40247-vm/api/scan_structure.php

<?php
declare(strict_types=1);
header("Content-Type: application/json; charset=utf-8");

require_once __DIR__ . "/../includes/curated_data.php";

$source = "https://spireason.neocities.org/";
$ids = [
  ["id"=>"growing-out-of-the-shadows", "icon"=>"⚛️", "title"=>"Growing Out of the Shadows"],
  ["id"=>"sheep", "icon"=>"🐑", "title"=>"Sheep / Laegna Counters"],
  ["id"=>"infinity", "icon"=>"♾️", "title"=>"Infinity"],
  ["id"=>"natura", "icon"=>"🌀", "title"=>"Natura"],
  ["id"=>"sunrise", "icon"=>"🔆", "title"=>"Sunrise"],
  ["id"=>"bigbang", "icon"=>"💥", "title"=>"Bigbang"],
  ["id"=>"yggdrasill", "icon"=>"🌳", "title"=>"Yggdrasill"],
  ["id"=>"spiritrise", "icon"=>"🔷", "title"=>"Spiritrise"],
  ["id"=>"laelab", "icon"=>"🔬", "title"=>"LaeLab"],
  ["id"=>"geneticar", "icon"=>"🧬", "title"=>"Geneticar"],
  ["id"=>"handheldcal", "icon"=>"🖩", "title"=>"HandheldCal"],
  ["id"=>"puzzled", "icon"=>"🧩", "title"=>"Puzzled"],
  ["id"=>"chakra", "icon"=>"✴️", "title"=>"Chakra"],
  ["id"=>"wheelsgoround", "icon"=>"⚙️", "title"=>"Wheels Go Round"],
  ["id"=>"laemedics", "icon"=>"🧘", "title"=>"LaeMedics"],
  ["id"=>"coffeeandcigarettes", "icon"=>"☕", "title"=>"Coffee and Cigarettes"],
];

$html = fetch_source($source);
if ($html === null) {
  http_response_code(502);
  echo json_encode(["success"=>false, "error"=>"Could not fetch source page"], JSON_UNESCAPED_SLASHES | JSON_UNESCAPED_UNICODE);
  exit;
}

$positions = [];
foreach ($ids as $item) {
  $pattern = '~\bid\s*=\s*["\x27]' . preg_quote($item["id"], '~') . '["\x27]~i';
  if (preg_match($pattern, $html, $m, PREG_OFFSET_CAPTURE)) {
    $item["pos"] = $m[0][1];
    $positions[] = $item;
  }
}
usort($positions, fn($a, $b) => $a["pos"] <=> $b["pos"]);

$sections = [];
$totalLinks = 0;
$utilityChrome = [];
foreach ($positions as $idx => $item) {
  $start = (int)$item["pos"];
  $end = isset($positions[$idx + 1]) ? (int)$positions[$idx + 1]["pos"] : strlen($html);
  $chunk = substr($html, $start, $end - $start);
  $ownedLinks = [];
  if ($item["id"] === "coffeeandcigarettes") {
    // Coffee is terminal prose. Links after it are footer/fixed/absolute utility chrome,
    // not branch links owned by the Coffee icon.
    $utilityChrome = parse_visible_links($chunk, $source);
  } else {
    $ownedLinks = parse_visible_links($chunk, $source);
    $totalLinks += count($ownedLinks);
  }
  $layout = $sectionLayout[$item["id"]] ?? null;
  $sections[] = [
    "id" => $item["id"],
    "icon" => $item["icon"],
    "title" => $item["title"],
    "source_anchor" => "#" . $item["id"],
    "owns_until" => $item["id"] === "coffeeandcigarettes" ? "utility chrome" : (isset($positions[$idx + 1]) ? "#" . $positions[$idx + 1]["id"] : "end"),
    "layout" => $layout,
    "excerpt" => unicode_limit(clean_text($chunk), 420),
    "link_count" => count($ownedLinks),
    "links" => $ownedLinks,
  ];
}

