pak parser in bash

3 files changed, 107 insertions, 0 deletions

diff --git a/parse-pak/README.md b/parse-pak/README.md
new file mode 100644
index 0000000..b241755
--- /dev/null
+++ b/parse-pak/README.md
@@ -0,0 +1,3 @@
+# Parses google chrome .pak resource files
+
+   ./parse-pak.sh en-US.pak
diff --git a/parse-pak/en-US.pak b/parse-pak/en-US.pak
new file mode 100644
index 0000000..689c22f
--- /dev/null
+++ b/parse-pak/en-US.pak
diff --git a/parse-pak/parse-pak.sh b/parse-pak/parse-pak.sh
new file mode 100755
index 0000000..9b5326e
--- /dev/null
+++ b/parse-pak/parse-pak.sh
@@ -0,0 +1,104 @@
+#!/bin/bash
+# Parse google-chrome .pak resourcefiles
+
+parse_int() {
+    ( echo "ibase=16; "; cat; echo -e "\n") | bc
+}
+
+endian_conversion() {
+    if [[ "$1" == "little" ]]; then
+        tac
+    else
+        cat
+    fi
+}
+
+read_bytes() {
+    local length=$1
+    local endian=$2
+    local i=0
+    while [ $i -lt $length ]; do
+        read -r -d ' ' byte
+        [ "$byte" == "" ] && read -r -d ' ' byte
+        echo $byte
+        let i+=1
+    done \
+        | endian_conversion "$endian" \
+        | tr -d '\n'
+}
+
+parse() {
+    local format=$1
+    local endian=big
+    local i=0
+    while [[ $i -lt ${#format} ]]; do
+        case "${format:$i:1}" in
+            "<") endian="little";;
+            ">") endian="big";;
+            "I") read_bytes "4" "$endian" | parse_int; let offset+=4;;
+            "H") read_bytes "2" "$endian" | parse_int; let offset+=2;;
+            "B") read_bytes "1" "$endian" | parse_int; let offset+=1;;
+            *)
+                echo "Invalid format: ${format:$i:1}" >&2;;
+        esac
+        let i+=1
+    done | tr '\n' ' '
+}
+
+read_file() {
+    local file=$1
+    local offset=$2
+    local format=$3
+    dd if=$file bs=1 skip=$offset count=$(format_len "$format") 2>/dev/null \
+        | od --output-duplicates --format=x1 --address-radix=n \
+        | tr 'a-z' 'A-Z' | tr -d '\n'
+}
+
+parse_file() {
+    read_file "$1" "$2" "$3" | parse "$3"
+}
+
+format_len() {
+    local format=$1
+    local len=0
+    local i=0
+    while [[ $i -lt ${#format} ]]; do
+        case "${format:$i:1}" in
+            "I") let len+=4 ;;
+            "H") let len+=2 ;;
+            "B") let len+=1 ;;
+            "<"|">") ;;
+            *) echo "Invalid format: ${format:$i:1}" >&2;;
+        esac
+        let i+=1
+    done
+    echo "$len"
+}
+
+##################
+INPUT=$1
+[ -e "$INPUT" ] || exit 1
+
+header_size=$(format_len "<IIB")
+entry_size=$(format_len "<HI")
+read -r version num_entries encoding < \
+    <(parse_file "$INPUT" 0 "<IIB")
+
+if [ "$version" != "4" ]; then
+    echo "Invalid version: $version" >&2
+    exit 1
+fi
+if [ "$encoding" != "1" ]; then
+    echo "Invalid encoding: $encoding" >&2
+    exit 1
+fi
+
+pos=$header_size
+i=0
+while [ $i -lt $num_entries ]; do
+    read -r id offset next_id next_offset < <(parse_file "$INPUT" $pos "<HIHI")
+    let pos+=entry_size
+    let entry_len=next_offset-offset
+    echo "$id: $(dd if=$INPUT bs=1 count=$entry_len skip=$offset 2>/dev/null)"
+    let i+=1
+done