Handle raw UTF-8 bytes in redirect headers #317
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Try to parse the `Location` header as UTF-8 bytes as a fallback if the header value is not valid US-ASCII. This is technically against the URI spec which requires all literal characters in the URI to be US-ASCII (see [RFC 3986, Section 4.1](https://tools.ietf.org/html/rfc3986#section-4.1)). This is also more or less against the HTTP spec, which historically allowed for ISO-8859-1 text in header values but since was restricted to US-ASCII plus opaque bytes. Never has UTF-8 been encouraged or allowed as-such. See [RFC 7230, Section 3.2.4](https://tools.ietf.org/html/rfc7230#section-3.2.4) for more info. However, some bad or misconfigured web servers will do this anyway, and most web browsers recover from this by allowing and interpreting UTF-8 characters as themselves even though they _should_ have been percent-encoded. The third-party URI parsers that we use have no such leniency, so we percent-encode such bytes (if legal UTF-8) ahead of time before handing them off to the URI parser. This is in the spirit of being generous with what we accept (within reason) while being strict in what we produce. Since real websites exhibit this out-of-spec behavior it is worth handling it. Note that the underlying `tiny_http` library that our HTTP test mocking is based on does not allow UTF-8 header values right now, so we can't really test this efficiently. We already have a couple tests out there doing some raw TCP munging for one reason or another, so in the future we need to make sure to rewrite `testserver` to allow such headers and then enable the test. For now I've manually verified that this works. Fixes #315.
Codecov Report
@@ Coverage Diff @@
## master #317 +/- ##
==========================================
- Coverage 75.25% 74.94% -0.32%
==========================================
Files 57 57
Lines 2983 2993 +10
==========================================
- Hits 2245 2243 -2
- Misses 738 750 +12
Continue to review full report at Codecov.
|
|
@teto-bot rustfmt |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Try to parse the
Locationheader as UTF-8 bytes as a fallback if the header value is not valid US-ASCII. This is technically against the URI spec which requires all literal characters in the URI to be US-ASCII (see RFC 3986, Section 4.1).This is also more or less against the HTTP spec, which historically allowed for ISO-8859-1 text in header values but since was restricted to US-ASCII plus opaque bytes. Never has UTF-8 been encouraged or allowed as-such. See RFC 7230, Section 3.2.4 for more info.
However, some bad or misconfigured web servers will do this anyway, and most web browsers recover from this by allowing and interpreting UTF-8 characters as themselves even though they should have been percent-encoded. The third-party URI parsers that we use have no such leniency, so we percent-encode such bytes (if legal UTF-8) ahead of time before handing them off to the URI parser.
This is in the spirit of being generous with what we accept (within reason) while being strict in what we produce. Since real websites exhibit this out-of-spec behavior it is worth handling it.
Note that the underlying
tiny_httplibrary that our HTTP test mocking is based on does not allow UTF-8 header values right now, so we can't really test this efficiently. We already have a couple tests out there doing some raw TCP munging for one reason or another, so in the future we need to make sure to rewritetestserverto allow such headers and then enable the test. For now I've manually verified that this works.Fixes #315.