#!/bin/bash

# Function to show help
show_help() {
    echo "Universal File Converter"
    echo "Supported conversions: DOCX→MD, MD→HTML, HTML→MD, MD→DOCX, HTML→DOCX"
    echo "Usage: $0 [-f] [-o outputdir] [-t format] [-T template] [--no-toc] input1.ext [input2.ext ...]"
    echo "  -f: Force overwrite existing output files"
    echo "  -o: Output directory (default: same as input)"
    echo "  -t: Target format (md, html, docx) - overrides auto-detection"
    echo "  -T: Template file path (default: templates/<template>.html, where <template>"
    echo "      comes from the doc's YAML front matter; falls back to templates/report.html)"
    echo "  --no-toc: Skip table of contents generation"
}

# Function to source ZDDC config files if they exist
source_config_file() {
    local config_file="$1"
    if [ -f "$config_file" ]; then
        echo "  → Loading ZDDC configuration from: $config_file"
        set -a  # automatically export all variables
        . "$config_file"
        set +a  # turn off automatic export
        return 0
    fi
    return 1
}

# Load ZDDC configuration file
load_zddc_config() {
    local search_dir="$1"
    
    # Search for zddc.conf then .zddc.conf in the search directory
    if source_config_file "$search_dir/zddc.conf"; then
        return 0
    elif source_config_file "$search_dir/.zddc.conf"; then
        return 0
    fi
    
    # No config file found - continue with defaults
    return 1
}

# Source global ZDDC config from current working directory
# This is called once at startup - do NOT call again inside convert functions
load_zddc_config "$(pwd)"

# Parse arguments
FORCE_OVERWRITE=false
OUTPUT_DIR=""
TARGET_FORMAT=""
CUSTOM_TEMPLATE=""
NO_TOC=false

while [ $# -gt 0 ]; do
    case $1 in
        -h|--help)
            show_help
            exit 0
            ;;
        -f)
            FORCE_OVERWRITE=true
            echo "Force overwrite mode: ON"
            shift
            ;;
        -o)
            OUTPUT_DIR="$2"
            echo "Output directory: $OUTPUT_DIR"
            shift 2
            ;;
        -t)
            TARGET_FORMAT="$2"
            echo "Target format: $TARGET_FORMAT"
            shift 2
            ;;
        -T)
            CUSTOM_TEMPLATE="$2"
            echo "Custom template: $CUSTOM_TEMPLATE"
            shift 2
            ;;
        --no-toc)
            NO_TOC=true
            echo "Table of contents: DISABLED"
            shift
            ;;
        -*)
            echo "Unknown option: $1"
            show_help
            exit 1
            ;;
        *)
            break
            ;;
    esac
done

if [ "$FORCE_OVERWRITE" = "false" ]; then
    echo "Force overwrite mode: OFF (will skip existing output files)"
fi

if [ -z "$OUTPUT_DIR" ]; then
    echo "Output directory: same as input files"
fi

if [ $# -eq 0 ]; then
    echo "Error: No input files specified"
    show_help
    exit 1
fi

# Validate target format if specified
if [ -n "$TARGET_FORMAT" ]; then
    TARGET_FORMAT_LOWER=$(echo "$TARGET_FORMAT" | tr '[:upper:]' '[:lower:]')
    if [ "$TARGET_FORMAT_LOWER" != "md" ] && [ "$TARGET_FORMAT_LOWER" != "html" ] && [ "$TARGET_FORMAT_LOWER" != "docx" ]; then
        echo "Error: Invalid target format '$TARGET_FORMAT'. Supported: md, html, docx"
        exit 1
    fi
    echo "Target format override: $TARGET_FORMAT_LOWER"
fi

echo "Processing $# files..."

TOTAL_FILES=$#
SUCCESSFUL=0
FAILED=0
SKIPPED=0

# Parse a ZDDC filename stem (no extension) into ZDDC_TRACKING / ZDDC_REVISION /
# ZDDC_STATUS / ZDDC_TITLE. Returns 0 on a full match, 1 otherwise.
# Each field is extracted with its own sed backref rather than a delimiter-joined
# string + cut, so a title containing the join character (e.g. '|') can't corrupt
# the split.
parse_zddc_filename() {
    local stem="$1"
    local sub='s/^\([^_]*\)_\([^ ]*\) *(\([^)]*\)) *- *\(.*\)$'
    # Gate on a full match before extracting (empty fields are otherwise ambiguous).
    printf '%s\n' "$stem" | grep -Eq '^[^_]+_[^ ]+ *\([^)]*\) *- *.+$' || return 1
    ZDDC_TRACKING=$(printf '%s\n' "$stem" | sed -n "${sub}/\\1/p")
    ZDDC_REVISION=$(printf '%s\n' "$stem" | sed -n "${sub}/\\2/p")
    ZDDC_STATUS=$(printf '%s\n'  "$stem" | sed -n "${sub}/\\3/p")
    ZDDC_TITLE=$(printf '%s\n'   "$stem" | sed -n "${sub}/\\4/p")
    return 0
}

