Add interactive language selection and command-line options for OCR4Linux

moheladwy · moheladwy · commit f24a245cc099 · 2025-09-06T01:25:44.000+03:00
- Implemented interactive language selection using rofi in the shell script.
- Added support for specifying languages via command-line arguments.
- Updated README.md to reflect new features and usage instructions.
diff --git a/OCR4Linux.py b/OCR4Linux.py
@@ -46,7 +46,7 @@ class TesseractConfig:
 
         langs (str): The languages to be used by Tesseract for OCR.
         custom_config (str): Custom configuration string for Tesseract.
-        ouput_encoding (str): The encoding to be used for the output file.
+        output_encoding (str): The encoding to be used for the output file.
 
     Methods:
         __init__(self, image_path: str, output_path: str):
@@ -60,28 +60,37 @@ class TesseractConfig:
             text extraction, and saves the extracted text to an output file. Returns 0 if successful, 1 otherwise.
     """
 
-    def __init__(self, image_path: str, output_path: str):
+    def __init__(self, image_path: str, output_path: str, langs: str = None):
         """
         Initializes the OCR4Linux class with command-line arguments.
 
         Attributes:
             image_path (str): The path to the input image file.
             output_path (str): The path to the output file where results will be saved.
+            langs (str): The languages to be used by Tesseract for OCR (optional).
             oem_mode (int): The OCR Engine Mode (OEM) for Tesseract.
             psm_mode (int): The Page Segmentation Mode (PSM) for Tesseract.
-            langs (str): The languages to be used by Tesseract for OCR.
             custom_config (str): Custom configuration string for Tesseract.
-            ouput_encoding (str): The encoding to be used for the output file.
+            output_encoding (str): The encoding to be used for the output file.
         """
         self.image_path = image_path
         self.output_path = output_path
         self.oem_mode = 3  # Default LSTM engine
         self.psm_mode = 6  # Uniform block of text
         self.available_langs = pytesseract.get_languages()
-        self.langs = '+'.join(filter(None, self.available_langs)
-                              ) if self.available_langs else 'eng'
+
+        # Use provided languages or default to all available languages
+        if langs and langs.strip():
+            self.langs = langs
+            print(f"Using specified languages: {langs}", file=sys.stderr)
+        else:
+            self.langs = '+'.join(filter(None, self.available_langs)
+                                  ) if self.available_langs else 'eng'
+            print(
+                f"Using all available languages: {self.langs}", file=sys.stderr)
+
         self.custom_config = f'--oem {self.oem_mode} --psm {self.psm_mode}'
-        self.ouput_encoding = 'utf-8'
+        self.output_encoding = 'utf-8'
 
     def extract_text_with_lines(self, image: Image) -> str:
         """
@@ -115,7 +124,7 @@ def main(self) -> int:
                 extracted_text = self.extract_text_with_lines(image)
 
                 # Save the extracted text to a file
-                with open(self.output_path, 'w', encoding=self.ouput_encoding) as file:
+                with open(self.output_path, 'w', encoding=self.output_encoding) as file:
                     file.write(extracted_text)
 
                 return 0
@@ -149,19 +158,21 @@ def __init__(self):
             "    and text extraction using Tesseract OCR. The script takes an input\n" + \
             "    based on the language in the image."
         self.useges = [
-            "python OCR4Linux.py <image_path> <output_path>",
+            "python OCR4Linux.py <image_path> <output_path> [--langs <languages>]",
             "python OCR4Linux.py [-l | --list-langs]",
             "python OCR4Linux.py [-h | --help]"
         ]
         self.examples = [
             "python OCR4Linux.py screenshot.png output.txt",
+            "python OCR4Linux.py screenshot.png output.txt --langs eng+fra+deu",
             "python OCR4Linux.py -l",
             "python OCR4Linux.py -h"
         ]
         self.arguments = [
             "file_path:         Path to the python script",
             "image_path:        Path to the image file",
             "output_path:       Path to the output text file",
+            "--langs:           Specify languages for OCR (e.g., eng+fra+deu)",
             "-l, --list-langs:  List all available languages for OCR in the system",
             "-h, --help:        Display this help message, then exit"
         ]
@@ -199,20 +210,24 @@ def check_arguments(self) -> int:
         Checks the command line arguments for validity.
 
