/assets /spot/public/assets
 COPY --from=spot-asset-builder /spot/public/uv /spot/public/uv
+COPY --from=pdfjs-installer /tmp/pdfjs /spot/public/pdf
+
+
+##
+# Target: fits-installer
+# !! This is a builder image, do not use directly !!
+#
+# @see https://github.com/harvard-lts/fits
+# @see https://github.com/samvera/hyrax/blob/3.x-stable/Dockerfile#L59-L65
+##
+FROM alpine:3 as fits-installer
+ARG FITS_VERSION="1.6.0"
+ENV FITS_VERSION="${FITS_VERSION}"
+ADD https://github.com/harvard-lts/fits/releases/download/${FITS_VERSION}/fits-${FITS_VERSION}.zip /tmp/fits.zip
+
+RUN unzip -d /tmp/fits /tmp/fits.zip && \
+    chmod a+x /tmp/fits/fits.sh
 
 
 ##
@@ -102,7 +135,7 @@ ENV MALLOC_ARENA_MAX=2
 # We don't need the entrypoint script to generate an SSL cert
 ENV SKIP_SSL_CERT=true
 
-RUN apt update && apt install -y --no-install-recommends \
+RUN apt-get update && apt-get install -y --no-install-recommends \
         bash \
         ffmpeg \
         ghostscript \
@@ -120,21 +153,7 @@ RUN ln -s /usr/bin/python3 /usr/bin/python
 # @see https://stackoverflow.com/questions/52998331/imagemagick-security-policy-pdf-blocking-conversion#comment110879511_59193253
 RUN sed -i '/disable ghostscript format types/,+6d' /etc/ImageMagick-6/policy.xml
 
-# Install FITS based on Hyrax's Dockerfile
-#
-# @see https://github.com/harvard-lts/fits
-# @see https://github.com/samvera/hyrax/blob/3.x-stable/Dockerfile#L59-L65
-ARG FITS_VERSION="1.6.0"
-ENV FITS_VERSION="${FITS_VERSION}"
-
-# (from https://github.com/samvera/hyrax/blob/3.x-stable/Dockerfile#L59-L65)
-RUN mkdir -p /usr/local/fits && \
-    cd /usr/local/fits && \
-    curl -Ls -o fits.zip "https://github.com/harvard-lts/fits/releases/download/${FITS_VERSION}/fits-${FITS_VERSION}.zip" && \
-    unzip fits.zip && \
-    rm fits.zip && \
-    chmod a+x /usr/local/fits/fits.sh
-
+COPY --from=fits-installer /tmp/fits /usr/local/fits
 ENV PATH="${PATH}:/usr/local/fits"
 
 CMD ["bundle", "exec", "sidekiq"]
@@ -149,6 +168,7 @@ RUN bundle config unset with &&\
     bundle install --jobs "$(nproc)"
 COPY . /spot/
 
+
 ##
 # Target: spot-worker-production
 # Copies compiled assets for use in production.

app/helpers/pdf_viewer_helper.rb

@@ -6,22 +6,34 @@ module PdfViewerHelper
   # @param [String] path
   # @return [String] URL to the viewer
   def viewer_url(path)
-    "/web/viewer.html?file=#{path}##{query_param}"
+    "/pdf/web/viewer.html?file=#{path}##{query_param}"
   end
 
-  # The search param used by PDFjs to prepopulate the search bar
+  # The search/page params used by PDFjs to prepopulate the search bar
   #
   # @return [String]
   def query_param
-    return unless search_query
-    "search=#{search_query}&phrase=true"
+    return unless search_query || page_param
+    qp = {}
+    qp[:page] = page_param if page_param
+
+    if search_query
+      qp[:search] = search_query
+      qp[:phrase] = true
+    end
+
+    URI.encode_www_form(qp)
   end
 
   private
 
+  def page_param
+    search&.query_params && search&.query_params&.try(:[], :page)
+  end
+
   # @return [String]
   def search_query
-    search&.query_params && search.query_params[:q]
+    search&.query_params && search.query_params&.try(:[], :q)
   end
 
   def search

config/initializers/mime_types.rb

@@ -7,3 +7,5 @@
 Mime::Type.register "application/ld+json", :jsonld
 Mime::Type.register "text/turtle", :ttl
 Mime::Type.register 'application/x-endnote-refer', :endnote
+
+Rack::Mime::MIME_TYPES['.mjs'] = 'application/javascript'

config/pdfjs/web/lafayette-viewer-enhancements.js

@@ -0,0 +1,18 @@
+// fills in the findInput search box with the
+// provided query term (when present), only on load
+document.addEventListener('pagerendered', function (ev) {
+  var findInput = document.getElementById('findInput')
+  var app = PDFViewerApplication
+
+  app.eventBus.on('pagesloaded', function () {
+    var findController = app.findController || {}
+    var state = findController.state || {}
+    var query = state.query || null
+
+    if (!query) {
+      return
+    }
+
+    findInput.value = query
+  })
+})