# Function to convert DOCX to Markdown
convert_docx_to_md() {
    local INPUT="$1"
    local OUTPUT_FILE="$2"
    local TEMP_FILE="$3"
    local MEDIA_DIR="$4"
    local BASENAME="$5"
    local FILENAME_NO_EXT="$6"
    
    # Convert using pandoc with proper extension stripping to temp file first
    if pandoc -f docx -t gfm --markdown-headings=atx --extract-media="$MEDIA_DIR" --wrap=none "$INPUT" -o "$TEMP_FILE"; then
        
        # Parse ZDDC filename pattern: trackingNumber_revision (status) - title.extension
        if parse_zddc_filename "$FILENAME_NO_EXT"; then
            TRACKING_NUMBER="$ZDDC_TRACKING"
            REVISION="$ZDDC_REVISION"
            STATUS="$ZDDC_STATUS"
            TITLE="$ZDDC_TITLE"

            echo "  → ZDDC metadata detected:"
            echo "    • Tracking: $TRACKING_NUMBER"
            echo "    • Revision: $REVISION"
            echo "    • Status: $STATUS"
            echo "    • Title: $TITLE"
            
            # Create YAML front matter and combine with content
            {
                echo "---"
                echo "client: \"${client:-}\""
                echo "project: \"${project:-}\""
                echo "tracking_number: \"$TRACKING_NUMBER\""
                echo "revision: \"$REVISION\""
                echo "status: \"$STATUS\""
                echo "title: \"$TITLE\""
                echo "source_file: \"$BASENAME\""
                echo "created: \"$(date -u +%Y-%m-%d)\""
                echo "---"
                echo ""
                cat "$TEMP_FILE"
            } > "$OUTPUT_FILE"
            
            rm "$TEMP_FILE"
        else
            # No ZDDC pattern detected, just move temp file to final location
            mv "$TEMP_FILE" "$OUTPUT_FILE"
        fi
        
        echo "  ✓ Successfully converted: $BASENAME (DOCX→MD)"
        return 0
    else
        echo "  ✗ Failed to convert: $BASENAME (DOCX→MD)"
        # Clean up temp file on failure
        [ -f "$TEMP_FILE" ] && rm "$TEMP_FILE"
        return 1
    fi
}

# Function to convert HTML to Markdown
convert_html_to_md() {
    local INPUT="$1"
    local OUTPUT_FILE="$2"
    local BASENAME="$3"
    
    if pandoc "$INPUT" -f html -t gfm --markdown-headings=atx --wrap=none -o "$OUTPUT_FILE"; then
        echo "  ✓ Successfully converted: $BASENAME (HTML→MD)"
        return 0
    else
        echo "  ✗ Failed to convert: $BASENAME (HTML→MD)"
        return 1
    fi
}

# Function to convert Markdown to DOCX
convert_md_to_docx() {
    local INPUT="$1"
    local OUTPUT_FILE="$2"
    local BASENAME="$3"
    
    if pandoc "$INPUT" -f gfm -t docx -o "$OUTPUT_FILE"; then
        echo "  ✓ Successfully converted: $BASENAME (MD→DOCX)"
        return 0
    else
        echo "  ✗ Failed to convert: $BASENAME (MD→DOCX)"
        return 1
    fi
}

# Function to convert HTML to DOCX
convert_html_to_docx() {
    local INPUT="$1"
    local OUTPUT_FILE="$2"
    local BASENAME="$3"
    
    if pandoc "$INPUT" -f html -t docx -o "$OUTPUT_FILE"; then
        echo "  ✓ Successfully converted: $BASENAME (HTML→DOCX)"
        return 0
    else
        echo "  ✗ Failed to convert: $BASENAME (HTML→DOCX)"
        return 1
    fi
}

# Function to convert Markdown to HTML
convert_md_to_html() {
    local INPUT="$1"
    local OUTPUT_FILE="$2"
    local BASENAME="$3"
    local INPUT_DIR="$4"
    
