#!/bin/bash

# Fetch and merge text and JSON files from URLs
# Usage: ./create.sh [output_file]

OUTPUT_FILE="${1:-merged_output.txt}"
TEMP_DIR=$(mktemp -d)
FINAL_TEMP="$TEMP_DIR/combined.txt"

# Color codes for output
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
BLUE='\033[0;34m'
NC='\033[0m' # No Color

# =============================================================================
# CONFIGURE YOUR URLS AND COMMENTS HERE
# =============================================================================

URLS_AND_COMMENTS=(
    "https://raw.githubusercontent.com/dsojevic/profanity-list/refs/heads/main/en.txt"
    "https://raw.githubusercontent.com/dsojevic/profanity-list/refs/heads/main/emoji.txt"
    "https://raw.githubusercontent.com/LDNOOBW/List-of-Dirty-Naughty-Obscene-and-Otherwise-Bad-Words/refs/heads/master/de"
    "https://raw.githubusercontent.com/LDNOOBW/List-of-Dirty-Naughty-Obscene-and-Otherwise-Bad-Words/refs/heads/master/en"
    "https://raw.githubusercontent.com/zacanger/profane-words/refs/heads/master/words.json"
)

# =============================================================================

# Check if jq is available for JSON processing
if ! command -v jq &> /dev/null; then
    echo -e "${YELLOW}Warning: jq not found. JSON files will be skipped.${NC}"
    echo -e "${YELLOW}Install jq with: apt-get install jq (Ubuntu/Debian) or brew install jq (macOS)${NC}"
    HAS_JQ=false
else
    HAS_JQ=true
fi

echo -e "${GREEN}Text/JSON File Merger${NC}"
echo "Processing $(echo "${URLS_AND_COMMENTS[@]}" | grep -c 'https://' || true) URLs..."

# Process URLs and comments
for line in "${URLS_AND_COMMENTS[@]}"; do
    # Handle comments
    if [[ "$line" =~ ^[[:space:]]*# ]]; then
        echo "$line" >> "$FINAL_TEMP"
        continue
    fi

    # Skip lines that don't look like URLs
    if [[ ! "$line" =~ ^https?:// ]]; then
        continue
    fi

    echo -e "${YELLOW}Fetching: $line${NC}"

    # Create temporary file for this URL
    URL_TEMP="$TEMP_DIR/url_content.tmp"

    # Fetch the URL
    if curl -s -f "$line" -o "$URL_TEMP" 2>/dev/null; then
        echo -e "${GREEN}✓ Successfully fetched${NC}"

        # Check if the content is JSON
        if [ "$HAS_JQ" = true ] && jq empty "$URL_TEMP" 2>/dev/null; then
            echo -e "${BLUE}  → Detected JSON format, extracting strings...${NC}"

            # Try to extract strings from JSON array or object
            # Handle different JSON structures
            if jq -e 'type == "array"' "$URL_TEMP" >/dev/null 2>&1; then
                # JSON array - extract all string values
                jq -r '.[] | select(type == "string")' "$URL_TEMP" >> "$FINAL_TEMP" 2>/dev/null || {
                    echo -e "${RED}  ✗ Failed to parse JSON array${NC}"
                    echo "# ERROR: Could not parse JSON from $line" >> "$FINAL_TEMP"
                }
            elif jq -e 'type == "object"' "$URL_TEMP" >/dev/null 2>&1; then
                # JSON object - extract all string values
                jq -r 'recurse | select(type == "string")' "$URL_TEMP" >> "$FINAL_TEMP" 2>/dev/null || {
                    echo -e "${RED}  ✗ Failed to parse JSON object${NC}"
                    echo "# ERROR: Could not parse JSON from $line" >> "$FINAL_TEMP"
                }
            else
                echo -e "${RED}  ✗ Unsupported JSON structure${NC}"
                echo "# ERROR: Unsupported JSON structure from $line" >> "$FINAL_TEMP"
            fi
        else
            # Regular text file - append as-is
            echo -e "${BLUE}  → Processing as text file...${NC}"
            cat "$URL_TEMP" >> "$FINAL_TEMP"
        fi

        # Clean up URL temp file
        rm -f "$URL_TEMP"
    else
        echo -e "${RED}✗ Failed to fetch: $line${NC}"
        echo "# ERROR: Could not fetch $line" >> "$FINAL_TEMP"
    fi
done

# Check if we have any content
if [ ! -s "$FINAL_TEMP" ]; then
    echo -e "${RED}No content to process${NC}"
    rm -rf "$TEMP_DIR"
    exit 1
fi

echo -e "${YELLOW}Processing content...${NC}"

# Remove duplicates, sort alphabetically, and save to output file
# Keep comments at the top, sort the rest
{
    grep '^#' "$FINAL_TEMP" 2>/dev/null || true
    grep -v '^#' "$FINAL_TEMP" 2>/dev/null | grep -v '^[[:space:]]*$' | sort -u
} > "$OUTPUT_FILE"

# Cleanup
rm -rf "$TEMP_DIR"

# Show results
line_count=$(wc -l < "$OUTPUT_FILE")
echo -e "${GREEN}✓ Complete! Merged content saved to: $OUTPUT_FILE${NC}"
echo -e "${GREEN}Total lines: $line_count${NC}"

# Show a preview of the content
if [ -s "$OUTPUT_FILE" ]; then
    echo -e "${YELLOW}Preview (first 10 non-comment lines):${NC}"
    grep -v '^#' "$OUTPUT_FILE" | head -10
fi