From: Steven Baltakatei Sandoval <baltakatei@gmail.com>
Date: Fri, 24 Apr 2026 04:13:19 +0000 (+0000)
Subject: feat(user/htmlz_to_cbz.sh): Support png files
X-Git-Url: https://zdv2.bktei.com/gitweb/BK-2020-03.git/commitdiff_plain/10992ec40e142ab5365624925c1dc95fcdfceca1?ds=inline

feat(user/htmlz_to_cbz.sh): Support png files

- feat(user/cbz_dedup.sh): Use jdupes to dedupe CBZ files
---

diff --git a/user/cbz_dedup.sh b/user/cbz_dedup.sh
new file mode 100644
index 0000000..fbcb34b
--- /dev/null
+++ b/user/cbz_dedup.sh
@@ -0,0 +1,73 @@
+#!/bin/bash
+# Desc: Deduplicates files within a CBZ file
+# Usage: cbz_dedup.sh [CBZ file]
+# Example: cbz_dedup.sh input.cbz
+# Version: 0.0.5
+# Depends: jdupes 1.27.3, unzip 6.00 by Debian
+
+
+yell() { echo "$0: $*" >&2; } # print script path and all args to stderr
+die() { yell "$*"; exit 111; } # same as yell() but non-zero exit status
+must() { "$@" || die "cannot $*"; } # runs args as command, reports args if command fails
+showUsage() {
+    # Desc: Display script usage information
+    # Usage: showUsage
+    # Version 0.0.2
+    # Input: none
+    # Output: stdout
+    # Depends: GNU-coreutils 8.30 (cat)
+    cat <<'EOF'
+    USAGE:
+        cbz_dedup.sh [FILE]
+
+    EXAMPLE:
+      cbz_dedup.sh input.cbz
+EOF
+}; # Display information on how to use this script.
+main() {
+     fout="output.cbz";
+    
+     # Check args
+     if [[ $# -ne 1 ]]; then showUsage; die "FATAL:Invalid arg count (should be 1):$#"; fi;
+     re1='\.cbz$';
+     if [[ ! "$1" =~ $re1 ]]; then showUsage; die "FATAL:Not a .cbz file:$1"; fi;
+     if [[ ! -f "$1" ]]; then die "FATAL:Not a file:$1"; fi;
+     fin="$1";
+     dir_tmp="$(mktemp -d)";
+     if [[ ! -d "$dir_tmp" ]]; then die "FATAL:Could not make temporary directory:${dir_tmp}"; fi;
+     trap 'rm -rf "${dir_tmp}"' EXIT;
+     pout="${dir_tmp}/${fout}";
+
+     # Check depends
+     for _cmd in unzip zip jdupes; do
+         if ! command -v "$_cmd" 1>/dev/random 2>&1; then
+             die "FATAL:Missing app:${_cmd}";
+         fi;
+     done;
+    
+     # Extract CBZ to temp dir
+     must unzip "$fin" -x / -d "$dir_tmp";
+
+     # Dedupe with jdupes
+     ## Check size before
+     tmp_size1="$(du -bd0 "$dir_tmp" | cut -f1; )";
+     must jdupes -dN "${dir_tmp}";
+     ## Check size after
+     tmp_size2="$(du -bd0 "$dir_tmp" | cut -f1; )";
+
+     # Replace CBZ if size changed
+     if [[ "$tmp_size1" -eq "$tmp_size2" ]]; then yell "STATUS:No deduplication detected."; return 0; fi;
+     
+     ## Recreate CBZ
+     must zip -j "$pout" "${dir_tmp}"/*;
+
+     ## Preserve original CBZ
+     must mv -n "$fin" "${fin%.*}_original.cbz";
+     
+     ## Move deduped CBZ
+     must mv -n "$pout" "$fin";
+
+     return 0;
+};
+
+main "$@";
diff --git a/user/htmlz_to_cbz.sh b/user/htmlz_to_cbz.sh
index b1c450e..de313f0 100755
--- a/user/htmlz_to_cbz.sh
+++ b/user/htmlz_to_cbz.sh
@@ -1,6 +1,6 @@
 #!/bin/bash
 # Desc: Collects .jpg/jpeg files from a Calibre .htmlz file into .cbz files
-# Version: 0.1.0
+# Version: 0.1.2
 # Depends: jdupes 1.27.3
 
 for fin in ./*.htmlz; do
@@ -8,31 +8,31 @@ for fin in ./*.htmlz; do
         dout="${fin%.*}";
         unzip "$fin" -x / -d "$dout";
         pushd "$dout";
-        mapfile -t images < <(cat index.html  | grep -E "(.jpg|.jpeg)" | sed -E -e 's#.+(images/[0-9]+.(jpeg|jpg)).+#\1#' | uniq; );
-        dout="./output";
-        if [[ -d "$dout" ]]; then
-            rm -r "$dout";
+        mapfile -t images < <(cat index.html  | grep -E "(.jpg|.jpeg|.png)" | sed -E -e 's#.+(images/[0-9]+.(jpeg|jpg|png)).+#\1#' | uniq; );
+        dout2="./output";
+        if [[ -d "$dout2" ]]; then
+            rm -r "$dout2";
         fi;
-        mkdir "$dout";
+        mkdir "$dout2";
         n=1;
         for path in "${images[@]}"; do
-            fnew="${dout}/$(printf "%06d" "$n").jpg";
+            fnew="${dout2}/$(printf "%06d" "$n").jpg";
             cp "$path" "$fnew";
             ((n++));
         done;
         # Add cover file if present
         if [[ -f cover.jpg ]]; then
-            cp -n cover.jpg "${dout}/000000.jpg";
+            cp -n cover.jpg "${dout2}/000000.jpg";
         fi;
         # Remove duplicate images
         if command -v jdupes 1>/dev/random 2>&1; then
-            jdupes -dN "$dout";
+            jdupes -dN "$dout2";
         fi;
         faout="output.cbz";
         if [[ -f "$faout" ]]; then
             rm "$fout";
         fi;
-        zip -j output.cbz "$dout"/*;
+        zip -j output.cbz "$dout2"/*;
     ) &
 done;
 wait && echo "STATUS:Finished." 1>&2;