From f369db5ae9a29a23299dff5f14768167230b0b79 Mon Sep 17 00:00:00 2001
From: Gary Pendergast <gary@pento.net>
Date: Thu, 9 Jan 2025 09:08:27 +1100
Subject: [PATCH] FIX: Ensure Onebox requests ask for the correct language.
 (#30637)

Onebox embeds currently default to accepting any language response from the destination, which can have some surprising behaviour. For example the `curl` equivalent of what Onebox does:

```
% curl -si -H "Accept-Language: *" 'https://developer.android.com/studio' | grep location:
location: /studio?hl=hi
```

This PR uses the value of `SiteSetting.default_locale` to populate the `Accept-Language` header, falling back to English if that isn't available, then finally accepting whatever language the destination makes available.
---
 lib/onebox/helpers.rb     | 18 +++++++++++++-----
 lib/oneboxer.rb           |  2 +-
 spec/lib/oneboxer_spec.rb |  7 +++++++
 3 files changed, 21 insertions(+), 6 deletions(-)

diff --git a/lib/onebox/helpers.rb b/lib/onebox/helpers.rb
index f80cca5fba3..eea070c2591 100644
--- a/lib/onebox/helpers.rb
+++ b/lib/onebox/helpers.rb
@@ -41,11 +41,10 @@ module Onebox
         canonical_uri = Addressable::URI.parse(canonical_link)
         if canonical_link && canonical_uri &&
              "#{canonical_uri.host}#{canonical_uri.path}" != "#{uri.host}#{uri.path}"
-          uri =
-            FinalDestination.new(
-              canonical_link,
-              Oneboxer.get_final_destination_options(canonical_link),
-            ).resolve
+          canonical_options = Oneboxer.get_final_destination_options(canonical_link)
+          canonical_options["extra_headers"] = { "Accept-Language" => accept_language }
+
+          uri = FinalDestination.new(canonical_link, canonical_options).resolve
           if uri.present?
             response =
               (
@@ -105,6 +104,7 @@ module Onebox
         headers ||= {}
 
         headers["User-Agent"] ||= user_agent if user_agent
+        headers["Accept-Language"] ||= accept_language if accept_language
 
         request = Net::HTTP::Get.new(uri.request_uri, headers)
         start_time = Time.now
@@ -236,6 +236,14 @@ module Onebox
       user_agent
     end
 
+    def self.accept_language
+      if SiteSetting.default_locale == "en"
+        "en;q=0.9, *;q=0.5"
+      else
+        "#{SiteSetting.default_locale.gsub(/_/, "-")};q=0.9, en;q=0.8, *;q=0.5"
+      end
+    end
+
     # Percent-encodes a URI string per RFC3986 - https://tools.ietf.org/html/rfc3986
     def self.uri_encode(url)
       return "" unless url
diff --git a/lib/oneboxer.rb b/lib/oneboxer.rb
index 422cc9603e5..a124e69023c 100644
--- a/lib/oneboxer.rb
+++ b/lib/oneboxer.rb
@@ -152,7 +152,7 @@ module Oneboxer
   end
 
   def self.redis_cached_response_body_key(uri)
-    "CACHED_RESPONSE_#{uri}"
+    "CACHED_RESPONSE_#{SiteSetting.default_locale}_#{uri}"
   end
 
   # Parse URLs out of HTML, returning the document when finished.
diff --git a/spec/lib/oneboxer_spec.rb b/spec/lib/oneboxer_spec.rb
index a2cde501cda..96f38fdc5f5 100644
--- a/spec/lib/oneboxer_spec.rb
+++ b/spec/lib/oneboxer_spec.rb
@@ -955,6 +955,13 @@ RSpec.describe Oneboxer do
       preview = Oneboxer.preview(url2, invalidate_oneboxes: true)
       expect(Oneboxer.cached_response_body_exists?(url2)).to eq(false)
     end
+
+    it "separates cache by default_locale" do
+      preview = Oneboxer.preview(url, invalidate_oneboxes: true)
+      expect(Oneboxer.cached_response_body_exists?(url)).to eq(true)
+      SiteSetting.default_locale = "fr"
+      expect(Oneboxer.cached_response_body_exists?(url)).to eq(false)
+    end
   end
 
   describe "register_local_handler" do