From 14b22769b97888979c8d8d5c02a918a8bc685c15 Mon Sep 17 00:00:00 2001
From: Joe Corall <jjc223@lehigh.edu>
Date: Fri, 15 Nov 2024 06:44:12 -0500
Subject: [PATCH] Add isle-leptonica to stack (#52)

---
 Dockerfile                     |  2 +-
 examples/imagemagick/README.md |  8 ++++++++
 examples/mergepdf/Dockerfile   | 35 ++++++++++++++++++----------------
 examples/mergepdf/README.md    |  7 +++++++
 examples/mergepdf/cmd.sh       |  4 ++--
 examples/tesseract/Dockerfile  |  7 +++++--
 examples/tesseract/README.md   |  8 ++++++++
 7 files changed, 50 insertions(+), 21 deletions(-)
 create mode 100644 examples/imagemagick/README.md
 create mode 100644 examples/tesseract/README.md

diff --git a/Dockerfile b/Dockerfile
index dd598d5..75d40af 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -25,7 +25,7 @@ RUN adduser -S -G nobody scyllaridae
 
 RUN apk update && \
     apk add --no-cache \
-      curl==8.10.1-r0 \
+      curl==8.11.0-r2 \
       bash==5.2.26-r0 \
       ca-certificates==20240705-r0 \
       openssl==3.3.2-r1
diff --git a/examples/imagemagick/README.md b/examples/imagemagick/README.md
new file mode 100644
index 0000000..979be7a
--- /dev/null
+++ b/examples/imagemagick/README.md
@@ -0,0 +1,8 @@
+# imagemagick
+
+imagemagick as a service
+
+## Attribution
+
+- [imagemagick](https://github.com/ImageMagick/ImageMagick) for image file manipulation
+- [islandora/imagemagick](https://github.com/Islandora-Devops/isle-imagemagick/) for a custom Alpine APK imagemagick package.
diff --git a/examples/mergepdf/Dockerfile b/examples/mergepdf/Dockerfile
index edd437b..c71b84b 100644
--- a/examples/mergepdf/Dockerfile
+++ b/examples/mergepdf/Dockerfile
@@ -1,23 +1,26 @@
 ARG TAG=main
 ARG DOCKER_REPOSITORY=local
+FROM islandora/leptonica:alpine-3.20.2-leptonica-1.84.1-r0 AS leptonica
 FROM ${DOCKER_REPOSITORY}/scyllaridae-imagemagick:${TAG} AS scyllaridae
 
-RUN apk update && \
-  apk add --no-cache \
-    ghostscript==10.04.0-r0 \
-    jq==1.7.1-r0 \
-    leptonica-dev==1.84.1-r0 \
-    tesseract-ocr==5.3.4-r0 \
-    tesseract-ocr-data-eng==5.3.4-r0 \
-    tesseract-ocr-data-fra==5.3.4-r0 \
-    tesseract-ocr-data-spa==5.3.4-r0 \
-    tesseract-ocr-data-ita==5.3.4-r0 \
-    tesseract-ocr-data-por==5.3.4-r0 \
-    tesseract-ocr-data-hin==5.3.4-r0 \
-    tesseract-ocr-data-deu==5.3.4-r0 \
-    tesseract-ocr-data-jpn==5.3.4-r0 \
-    tesseract-ocr-data-rus==5.3.4-r0 \
-    poppler-utils==24.02.0-r1
+RUN --mount=type=bind,from=leptonica,source=/packages,target=/packages \
+    --mount=type=bind,from=leptonica,source=/etc/apk/keys,target=/etc/apk/keys \
+    apk update && \
+    apk add --no-cache \
+        /packages/leptonica-*.apk \
+        ghostscript==10.04.0-r0 \
+        jq==1.7.1-r0 \
+        tesseract-ocr==5.3.4-r0 \
+        tesseract-ocr-data-eng==5.3.4-r0 \
+        tesseract-ocr-data-fra==5.3.4-r0 \
+        tesseract-ocr-data-spa==5.3.4-r0 \
+        tesseract-ocr-data-ita==5.3.4-r0 \
+        tesseract-ocr-data-por==5.3.4-r0 \
+        tesseract-ocr-data-hin==5.3.4-r0 \
+        tesseract-ocr-data-deu==5.3.4-r0 \
+        tesseract-ocr-data-jpn==5.3.4-r0 \
+        tesseract-ocr-data-rus==5.3.4-r0 \
+        poppler-utils==24.02.0-r1
 
