import os
import re
import gzip

SRC_DIR = "src"

def minify_html(html: str) -> str:
    # Tags where whitespace should be preserved
    preserve_tags = ['pre', 'code', 'textarea', 'script', 'style']
    preserve_regex = '|'.join(preserve_tags)

    # Protect preserve blocks with placeholders
    preserve_blocks = []
    def preserve(match):
        preserve_blocks.append(match.group(0))
        return f"__PRESERVE_BLOCK_{len(preserve_blocks)-1}__"

    html = re.sub(rf'<({preserve_regex})[\s\S]*?</\1>', preserve, html, flags=re.IGNORECASE)

    # Remove HTML comments
    html = re.sub(r'<!--.*?-->', '', html, flags=re.DOTALL)

    # Collapse all whitespace between tags
    html = re.sub(r'>\s+<', '><', html)

    # Collapse multiple spaces inside tags
    html = re.sub(r'\s+', ' ', html)

    # Restore preserved blocks
    for i, block in enumerate(preserve_blocks):
        html = html.replace(f"__PRESERVE_BLOCK_{i}__", block)

    return html.strip()

for root, _, files in os.walk(SRC_DIR):
    for file in files:
        if file.endswith(".html"):
            html_path = os.path.join(root, file)
            with open(html_path, "r", encoding="utf-8") as f:
                html_content = f.read()

            # minified = regex.sub("\g<1>", html_content)
            minified = minify_html(html_content)

            # Compress with gzip (compresslevel 9 is maximum compression)
            # IMPORTANT: we don't use brotli because Firefox doesn't support brotli with insecured context (only supported on HTTPS)
            compressed = gzip.compress(minified.encode('utf-8'), compresslevel=9)

            base_name = f"{os.path.splitext(file)[0]}Html"
            header_path = os.path.join(root, f"{base_name}.generated.h")

            with open(header_path, "w", encoding="utf-8") as h:
                h.write(f"// THIS FILE IS AUTOGENERATED, DO NOT EDIT MANUALLY\n\n")
                h.write(f"#pragma once\n")
                h.write(f"#include <cstddef>\n\n")

                # Write the compressed data as a byte array
                h.write(f"constexpr char {base_name}[] PROGMEM = {{\n")

                # Write bytes in rows of 16
                for i in range(0, len(compressed), 16):
                    chunk = compressed[i:i+16]
                    hex_values = ', '.join(f'0x{b:02x}' for b in chunk)
                    h.write(f"  {hex_values},\n")

                h.write(f"}};\n\n")
                h.write(f"constexpr size_t {base_name}CompressedSize = {len(compressed)};\n")
                h.write(f"constexpr size_t {base_name}OriginalSize = {len(minified)};\n")

            print(f"Generated: {header_path}")
            print(f"  Original: {len(html_content)} bytes")
            print(f"  Minified: {len(minified)} bytes ({100*len(minified)/len(html_content):.1f}%)")
            print(f"  Compressed: {len(compressed)} bytes ({100*len(compressed)/len(html_content):.1f}%)")