Repository navigation
fix: decoded host size for IPv6 zone IDs - #1005
Merged
alandefreitas merged 1 commit intoOct 13, 2026
Merged
Conversation
|
An automated preview of the documentation is available at https://1005.url.prtest2.cppalliance.org/index.html If more commits are pushed to the pull request, the docs will rebuild at the same URL. 2026-10-13 02:35:11 UTC |
|
GCOVR code coverage report https://1005.url.prtest2.cppalliance.org/gcovr/index.html Build time: 2026-10-13 02:48:04 UTC |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## develop #1005 +/- ##
========================================
Coverage 99.12% 99.12%
========================================
Files 154 154
Lines 10093 10103 +10
========================================
+ Hits 10005 10015 +10
Misses 88 88
... and 1 file with indirect coverage changes Continue to review full report in Codecov by Harness.
🚀 New features to boost your workflow:
|
The host rule stored the encoded size of an IPv6 literal as its
decoded size. A zone ID shrinks when decoded ("%25" becomes "%",
and the zone ID itself may contain pct-encoded octets), so the
stored size overshot and every accessor that decodes the host
produced a string with trailing garbage:
url_view u("http://[fe80::1%25eth0]/");
u.host_address(); // "fe80::1%eth0\0\0", size 14 instead of 12
The ip_literal_rule now forwards the zone ID parsed by
ipv6_addrz_rule, and the host rule derives the decoded size from it.
The assertion in encoded_host_address() compared the decoded size
with the encoded size, which only holds without a zone ID, and had
been relaxed in url_view_base to tolerate the wrong value. It is
removed from both url_view_base and authority_view: the invariant
is covered by tests, like every other encoded_* accessor. Checking
it exposed three more bookkeeping bugs in the same family, fixed
here as well:
- remove_authority() left the decoded sizes of the removed user,
password and host behind, so encoded_authority().decoded_size()
was non-zero on a URL without an authority.
- encoded_authority() never counted the "@" separator nor the port,
so url_view("http://host:8080/").encoded_authority().decode()
returned "host".
- format() stored the bracketed host size without the closing "]",
and counted the ":" delimiter in the password and port sizes, so
format("http://[fe80::1]/").host() returned "[fe80::1".
The existing issue boostorg#919 regression test is kept as is. New tests
cover the parser, authority_view, the url setters, remove_authority
and format with zone IDs, pct-encoded zone IDs, userinfo and ports.
Thanks to Sergei Zimmerman for identifying the issue in boostorg#1004.
Closes boostorg#1004
alandefreitas
force-pushed
the
fix/ipv6-zone-id-decoded-size
branch
from
October 13, 2026 02:32
dbac83d to
841f0a5
Compare
This file contains hidden or 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
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.
The host rule stored the encoded size of an IPv6 literal as its decoded size. A zone ID shrinks when decoded ("%25" becomes "%", and the zone ID itself may contain pct-encoded octets), so the stored size overshot and every accessor that decodes the host produced a string with trailing garbage:
The ip_literal_rule now forwards the zone ID parsed by ipv6_addrz_rule, and the host rule derives the decoded size from it.
The assertion in encoded_host_address() compared the decoded size with the encoded size, which only holds without a zone ID, and had been relaxed in url_view_base to tolerate the wrong value. It is removed from both url_view_base and authority_view: the invariant is covered by tests, like every other encoded_* accessor. Checking it exposed three more bookkeeping bugs in the same family, fixed here as well:
The existing issue #919 regression test is kept as is. New tests cover the parser, authority_view, the url setters, remove_authority and format with zone IDs, pct-encoded zone IDs, userinfo and ports.
Thanks to Sergei Zimmerman for identifying the issue in #1004.
Closes #1004