$shadowBonus = [
  ["icon"=>"🎭", "title"=>"Dancing Shadows", "href"=>"https://material-psychic-gam-8mo3.bolt.host/"],
  ["icon"=>"📘", "title"=>"Book of Shadows 2", "href"=>"https://app-bxfrqbbqegap.appmedo.com/"],
  ["icon"=>"📕", "title"=>"Bulk Load Book of the Dead 3", "href"=>"https://app-by9gm7mu9ssh.appmedo.com/"],
];

echo json_encode([
  "success" => true,
  "source" => $source,
  "scanned_at" => gmdate("Y-m-d H:i:s") . " UTC",
  "model" => "CSS-aware: first 100vh console and fixed/absolute utility chrome are separated; document-flow icon owners keep links until the next meaningful icon/title; Coffee is terminal prose and owns no branch links.",
  "layout_zones" => $layoutZones ?? [],
  "section_count" => count($sections),
  "link_count" => $totalLinks,
  "utility_count" => count($utilityChrome),
  "utility_chrome" => $utilityChrome,
  "shadow_bonus" => $shadowBonus,
  "sections" => $sections,
], JSON_UNESCAPED_SLASHES | JSON_UNESCAPED_UNICODE);

function fetch_source(string $url): ?string
{
  $context = stream_context_create([
    "http" => [
      "timeout" => 18,
      "header" => "User-Agent: LandScaper structural scanner\r\n",
    ],
  ]);
  $html = @file_get_contents($url, false, $context);
  return is_string($html) && $html !== "" ? $html : null;
}

function parse_visible_links(string $chunk, string $base): array
{
  $links = [];
  if (!preg_match_all("~<a\b([^>]*)>(.*?)</a>~is", $chunk, $matches, PREG_SET_ORDER | PREG_OFFSET_CAPTURE)) {
    return $links;
  }
  foreach ($matches as $match) {
    $attrs = $match[1][0] ?? "";
    $inner = $match[2][0] ?? "";
    $offset = (int)($match[0][1] ?? 0);
    if (is_hidden_anchor($attrs)) {
      continue;
    }
    $href = attr_value($attrs, "href");
    $label = visible_label($attrs, $inner);
    if ($href === "" || $label === "") {
      continue;
    }
    $absolute = absolute_url($href, $base);
    $reason = skip_reason($href, $absolute);
    if ($reason !== "") {
      continue;
    }
    $host = strtolower((string)(parse_url($absolute, PHP_URL_HOST) ?: ""));
    $category = categorize_link($absolute, $label);
    $links[] = [
      "text" => unicode_limit($label, 150),
      "href" => $absolute,
      "host" => $host,
      "category" => $category,
      "category_label" => category_label($category),
      "context" => context_hint($chunk, $offset),
      "layout_hint" => layout_hint($attrs, $chunk, $offset),
      "safety" => safety_class($host),
    ];
  }
  return $links;
}


function context_hint(string $chunk, int $offset): string
{
  $prefix = substr($chunk, max(0, $offset - 1800), min(1800, $offset));
  if (!preg_match_all('~<(article|section|blockquote|div|p|big|center)\b([^>]*)>~i', $prefix, $tags, PREG_SET_ORDER)) {
    return 'flow text';
  }
  $tag = end($tags);
  $name = strtolower($tag[1] ?? 'flow');
  $attrs = $tag[2] ?? '';
  $id = attr_value($attrs, 'id');
  $class = attr_value($attrs, 'class');
  $label = $name;
  if ($id !== '') $label .= '#' . $id;
  if ($class !== '') $label .= '.' . preg_replace('/\s+/', '.', trim($class));
  return $label;
}