         Handles the following options:
-        - Standard usage: <image_path> <output_path>
+        - Standard usage: <image_path> <output_path> [--langs <languages>]
         - Help: -h or --help
         - List languages: -l or --list-langs
 
         Returns:
-            bool: True if arguments are valid, False otherwise.
+            int: 0 if help/list was shown, 1 if error, 2 if valid arguments for processing.
         """
         if len(sys.argv) == 2 and sys.argv[1] in ['-l', '--list-langs']:
             self.list_available_languages()
             return 0
         elif len(sys.argv) == 2 and sys.argv[1] in ['-h', '--help']:
             self.help()
             return 0
-        elif len(sys.argv) != self.args_num:
+        elif len(sys.argv) < self.args_num or len(sys.argv) > 5:
+            # Valid patterns:
+            # 3 args: script image_path output_path
+            # 4 args: script image_path output_path --langs=languages
+            # 5 args: script image_path output_path --langs languages
             self.help()
             return 1
         return 2
@@ -252,7 +267,8 @@ def main(self):
         This function performs the following steps:
         1. Checks if the correct number of arguments is provided.
         2. Verifies if the image file exists.
-        3. Creates an instance of the TesseractConfig class and runs the OCR process.
+        3. Parses language arguments if provided.
+        4. Creates an instance of the TesseractConfig class and runs the OCR process.
 
         Returns:
             int: Returns 1 if there is an error with the arguments or image path, otherwise returns the result of the TesseractConfig main function.
@@ -268,8 +284,15 @@ def main(self):
         if not self.check_image_path(sys.argv[1]):
             return 1
 
+        # Parse language arguments
+        langs = None
+        if len(sys.argv) >= 4 and sys.argv[3] == '--langs' and len(sys.argv) == 5:
+            langs = sys.argv[4]
+        elif len(sys.argv) == 4 and sys.argv[3].startswith('--langs='):
+            langs = sys.argv[3].split('=', 1)[1]
+
         # Create an instance of the TesseractConfig class
-        tesseract = TesseractConfig(sys.argv[1], sys.argv[2])
+        tesseract = TesseractConfig(sys.argv[1], sys.argv[2], langs)
         return tesseract.main()
 
 
diff --git a/OCR4Linux.sh b/OCR4Linux.sh
@@ -23,6 +23,7 @@
 #     - tesseract-ocr: For text extraction
 #     - grimblast/scrot: For screenshot capture
 #     - wl-clipboard/xclip: For clipboard operations
+#     - rofi: For language selection menu
 #     - Python 3.x: For image processing
 #
 # Usage:
@@ -39,37 +40,61 @@ LOGS_FILE_NAME="OCR4Linux.log"
 SLEEP_DURATION=0.5
 REMOVE_SCREENSHOT=false
 KEEP_LOGS=false
+LANG_SPECIFIED=false
+SPECIFIED_LANGS=""
+
+langs=()
 
 # Display help message
 show_help() {
     echo "Usage: $(basename "$0") [OPTIONS]"
     echo "Options:"
-    echo "  -r            Remove screenshot in the screenshot directory"
-    echo "  -d DIRECTORY  Set screenshot directory (default: $SCREENSHOT_DIRECTORY)"
-    echo "  -l            Keep logs"
-    echo "  -h            Show this help message, then exit"
+    echo "  -r                Remove screenshot in the screenshot directory"
+    echo "  -d DIRECTORY      Set screenshot directory (default: $SCREENSHOT_DIRECTORY)"
+    echo "  -l                Keep logs"
+    echo "  --lang LANGUAGES  Specify OCR languages (e.g., 'all', 'eng', 'eng+ara')"
+    echo "  -h                Show this help message, then exit"
     echo "Example:"
-    echo "  OCR4Linux.sh -s -d $HOME/screenshots -l"
-    echo "  OCR4Linux.sh -s -l"
+    echo "  OCR4Linux.sh -d $HOME/screenshots -l"
+    echo "  OCR4Linux.sh --lang eng+ara"
+    echo "  OCR4Linux.sh --lang all -l"
     echo "  OCR4Linux.sh -h"
     echo "Note:"
-    echo "  if you run \`OCR4Linux.sh\` only without any arguments, it will save the screenshot in the default directory $SCREENSHOT_DIRECTORY."
+    echo "  - If --lang is not specified, an interactive language selection menu will appear"
+    echo "  - Use 'all' to select all available languages"
+    echo "  - Use '+' to separate multiple languages (e.g., 'eng+ara+fra')"
+    echo "  - Without arguments, screenshots are saved to $SCREENSHOT_DIRECTORY"
 }
 
 # Parse command line arguments
-while getopts "rd:lh" opt; do
-    case $opt in
-    r) REMOVE_SCREENSHOT=true ;;
-    d) SCREENSHOT_DIRECTORY="$OPTARG" ;;
-    l) KEEP_LOGS=true ;;
-    h)
-        show_help
-        exit 0
-        ;;
-    *)
-        show_help
-        exit 1
-        ;;
+while [[ $# -gt 0 ]]; do
+    case $1 in
+        -r)
+            REMOVE_SCREENSHOT=true
+            shift
+            ;;
+        -d)
+            SCREENSHOT_DIRECTORY="$2"
+            shift 2
+            ;;
+        -l)
+            KEEP_LOGS=true
+            shift
+            ;;
+        --lang)
+            SPECIFIED_LANGS="$2"
+            LANG_SPECIFIED=true
+            shift 2
+            ;;
+        -h)
+            show_help
+            exit 0
+            ;;
+        *)
+            echo "Unknown option: $1"
+            show_help
+            exit 1
+            ;;
     esac
 done
 
@@ -89,6 +114,12 @@ log_message() {
 check_if_files_exist() {
     log_message "Checking required files and directories..."
 