    # No need to reload config - already loaded at startup
    # Config variables from zddc.conf are already in environment
    
    # Get absolute paths - use pwd-based approach for POSIX compatibility
    case "$INPUT" in
        /*) INPUT_ABS="$INPUT" ;;
        *) INPUT_ABS="$(pwd)/$INPUT" ;;
    esac
    case "$OUTPUT_FILE" in
        /*) OUTPUT_ABS="$OUTPUT_FILE" ;;
        *) OUTPUT_ABS="$(pwd)/$OUTPUT_FILE" ;;
    esac
    # Determine template to use
    if [ -n "$CUSTOM_TEMPLATE" ]; then
        # Use custom template if specified
        if [ -f "$CUSTOM_TEMPLATE" ]; then
            TEMPLATE_ABS="$CUSTOM_TEMPLATE"
            echo "  → Using custom template: $TEMPLATE_ABS"
        else
            echo "  ⚠ Warning: Custom template not found: $CUSTOM_TEMPLATE, using default discovery"
            CUSTOM_TEMPLATE=""
        fi
    fi
    
    # Default template discovery if no custom template or custom template not found.
    # Named templates live in a templates/ dir (report.html, letter.html,
    # specification.html, sharing _head/_doc/_scripts partials). The document
    # selects one via a `template:` field in its YAML front matter; default report.
    if [ -z "$CUSTOM_TEMPLATE" ]; then
        # Convert script directory to absolute path
        SCRIPT_DIR=$(cd "$(dirname "$0")" && pwd)

        # Check if script is a symlink and resolve target directory
        SCRIPT_TARGET_DIR=""
        if [ -L "$0" ]; then
            # readlink -f is available on Linux with GNU coreutils
            SCRIPT_TARGET=$(readlink -f "$0")
            SCRIPT_TARGET_DIR=$(dirname "$SCRIPT_TARGET")
        fi

        # Template name from the doc's front matter (sanitized to a bare basename).
        TEMPLATE_NAME=$(sed -n '/^---[[:space:]]*$/,/^---[[:space:]]*$/ s/^template:[[:space:]]*"\{0,1\}\([A-Za-z0-9_-]\{1,\}\)"\{0,1\}[[:space:]]*$/\1/p' "$INPUT_ABS" | head -1)
        [ -n "$TEMPLATE_NAME" ] || TEMPLATE_NAME="report"

        # Search order: input dir, script dir, symlink target dir — each a templates/
        # subdir. Use absolute paths since pandoc runs after a cd into the input dir.
        INPUT_DIR_ABS=$(dirname "$INPUT_ABS")
        TEMPLATE_ABS=""
        for _tdir in "$INPUT_DIR_ABS/templates" "$SCRIPT_DIR/templates" "$SCRIPT_TARGET_DIR/templates"; do
            [ -n "$_tdir" ] || continue
            if [ -f "$_tdir/$TEMPLATE_NAME.html" ]; then
                TEMPLATE_ABS="$_tdir/$TEMPLATE_NAME.html"
                echo "  → Using template: $TEMPLATE_ABS"
                break
            elif [ -f "$_tdir/report.html" ]; then
                TEMPLATE_ABS="$_tdir/report.html"
                echo "  ⚠ Template '$TEMPLATE_NAME' not found; using $TEMPLATE_ABS"
                break
            fi
        done
        if [ -z "$TEMPLATE_ABS" ]; then
            echo "  ⚠ Warning: templates/ not found, using pandoc default template"
        fi
    fi
    
    # Change to input directory so pandoc can find relative resources
    ORIGINAL_DIR=$(pwd)
    cd "$INPUT_DIR"
    
    # Build pandoc command as an argument array (safe form, no eval — each value
    # is a separate array element so it can't be re-split or injected by the shell).
    PANDOC_ARGS=()
    PANDOC_ARGS+=("--from" "markdown+yaml_metadata_block")
    PANDOC_ARGS+=("--standalone")
    PANDOC_ARGS+=("--embed-resources")
    PANDOC_ARGS+=("--section-divs")
    
    # Add TOC options if not disabled
    if [ "$NO_TOC" = "false" ]; then
        PANDOC_ARGS+=("--toc" "--toc-depth=6")
    fi
    
    if [ -n "$TEMPLATE_ABS" ]; then
        PANDOC_ARGS+=("--template" "$TEMPLATE_ABS")
    fi
    
    # Generate timestamp for conversion (force English locale)
    GENERATION_TIME=$(LC_TIME=C date '+%B %d, %Y at %I:%M:%S %p %Z')
    
    # Extract ZDDC metadata from filename for template variables
    FILENAME_NO_EXT=$(basename "$INPUT" .md)
    if parse_zddc_filename "$FILENAME_NO_EXT"; then
        TRACKING_NUMBER="$ZDDC_TRACKING"
        REVISION="$ZDDC_REVISION"
        STATUS="$ZDDC_STATUS"
        TITLE="$ZDDC_TITLE"