function layout_hint(string $attrs, string $chunk, int $offset): string
{
  $style = strtolower(attr_value($attrs, 'style'));
  $class = strtolower(attr_value($attrs, 'class'));
  $prefix = strtolower(substr($chunk, max(0, $offset - 1300), min(1300, $offset)));
  $haystack = $style . ' ' . $class . ' ' . $prefix;
  if (str_contains($haystack, 'position: fixed') || str_contains($haystack, 'position:fixed')) return 'fixed chrome';
  if (str_contains($haystack, 'position: absolute') || str_contains($haystack, 'position:absolute') || str_contains($haystack, 'calc(100vh')) return 'absolute-positioned chrome';
  if (str_contains($haystack, 'position: relative') || str_contains($haystack, 'position:relative')) return 'relative positioned block';
  if (str_contains($haystack, '<article')) return 'article branch link';
  if (str_contains($haystack, '<section')) return 'section branch link';
  if (str_contains($haystack, '<blockquote')) return 'blockquote branch link';
  return 'document-flow link';
}

function attr_value(string $attrs, string $name): string
{
  $pattern = '~\b' . preg_quote($name, '~') . '\s*=\s*(["\x27])(.*?)\1~is';
  if (preg_match($pattern, $attrs, $m)) {
    return html_entity_decode(trim($m[2]), ENT_QUOTES | ENT_HTML5, "UTF-8");
  }
  return "";
}

function is_hidden_anchor(string $attrs): bool
{
  if (preg_match("~(?:^|\s)hidden(?:\s|=|$)~i", $attrs)) return true;
  if (preg_match('~\baria-hidden\s*=\s*(["\x27])true\1~i', $attrs)) return true;
  $style = attr_value($attrs, "style");
  if ($style !== "") {
    $style = strtolower($style);
    return str_contains($style, "display:none") || str_contains($style, "display: none") || str_contains($style, "visibility:hidden") || str_contains($style, "visibility: hidden") || str_contains($style, "opacity:0") || str_contains($style, "opacity: 0");
  }
  return false;
}

function visible_label(string $attrs, string $inner): string
{
  $cleanInner = preg_replace("~<(script|style)\b[^>]*>.*?</\1>~is", " ", $inner) ?? $inner;
  $text = html_entity_decode(strip_tags($cleanInner), ENT_QUOTES | ENT_HTML5, "UTF-8");
  $text = trim(preg_replace("/\s+/u", " ", $text) ?? "");
  if ($text !== "") return $text;
  foreach (["aria-label", "title", "alt"] as $attr) {
    $candidate = attr_value($attrs, $attr);
    if ($candidate !== "") return trim(preg_replace("/\s+/u", " ", $candidate) ?? "");
  }
  return "";
}

function skip_reason(string $href, string $absolute): string
{
  $hrefLower = strtolower(trim($href));
  if ($hrefLower === "" || $hrefLower[0] === "#") return "fragment";
  foreach (["javascript:", "mailto:", "tel:", "data:", "blob:"] as $scheme) {
    if (str_starts_with($hrefLower, $scheme)) return "scheme";
  }
  if (!preg_match("~^https?://~i", $absolute)) return "scheme";
  $path = (string)(parse_url($absolute, PHP_URL_PATH) ?: "");
  if (preg_match("~\.(?:js|mjs|css|map|png|jpe?g|gif|svg|webp|ico|woff2?|ttf|eot)(?:$|[?#])~i", $path)) return "asset";
  return "";
}

function absolute_url(string $href, string $base): string
{
  $href = html_entity_decode(trim($href), ENT_QUOTES | ENT_HTML5, "UTF-8");
  if (preg_match("~^https?://~i", $href)) return $href;
  if (str_starts_with($href, "//")) return "https:" . $href;
  $parts = parse_url($base);
  $scheme = $parts["scheme"] ?? "https";
  $host = $parts["host"] ?? "spireason.neocities.org";
  if (str_starts_with($href, "/")) return $scheme . "://" . $host . $href;
  $basePath = $parts["path"] ?? "/";
  $dir = rtrim(dirname($basePath), "/");
  if ($dir === "") $dir = "/";
  return $scheme . "://" . $host . rtrim($dir, "/") . "/" . str_replace(" ", "%20", $href);
}

