From 6901c66bcfc911d5f25a85bce1efa1f6033c6aad Mon Sep 17 00:00:00 2001
From: jim kolpack <james.kolpack@gmail.com>
Date: Fri, 3 Jul 2026 19:32:04 -0400
Subject: [PATCH] Skip already-processed ZIPs by sampling first JPEG comment

Extracts one JPEG from each ZIP to check for the processing comment
before doing a full extract/resize/rezip cycle. Shared constants moved
to config.sh as single source of truth. Trap and cleanup extended to
cover the sample temp file.
---
 config.sh            |  5 +++++
 images_resize.sh     |  6 +++---
 zip_images_resize.sh | 30 +++++++++++++++++++++++++++---
 3 files changed, 35 insertions(+), 6 deletions(-)
 create mode 100644 config.sh

diff --git a/config.sh b/config.sh
new file mode 100644
index 0000000..e3124ed
--- /dev/null
+++ b/config.sh
@@ -0,0 +1,5 @@
+#!/bin/bash
+# Shared configuration for image processing scripts
+
+MAX_PIXELS=8000000
+MAGICK_COMMENT="magick:resize=8mp,quality=85,sampling=4:2:0"
diff --git a/images_resize.sh b/images_resize.sh
index 7f91fbf..dc75643 100755
--- a/images_resize.sh
+++ b/images_resize.sh
@@ -5,10 +5,10 @@
 #
 # Usage: xxx.sh
 
-echo "Resizing large JPGs in $(pwd)"
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+source "$SCRIPT_DIR/config.sh"
 
-MAX_PIXELS=8000000
-MAGICK_COMMENT="magick:resize=8mp,quality=85,sampling=4:2:0"
+echo "Resizing large JPGs in $(pwd)"
 
 find . -regextype posix-egrep -regex ".*\.(jpg|JPG|JPEG|jpeg)$" -print0 | while IFS= read -r -d '' f
 do
diff --git a/zip_images_resize.sh b/zip_images_resize.sh
index 5b5d755..9447991 100755
--- a/zip_images_resize.sh
+++ b/zip_images_resize.sh
@@ -11,6 +11,7 @@ if [ $# -eq 0 ]; then
 fi
 
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+source "$SCRIPT_DIR/config.sh"
 
 echo "Resizing large JPGs in ZIPs"
 
@@ -19,6 +20,9 @@ WORKDIR="/dev/shm/"
 
 # Cleanup function
 cleanup_temp() {
+    if [ -n "$SAMPLE" ] && [ -f "$SAMPLE" ]; then
+        rm -f "$SAMPLE"
+    fi
     if [ -n "$TEMPDIR" ] && [ -d "$TEMPDIR" ]; then
         rm -r "$TEMPDIR"
     fi
@@ -35,6 +39,29 @@ for INFILE in "$@"; do
     INFILE="$(realpath "$INFILE")"
     echo "Processing file $CURRENT_FILE of $TOTAL_FILES: $INFILE"
 
+    SAMPLE=""
+    TEMPDIR=""
+    trap cleanup_temp EXIT INT TERM
+
+    # Check if already processed by sampling the first JPEG in the archive
+    FIRST_JPEG=$(unzip -l "$INFILE" 2>/dev/null | grep -iE '\.(jpg|jpeg)$' | head -1 | awk '{print $NF}')
+    if [ -n "$FIRST_JPEG" ]; then
+        SAMPLE=$(mktemp -p "$WORKDIR" --suffix=.jpg)
+        if unzip -p "$INFILE" "$FIRST_JPEG" > "$SAMPLE" 2>/dev/null; then
+            COMMENT=$(~/Applications/magick identify -format "%c" "$SAMPLE" 2>/dev/null)
+            rm -f "$SAMPLE"
+            SAMPLE=""
+            if [[ "$COMMENT" == "$MAGICK_COMMENT" ]]; then
+                echo "Skipping $INFILE: already processed"
+                trap - EXIT INT TERM
+                continue
+            fi
+        else
+            rm -f "$SAMPLE"
+            SAMPLE=""
+        fi
+    fi
+
     OUTFILE="$(realpath "${INFILE%.zip}-compress.zip")"
 
     # Remove output file if it already exists
@@ -45,9 +72,6 @@ for INFILE in "$@"; do
     # Set name for the temp dir. This directory will be created under WORKDIR
     TEMPDIR=$(mktemp -p "$WORKDIR" -d)
 
-    # Set up cleanup trap
-    trap cleanup_temp EXIT INT TERM
-
     # Create a temporary folder for unzipped files
     echo "Extracting $INFILE"