        # Pass ZDDC variables to template (each as separate args to avoid injection)
        PANDOC_ARGS+=("--variable" "tracking_number=$TRACKING_NUMBER")
        PANDOC_ARGS+=("--variable" "revision=$REVISION")
        PANDOC_ARGS+=("--variable" "status=$STATUS")
        PANDOC_ARGS+=("--variable" "generation_time=$GENERATION_TIME")
        PANDOC_ARGS+=("--variable" "title=$TITLE")
        case "$REVISION" in
            *~*)
                PANDOC_ARGS+=("--variable" "is_draft=true")
                ;;
        esac
    else
        # Still pass generation time even if no ZDDC match
        PANDOC_ARGS+=("--variable" "generation_time=$GENERATION_TIME")
    fi

    # Add ZDDC configuration variables from zddc.conf
    if [ -n "$client" ]; then
        PANDOC_ARGS+=("--variable" "client=$client")
    fi
    if [ -n "$project" ]; then
        PANDOC_ARGS+=("--variable" "project=$project")
    fi
    if [ -n "$contractor" ]; then
        PANDOC_ARGS+=("--variable" "contractor=$contractor")
    fi
    if [ -n "$project_number" ]; then
        PANDOC_ARGS+=("--variable" "project_number=$project_number")
    fi
    
    # Pass TOC status to template
    if [ "$NO_TOC" = "true" ]; then
        PANDOC_ARGS+=("--variable" "no-toc=true")
    fi
    
    # (--section-divs already added above)
    PANDOC_ARGS+=("--html-q-tags")

    # Run pandoc with positional arguments (safe form, no eval)
    # All variables passed as separate arguments to avoid shell injection
    if pandoc "$(basename "$INPUT_ABS")" -o "$OUTPUT_ABS" "${PANDOC_ARGS[@]}"; then
        
        echo "  ✓ Successfully converted: $BASENAME (MD→HTML)"
        cd "$ORIGINAL_DIR"
        return 0
    else
        echo "  ✗ Failed to convert: $BASENAME (MD→HTML)"
        cd "$ORIGINAL_DIR"
        return 1
    fi
}

for INPUT in "$@"; do
    echo ""
    echo "Processing: $INPUT"
    
    # Validate input file exists
    if [ ! -f "$INPUT" ]; then
        echo "  ✗ Input file not found: $INPUT"
        FAILED=$((FAILED + 1))
        continue
    fi
    
    # Extract filename without path and extension 
    BASENAME=$(basename "$INPUT")
    FILENAME_NO_EXT="${BASENAME%.*}"
    EXTENSION="${BASENAME##*.}"
    INPUT_DIR=$(dirname "$INPUT")
    
    # Convert extension to lowercase for comparison
    EXTENSION_LOWER=$(echo "$EXTENSION" | tr '[:upper:]' '[:lower:]')
    
    # Determine conversion type based on target format override or auto-detection
    if [ -n "$TARGET_FORMAT" ]; then
        # Target format specified - determine conversion type
        TARGET_EXT="$TARGET_FORMAT_LOWER"
        case "$EXTENSION_LOWER" in
            docx)
                if [ "$TARGET_EXT" = "md" ]; then
                    CONVERSION_TYPE="docx2md"
                elif [ "$TARGET_EXT" = "html" ]; then
                    echo "  ✗ Direct DOCX→HTML conversion not supported. Convert to MD first."
                    FAILED=$((FAILED + 1))
                    continue
                elif [ "$TARGET_EXT" = "docx" ]; then
                    echo "  ⚠ Skipping: $BASENAME (already DOCX format)"
                    SKIPPED=$((SKIPPED + 1))
                    continue
                fi
                ;;
            md)
                if [ "$TARGET_EXT" = "html" ]; then
                    CONVERSION_TYPE="md2html"
                elif [ "$TARGET_EXT" = "docx" ]; then
                    CONVERSION_TYPE="md2docx"
                elif [ "$TARGET_EXT" = "md" ]; then
                    echo "  ⚠ Skipping: $BASENAME (already MD format)"
                    SKIPPED=$((SKIPPED + 1))
                    continue
                fi
                ;;
            html|htm)
                if [ "$TARGET_EXT" = "md" ]; then
                    CONVERSION_TYPE="html2md"
                elif [ "$TARGET_EXT" = "docx" ]; then
                    CONVERSION_TYPE="html2docx"
                elif [ "$TARGET_EXT" = "html" ]; then
                    echo "  ⚠ Skipping: $BASENAME (already HTML format)"
                    SKIPPED=$((SKIPPED + 1))
                    continue
                fi
                ;;
            *)
                echo "  ✗ Unsupported input file type: .$EXTENSION (supported: .docx, .md, .html, .htm)"
                FAILED=$((FAILED + 1))
                continue
                ;;
        esac
    else
        # Auto-detect conversion type based on input extension
        case "$EXTENSION_LOWER" in
            docx)
                CONVERSION_TYPE="docx2md"
                TARGET_EXT="md"
                ;;
            md)
                CONVERSION_TYPE="md2html"
                TARGET_EXT="html"
                ;;
            html|htm)
                CONVERSION_TYPE="html2md"
                TARGET_EXT="md"
                ;;
            *)
                echo "  ✗ Unsupported file type: .$EXTENSION (supported: .docx, .md, .html, .htm)"
                FAILED=$((FAILED + 1))
                continue
                ;;
        esac
    fi
    