+    # Check if rofi is installed
+    if ! command -v rofi &> /dev/null; then
+        log_message "ERROR: rofi is not installed. Please install rofi to use language selection."
+        exit 1
+    fi
+
     # Validate screenshot directory
     if [ ! -d "$SCREENSHOT_DIRECTORY" ]; then
         log_message "Creating screenshot directory: $SCREENSHOT_DIRECTORY since it does not exist."
@@ -112,6 +143,50 @@ check_if_files_exist() {
     fi
 }
 
+# Process specified languages from command line
+process_specified_langs() {
+    log_message "Processing specified languages: $SPECIFIED_LANGS"
+    
+    # Handle "all" case
+    if [[ "$SPECIFIED_LANGS" == "all" ]]; then
+        mapfile -t langs < <(tesseract --list-langs | awk 'FNR>1')
+        log_message "Using ALL available languages: $(IFS=+ ; echo "${langs[*]}")"
+    else
+        # Split the language string by '+' and populate the langs array
+        IFS='+' read -ra langs <<< "$SPECIFIED_LANGS"
+        log_message "Using specified languages: $(IFS=+ ; echo "${langs[*]}")"
+        
+        # Validate that the specified languages are available
+        available_langs=$(tesseract --list-langs | awk 'FNR>1')
+        for lang in "${langs[@]}"; do
+            if ! echo "$available_langs" | grep -q "^$lang$"; then
+                log_message "WARNING: Language '$lang' is not available on this system"
+            fi
+        done
+    fi
+}
+
+# Choose languages for OCR using rofi
+choose_lang() {
+    log_message "Fetching available languages for OCR selection..."
+    
+    # Get available languages and add "ALL" option at the beginning
+    mapfile -t langs < <(tesseract --list-langs | awk 'BEGIN {print "ALL" } FNR>1' | rofi -dmenu -multi-select -p "Select OCR Languages:")
+
+    if [ ${#langs[@]} -eq 0 ]; then
+        log_message "CANCELLED: User aborted language selection"
+        exit 1
+    fi
+
+    # If "ALL" is selected, use all available languages
+    if [[ " ${langs[*]} " =~ " ALL " ]]; then
+        mapfile -t langs < <(tesseract --list-langs | awk 'FNR>1')
+        log_message "Selected ALL languages: $(IFS=+ ; echo "${langs[*]}")"
+    else
+        log_message "Selected languages: $(IFS=+ ; echo "${langs[*]}")"
+    fi
+}
+
 # take shots using grimblast for wayland
 takescreenshot_wayland() {
     sleep $SLEEP_DURATION
@@ -136,9 +211,22 @@ takescreenshot() {
 
 # Pass the screenshot to OCR tool to extract text from the image.
 extract_text() {
-    python "$OCR4Linux_HOME/$OCR4Linux_PYTHON_NAME" \
-        "$SCREENSHOT_DIRECTORY/$SCREENSHOT_NAME" \
-        "$OCR4Linux_HOME/$TEXT_OUTPUT_FILE_NAME"
+    # Create language string for passing to Python script
+    local lang_string=""
+    if [ ${#langs[@]} -gt 0 ]; then
+        lang_string=$(IFS=+; echo "${langs[*]}")
+    fi
+    
+    if [ -n "$lang_string" ]; then
+        python "$OCR4Linux_HOME/$OCR4Linux_PYTHON_NAME" \
+            "$SCREENSHOT_DIRECTORY/$SCREENSHOT_NAME" \
+            "$OCR4Linux_HOME/$TEXT_OUTPUT_FILE_NAME" \
+            --langs "$lang_string"
+    else
+        python "$OCR4Linux_HOME/$OCR4Linux_PYTHON_NAME" \
+            "$SCREENSHOT_DIRECTORY/$SCREENSHOT_NAME" \
+            "$OCR4Linux_HOME/$TEXT_OUTPUT_FILE_NAME"
+    fi
     log_message "Text extraction completed successfully"
 }
 
@@ -175,6 +263,14 @@ remove_image() {
 # Run the functions
 main() {
     check_if_files_exist
+    
+    # Handle language selection
+    if [ "$LANG_SPECIFIED" = true ]; then
+        process_specified_langs
+    else
+        choose_lang
+    fi
+    
     takescreenshot
     extract_text
     run_copy_to_clipboard
diff --git a/README.md b/README.md