function clean_text(string $html): string
{
  $html = preg_replace("~<(script|style)\b[^>]*>.*?</\1>~is", " ", $html) ?? $html;
  $text = html_entity_decode(strip_tags($html), ENT_QUOTES | ENT_HTML5, "UTF-8");
  $text = preg_replace("~\bid=\"[^\"]+\"\s*>~u", " ", $text) ?? $text;
  $text = preg_replace("~[#.][A-Za-z0-9_-][^{}]{0,220}\\{[^{}]*\\}~u", " ", $text) ?? $text;
  $text = preg_replace("~\b(?:font|color|background|padding|margin|position|display|border|box-shadow|width|height|transform|opacity|z-index|transition)[^.;]{0,180}[.;]~iu", " ", $text) ?? $text;
  $text = preg_replace("~/\\*.*?\\*/~s", " ", $text) ?? $text;
  $text = preg_replace("~@media[^{}]*\\{\\s*\\}~u", " ", $text) ?? $text;
  return trim(preg_replace("/\s+/u", " ", $text) ?? "");
}

function categorize_link(string $url, string $label): string
{
  $host = strtolower((string)(parse_url($url, PHP_URL_HOST) ?: ""));
  $path = strtolower((string)(parse_url($url, PHP_URL_PATH) ?: ""));
  $haystack = strtolower($url . " " . $label);
  if (str_ends_with($path, ".pdf")) return "pdf";
  if (str_ends_with($path, ".zip")) return "archive";
  if (str_ends_with($path, ".json") || str_contains($path, "numberdatabase")) return "dataset";
  if (str_ends_with($path, ".py") || str_contains($haystack, "script") || str_contains($haystack, "github")) return "source";
  if (str_contains($haystack, "frequency") || str_contains($haystack, "frequential") || str_contains($haystack, "octave") || str_contains($haystack, "calculator") || str_contains($haystack, "counter")) return "tool";
  if (str_contains($haystack, "bot") || str_contains($host, "perplexity.ai")) return "bot";
  if (str_contains($host, "github")) return "github";
  if (str_contains($path, "apples") || str_contains($haystack, "app") || str_contains($host, "lovable.app") || str_contains($host, "bolt.host") || str_contains($host, "appmedo.com")) return "applet";
  if (str_contains($host, "prezi") || str_contains($host, "docs.google")) return "presentation";
  if (str_contains($host, "youtube") || str_contains($host, "notion")) return "media";
  if ($host !== "spireason.neocities.org") return "external";
  return "onsite";
}

function category_label(string $category): string
{
  return [
    "pdf" => "PDF text",
    "archive" => "Archive",
    "dataset" => "Dataset",
    "source" => "Source file",
    "tool" => "Tool",
    "bot" => "External bot/system",
    "github" => "Repository",
    "applet" => "Applet",
    "presentation" => "Presentation",
    "media" => "Media",
    "external" => "External system",
    "onsite" => "On-site branch",
  ][$category] ?? ucfirst(str_replace("-", " ", $category));
}

function safety_class(string $host): string
{
  if ($host === "spireason.neocities.org") return "source";
  $knownHosts = [
    "laegna.notaku.site",
    "prezi.com",
    "www.perplexity.ai",
    "huggingface.co",
    "assorted-canopy-961.notion.site",
    "www.youtube.com",
    "youtube.com",
    "github.com",
    "tambetvali.github.io",
    "archive.org",
  ];
  foreach ($knownHosts as $knownHost) {
    if ($host === $knownHost || str_ends_with($host, "." . $knownHost)) return "known-external";
  }
  return "review";
}

function unicode_limit(string $text, int $limit): string
{
  if (preg_match_all("/./us", $text, $chars) && count($chars[0]) > $limit) {
    return implode("", array_slice($chars[0], 0, max(0, $limit - 1))) . "…";
  }
  return $text;
}