    # Determine output location
    if [ -n "$OUTPUT_DIR" ]; then
        OUTPUT_FILE="$OUTPUT_DIR/$FILENAME_NO_EXT.$TARGET_EXT"
        if [ "$CONVERSION_TYPE" = "docx2md" ]; then
            TEMP_FILE="$OUTPUT_DIR/temp_$FILENAME_NO_EXT.md"
            MEDIA_DIR="$OUTPUT_DIR/$FILENAME_NO_EXT"
        fi
    else
        OUTPUT_FILE="$INPUT_DIR/$FILENAME_NO_EXT.$TARGET_EXT"
        if [ "$CONVERSION_TYPE" = "docx2md" ]; then
            TEMP_FILE="$INPUT_DIR/temp_$FILENAME_NO_EXT.md"
            MEDIA_DIR="$INPUT_DIR/$FILENAME_NO_EXT"
        fi
    fi
    
    echo "  → Output file: $OUTPUT_FILE"
    if [ "$CONVERSION_TYPE" = "docx2md" ]; then
        echo "  → Media dir: $MEDIA_DIR/"
    fi
    
    # Create output directory if needed
    OUTPUT_FILE_DIR=$(dirname "$OUTPUT_FILE")
    if [ ! -d "$OUTPUT_FILE_DIR" ]; then
        mkdir -p "$OUTPUT_FILE_DIR"
    fi
    
    # Check if output file exists and handle accordingly
    if [ -f "$OUTPUT_FILE" ] && [ "$FORCE_OVERWRITE" = "false" ]; then
        echo "  ⚠ Skipped (file exists): $BASENAME"
        SKIPPED=$((SKIPPED + 1))
        continue
    fi
    
    # Perform conversion based on type
    case "$CONVERSION_TYPE" in
        docx2md)
            if convert_docx_to_md "$INPUT" "$OUTPUT_FILE" "$TEMP_FILE" "$MEDIA_DIR" "$BASENAME" "$FILENAME_NO_EXT"; then
                SUCCESSFUL=$((SUCCESSFUL + 1))
            else
                FAILED=$((FAILED + 1))
            fi
            ;;
        md2html)
            if convert_md_to_html "$INPUT" "$OUTPUT_FILE" "$BASENAME" "$INPUT_DIR"; then
                SUCCESSFUL=$((SUCCESSFUL + 1))
            else
                FAILED=$((FAILED + 1))
            fi
            ;;
        html2md)
            if convert_html_to_md "$INPUT" "$OUTPUT_FILE" "$BASENAME"; then
                SUCCESSFUL=$((SUCCESSFUL + 1))
            else
                FAILED=$((FAILED + 1))
            fi
            ;;
        md2docx)
            if convert_md_to_docx "$INPUT" "$OUTPUT_FILE" "$BASENAME"; then
                SUCCESSFUL=$((SUCCESSFUL + 1))
            else
                FAILED=$((FAILED + 1))
            fi
            ;;
        html2docx)
            if convert_html_to_docx "$INPUT" "$OUTPUT_FILE" "$BASENAME"; then
                SUCCESSFUL=$((SUCCESSFUL + 1))
            else
                FAILED=$((FAILED + 1))
            fi
            ;;
        *)
            echo "  ✗ Unknown conversion type: $CONVERSION_TYPE"
            FAILED=$((FAILED + 1))
            ;;
    esac
done

echo ""
echo "=========================================="
echo "CONVERSION SUMMARY"
echo "=========================================="
echo "Total files processed: $TOTAL_FILES"
echo "Successful conversions: $SUCCESSFUL"
echo "Failed conversions: $FAILED"
echo "Skipped (existing files): $SKIPPED"
echo "=========================================="