From 63ea99d38ade254bd428b5d031be0feec123315f Mon Sep 17 00:00:00 2001
From: Nutomic <me@nutomic.com>
Date: Tue, 19 Nov 2024 15:48:44 +0100
Subject: [PATCH] Guess image mime type from file extension (fixes #5196)
 (#5212)

* Guess image mime type from file extension (fixes #5196)

* Mime check fixes. (#5213)

* Mime check fixes.

* Adding back comment.

---------

Co-authored-by: Dessalines <dessalines@users.noreply.github.com>
---
 Cargo.lock                       | 19 ++++++++++++++++++-
 crates/api_common/Cargo.toml     |  1 +
 crates/api_common/src/request.rs | 20 ++++++++++++++------
 3 files changed, 33 insertions(+), 7 deletions(-)

diff --git a/Cargo.lock b/Cargo.lock
index 54857b3a0..b62cf624c 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -1,6 +1,6 @@
 # This file is automatically @generated by Cargo.
 # It is not intended for manual editing.
-version = 3
+version = 4
 
 [[package]]
 name = "accept-language"
@@ -2515,6 +2515,7 @@ dependencies = [
  "lemmy_db_views_moderator",
  "lemmy_utils",
  "mime",
+ "mime_guess",
  "moka",
  "pretty_assertions",
  "regex",
@@ -3147,6 +3148,16 @@ version = "0.3.17"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "6877bb514081ee2a7ff5ef9de3281f14a4dd4bceac4c09388074a6b5df8a139a"
 
+[[package]]
+name = "mime_guess"
+version = "2.0.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f7c44f8e672c00fe5308fa235f821cb4198414e1c77935c1ab6948d3fd78550e"
+dependencies = [
+ "mime",
+ "unicase",
+]
+
 [[package]]
 name = "minimal-lexical"
 version = "0.2.1"
@@ -5300,6 +5311,12 @@ version = "1.17.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "42ff0bf0c66b8238c6f3b578df37d0b7848e55df8577b3f74f92a69acceeb825"
 
+[[package]]
+name = "unicase"
+version = "2.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7e51b68083f157f853b6379db119d1c1be0e6e4dec98101079dec41f6f5cf6df"
+
 [[package]]
 name = "unicode-bidi"
 version = "0.3.15"
diff --git a/crates/api_common/Cargo.toml b/crates/api_common/Cargo.toml
index 7b987cbb7..76f531ddb 100644
--- a/crates/api_common/Cargo.toml
+++ b/crates/api_common/Cargo.toml
@@ -64,6 +64,7 @@ actix-web = { workspace = true, optional = true }
 enum-map = { workspace = true }
 urlencoding = { workspace = true }
 mime = { version = "0.3.17", optional = true }
+mime_guess = "2.0.5"
 webpage = { version = "2.0", default-features = false, features = [
   "serde",
 ], optional = true }
diff --git a/crates/api_common/src/request.rs b/crates/api_common/src/request.rs
index 36010f760..cc506b896 100644
--- a/crates/api_common/src/request.rs
+++ b/crates/api_common/src/request.rs
@@ -23,7 +23,6 @@ use lemmy_utils::{
   REQWEST_TIMEOUT,
   VERSION,
 };
-use mime::Mime;
 use reqwest::{
   header::{CONTENT_TYPE, RANGE},
   Client,
@@ -64,11 +63,20 @@ pub async fn fetch_link_metadata(url: &Url, context: &LemmyContext) -> LemmyResu
     .await?
     .error_for_status()?;
 
-  let content_type: Option<Mime> = response
-    .headers()
-    .get(CONTENT_TYPE)
-    .and_then(|h| h.to_str().ok())
-    .and_then(|h| h.parse().ok());
+  // In some cases servers send a wrong mime type for images, which prevents thumbnail
+  // generation. To avoid this we also try to guess the mime type from file extension.
+  let content_type = mime_guess::from_path(url.path())
+    .first()
+    // If you can guess that its an image type, then return that first.
+    .filter(|guess| guess.type_() == mime::IMAGE)
+    // Otherwise, get the content type from the headers
+    .or(
+      response
+        .headers()
+        .get(CONTENT_TYPE)
+        .and_then(|h| h.to_str().ok())
+        .and_then(|h| h.parse().ok()),
+    );
 
   let opengraph_data = {
     // if the content type is not text/html, we don't need to parse it