 COPY . /app
 
diff --git a/examples/mergepdf/README.md b/examples/mergepdf/README.md
index fad2c1f..9eee2a1 100644
--- a/examples/mergepdf/README.md
+++ b/examples/mergepdf/README.md
@@ -1,3 +1,10 @@
 # mergepdf
 
 Merge children of a paged content item into a single PDF
+
+## Attribution
+
+- [imagemagick](https://github.com/ImageMagick/ImageMagick) for image file manipulation
+- [tesseract](https://github.com/tesseract-ocr/tesseract) OCR engine
+- [islandora/imagemagick](https://github.com/Islandora-Devops/isle-imagemagick/) for a custom Alpine APK imagemagick package.
+- [islandora/leptonica](https://github.com/Islandora-Devops/isle-leptonica/) for a custom Alpine APK leptonica package.
diff --git a/examples/mergepdf/cmd.sh b/examples/mergepdf/cmd.sh
index dc84797..4ed3564 100755
--- a/examples/mergepdf/cmd.sh
+++ b/examples/mergepdf/cmd.sh
@@ -4,11 +4,11 @@ set -eou pipefail
 
 TMP_DIR=$(mktemp -d)
 I=0
-MAX_THREADS=7
+MAX_THREADS=${MAX_THREADS:-5}
 PIDS=()
 
 # iterate over all images in the IIIF manifest
-URLS=$(curl -s "$1/book-manifest" | jq -r '.sequences[0].canvases[].images[0].resource."@id"')
+URLS=$(curl -s "$1/book-manifest" | jq -r '.sequences[0].canvases[].images[0].resource."@id"' | awk -F '/' '{print $7}'|sed -e 's/%2F/\//g' -e 's/%3A/:/g')
 while read -r URL; do
   # If we have reached the max thread limit, wait for any one job to finish
   if [ "${#PIDS[@]}" -ge "$MAX_THREADS" ]; then
diff --git a/examples/tesseract/Dockerfile b/examples/tesseract/Dockerfile
index 434dde1..3811632 100644
--- a/examples/tesseract/Dockerfile
+++ b/examples/tesseract/Dockerfile
@@ -1,10 +1,13 @@
 ARG TAG=main
 ARG DOCKER_REPOSITORY=local
+FROM islandora/leptonica:alpine-3.20.2-leptonica-1.84.1-r0 AS leptonica
 FROM ${DOCKER_REPOSITORY}/scyllaridae:${TAG}
 
-RUN apk update && \
+RUN --mount=type=bind,from=leptonica,source=/packages,target=/packages \
+    --mount=type=bind,from=leptonica,source=/etc/apk/keys,target=/etc/apk/keys \
+    apk update && \
     apk add --no-cache \
-        leptonica-dev==1.84.1-r0 \
+        /packages/leptonica-*.apk \
         tesseract-ocr==5.3.4-r0 \
         tesseract-ocr-data-eng==5.3.4-r0 \
         tesseract-ocr-data-fra==5.3.4-r0 \
diff --git a/examples/tesseract/README.md b/examples/tesseract/README.md
new file mode 100644
index 0000000..4ba308c
--- /dev/null
+++ b/examples/tesseract/README.md
@@ -0,0 +1,8 @@
+# tesseract
+
+Tesseract as a service
+
+## Attribution
+
+- [tesseract](https://github.com/tesseract-ocr/tesseract) OCR engine
+- [islandora/leptonica](https://github.com/Islandora-Devops/isle-leptonica/) for a custom Alpine APK leptonica package.