Update CHANGELOG.md, Hotfix rolling back checking magic number for syms

Add word fixed


Hotfix rolling back checking magic number for symbols

Needs more testing, as it broke symbol uploads.

CHANGELOG.md

@@ -6,6 +6,22 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.2.0.4] - 2021-09-25
+### Added
+- Added `.docker` folder, which defines a docker based development environment.
+- Redis+RQ task queue to decode minidumps
+- Added prism.js for code formatting (though feature is WIP)
+
+### Changed
+- Updated Crash Report page to look cleaner and show dump metadata or decode progress.
+- Updated Crash Report page to have a slim view for mobile pages.
+
+### Removed
+- Huey as a task queue
+
+### Fixed
+- Fixed bug preventing from upload webpage from uploading minidumps
+
 ## [0.2.0.3] - 2021-09-19
 ### Fixed
 - Flush symbol after stored in database to get "symbol_location" from project

crashserver/webapp/api.py

@@ -44,14 +44,8 @@ def upload_minidump(project):
 def upload_symbol(project, version):
     symbol_file = request.files.get("symbol_file")
 
-    # Ensure the uploaded file is plain text
-    symbol_file_bytes = symbol_file.stream.read()
-    magic_number = magic.from_buffer(symbol_file_bytes, mime=True)
-    if magic_number != "text/plain":
-        logger.error("Symbol rejected from {}. File detected as {}", request.remote_addr, magic_number)
-        return make_response({"error": "Bad symbol file."}, 400)
-
     # Use charset_normalizer to get a readable version of the text.
+    symbol_file_bytes = symbol_file.stream.read()
     char_res = char_norm.from_bytes(symbol_file_bytes)
     decoded = char_res.best().output()
     first_line_str = decoded[: decoded.find("\n".encode())].decode("utf-8")