From a54c90f7c014510b5a37b0fcf68a45ace055070f Mon Sep 17 00:00:00 2001
From: Jose Falanga <jose.falanga@gmail.com>
Date: Wed, 19 Aug 2026 21:19:33 -0300
Subject: [PATCH] refactor: simplify _fetch_game_page_metadata using shared
 helpers

- Replace manual HTTP polling with _download_url_bytes
- Extract _is_cloudflare_challenge and _extract_screenshots_from_html
- All functions now under 50 lines
---
 core/boxart_itch.gd | 61 ++++++++++-----------------------------------
 1 file changed, 13 insertions(+), 48 deletions(-)

diff --git a/core/boxart_itch.gd b/core/boxart_itch.gd
index 72d77ed..508d61c 100644
--- a/core/boxart_itch.gd
+++ b/core/boxart_itch.gd
@@ -138,7 +138,7 @@ func get_boxart(item: LibraryItem, kind: LAYOUT) -> Texture2D:
 	var meta: Dictionary = _enriched_meta.get(game_id, {})
 	if meta.is_empty() and game.has("url"):
 		logger.info("Enriching metadata for: %s (url=%s)" % [title, game.get("url", "")])
-		meta = _fetch_game_page_metadata(game)
+		meta = await _fetch_game_page_metadata(game)
 		# Cache both successes and failures to avoid repeated requests.
 		_enriched_meta[game_id] = meta
 		_save_enriched_meta(game_id, meta)
@@ -605,67 +605,32 @@ func _fetch_game_page_metadata(game: Dictionary) -> Dictionary:
 		return {}
 
 	logger.info("Scraping page for screenshots: " + page_url)
-	var http := HTTPClient.new()
-	var parts := page_url.split("/")
-	var use_tls := parts[0] == "https:"
-	var err := http.connect_to_host(parts[2], 443 if use_tls else 80, TLSOptions.client() if use_tls else null)
-	if err != OK:
-		logger.warn("HTTP connect failed for %s: %d" % [page_url, err])
+	var body := await _download_url_bytes(page_url)
+	if body.is_empty():
 		return {}
-	var deadline := Time.get_ticks_msec() + HTTP_TIMEOUT_MS
-	while http.get_status() == HTTPClient.STATUS_CONNECTING or http.get_status() == HTTPClient.STATUS_RESOLVING:
-		http.poll()
-		if Time.get_ticks_msec() > deadline:
-			http.close()
-			logger.warn("HTTP connect timed out for: " + page_url)
-			return {}
-		OS.delay_msec(CONNECT_POLL_DELAY_MS)
-	if http.get_status() != HTTPClient.STATUS_CONNECTED:
-		logger.warn("HTTP not connected for %s: status=%d" % [page_url, http.get_status()])
-		http.close()
-		return {}
-
-	http.request(HTTPClient.METHOD_GET, "/" + "/".join(parts.slice(3)), PackedStringArray())
-	deadline = Time.get_ticks_msec() + HTTP_TIMEOUT_MS
-	while http.get_status() == HTTPClient.STATUS_REQUESTING:
-		http.poll()
-		if Time.get_ticks_msec() > deadline:
-			http.close()
-			logger.warn("HTTP request timed out for: " + page_url)
-			return {}
-		OS.delay_msec(READ_POLL_DELAY_MS)
-
-	var body := PackedByteArray()
-	while http.get_status() == HTTPClient.STATUS_BODY:
-		http.poll()
-		var chunk: PackedByteArray = http.read_response_body_chunk()
-		if chunk.is_empty():
-			if Time.get_ticks_msec() > deadline:
-				http.close()
-				logger.warn("HTTP body read timed out for: " + page_url)
-				return {}
-			OS.delay_msec(READ_POLL_DELAY_MS)
-			continue
-		body.append_array(chunk)
-	http.close()
-
 	var html := body.get_string_from_utf8()
 	if html.is_empty():
 		logger.warn("Empty HTML response for: " + page_url)
 		return {}
-
-	# Detect Cloudflare challenge pages — no point extracting screenshots.
-	if html.find("challenge-platform") != -1 or html.find("cf-browser-verification") != -1:
+	if _is_cloudflare_challenge(html):
 		logger.warn("Cloudflare challenge detected for: %s — falling back to cover art" % page_url)
 		return {}
+	return _extract_screenshots_from_html(html, page_url)
 
+
+## Returns true if the HTML contains a Cloudflare challenge page.
+func _is_cloudflare_challenge(html: String) -> bool:
+	return html.find("challenge-platform") != -1 or html.find("cf-browser-verification") != -1
+
+
+## Extracts screenshot URLs from the itch.io page HTML using regex.
+func _extract_screenshots_from_html(html: String, page_url: String) -> Dictionary:
 	var shot_re := RegEx.new()
 	shot_re.compile(SCREENSHOT_REGEX_PATTERN)
 	var screenshots: Array = []
 	for m in shot_re.search_all(html):
 		var url: String = _sanitize_url(m.get_string(0))
 		screenshots.append(url)
-
 	logger.info("Found %d screenshots for page %s" % [screenshots.size(), page_url])
 	return {"screenshots": screenshots}