When a user supplies an IP address to use for the HAPROXY protocol,
the IP version reported must be deduced from the address and has
no relation to the IP version used for the upstream connection.
Add test3220 to verify.
Fixes#21340
Reported-by: Fiona Klute
Closes#21341
Support HTTPS records in dnsd that have ALPN preferences.
Add pytest test_22_* to verify that HTTPS record ALPN
preferencces take effect if the HTTPS resolve arrives in time.
Fix HTTPS eyeballing use of timeouts for second attempt. Also,
make an initial HTTP/1.1 attempt switch off HTTP/2.
Closes#21329
Test has been flaky due to not waiting for the test server to be
listening. On slow CI systems, the started thread is late and the curl
command gets a refused connection.
Closes#21337
Add new internal `curl_url_same_origin()` to check if a href has the
same origin as a base URL. Add test cases in test1675 and use this in
http2 push handling.
Closes#21328
The function libssh2_userauth_keyboard_interactive_ex() calls the
callback and is documented to call free() on the memory returned to
libssh2 from the callback. libcurl can therefore not use the regular
curlx_strdup() for this, as that is not compatible in debug builds or
when curl_global_init_mem() is used.
Fixes#21336Closes#21338
- make sure all UNITTEST prototypes mark in which unit test they are used,
with "@unittest" markup
- make sure all UNITTEST functions do not use Curl_ prefix, as that is a
prefix we use for global private functions and these functions are static
and therefore not global and the prefix is wrong
- drop UNITTEST for functions not used in unit tests
- make the extract-unit-protos script highlight the above issues if found
- extract-unit-protos now also outputs the unit test number for all the
generated protos in lib/unitprotos.h to aid readers. It also adds the source
file and line number where the proto originates from.
- extract-unit-protos now exits with a non-zero value if any of the above
warnings are triggered
- cf-dns: Curl_cf_dns_result => static cf_dns_result
- hostip: Curl_ipv6works => static ipv6works
- url: remove Curl_setup_conn() - not used anymore
- connect: Curl_timeleft_now_ms => UNITTEST timeleft_now_ms
Closes#21330
This fixes the torture failures in 1675
Also, make it allocate the scheme *after* the path has been verified, so
that it is avoided in the common error cases.
Fixes#21326Closes#21324
To make it scoped for the single request appropriately.
Reported-by: Muhamad Arga Reksapati
Verify with libtest 2504: a custom Host *disabled* on reused handle
Closes#21312
Add "Delay-A: ms", "Delay-AAAA: ms" and "Delay-HTTPS: ms" to the test
dnsd config and implement delayed response handling.
Add test_21_09 and test_21_10 to check that delayed responses connect
using the undelayed address family.
Closes#21299
Use a dns connection filter to resolve hostnames where their addresses
are locally resolved and forwarded to the SOCKS proxy.
This makes all improvements, like in #21295 for example, also apply
to socks connections. Curl_resolv() is now only called from a DNS
filter.
(ftp still calls Curl_resolv_blocking() in two places, one of which
may be replaceable with a DNS filter as well to remove the block, tbd)
Closes#21297
There is no reason we should treat this part different on Windows. Noe
anything except blank, localhost or 127.0.0.1 cause error there as well.
Also: fix query handling in urlencode_str
Closes#21296
- enable three checks:
- bugprone-invalid-enum-default-initialization
- bugprone-sizeof-expression
- readability-inconsistent-declaration-parameter-name (strict)
- fix remaining discrepancies with arg names in prototypes
and implementation, in strict mode.
- document reason for some checks tested but not enabled.
Closes#20794
- examples: sync debug output printf masks.
- INSTALL-CMAKE.md: tidy up section for some options.
- curl_sha512_256: delete comment suggesting an optimization.
- vtls/keylog: scope a variable.
- vtls/openssl: make a source code URL a permalink.
- vtls/schannel: drop redundant parentheses.
- test1119.pl: robustify `$1` -> `$s`.
- sync arg names in comments to match the code.
- tidy up and minor fixes to comments.
- fix formatting/indenting/comment/newline/include nits.
- move `UNITTEST` protos next to definitions, sync their argument names.
- make variables static.
- add parentheses to Perl `open()` calls.
- drop unnecessary double quotes in Perl.
- clang-format.
Closes#21000
When a user supplied an ip address in a URL as hostname, use that even
when address family restrictions like -4 or -6 are set.
Add test_10_15/16 to verify with a local proxy server.
Fixes#21146
Reported-by: Terrance Wong
How:
- cf-dns: on see the hostname is an ip(v6) address, add the respective
A/AAAA to the dns query bits
- cf-dns/hostip: only hand out addrinfos for a family if that family
is part of the DNS queries. That prevents for example ipv6 addresses
to show up from dns cache entries
- change cf-ip-happy to no longer check for "ip_version" and instead
use all addresses that cf-dns hands out
Closes#21295
Fix http auth to set the `picked` auth when sending the corresponding
headers. Fix reporting Digest as avail.
Fixes#21274
Reported-by: sergio-nsk on github
Closes#21284
This should return a SSL_CTX pointer but it was accidentally broken.
Verify with test 1587
Follow-up to 2db8ae480f
Spotted by Codex Security
Closes#21290
... unless it is a POST and the user explicitly asked to keep doing
POST.
Add test1983/1984: verify --follow with 303 and PUT + custom GET
Fixes#20715
Reported-by: Dan Arnfield
Closes#21280
Connection filters can now carry the flag CF_TYPE_SETUP, indicating that
they are only needed during connection setup, e.g. connect.
Once the connection is fully established, those filter are removed
again. This frees resources and also makes the filter (call) chains
shorter.
Closes#21269
When HTTPS-RR is needed for the HTTP/3 handshake, delay the connect
until it arrives. Relevant only for TLS backends that support ECH, for
now.
Closes#21253
Make cf-https-connect work async correctly:
- only start first baller when at least one A/AAAA address
is available
- select first connect attempt after that with HTTPS-RR info
there or not.
- select second connect attempt only when HTTPS-RR is resolved
(may have resolved to "not known") and select possible ALPN
from things known by then. May not select any second attempt
when first already covers everything.
This means when the HTTPS-RR is known at/before the first address
is resolved, everything behaves as before. When the HTTPS-RR is
late, a first connection attempt will have been started. Any
ALPN preference from the HTTPS-RR that is not already ongoing will
then start the second attempt.
For HTTPS-RRs that recommend 2 or more ALPNs, the first will always
be attempted: either it is already ongong or it will be the ALPN
for the second attempt. The 2nd ALPN recommendation from HTTPS-RR
*may* be honored or not, depending on what is already selected.
The difference in behaviour between early/late HTTPS-RR resolve
cannot be helped - unless we do not perform any attempts before
it arrives. Trade offs.
Closes#21267
The making of the TLS session cache key should use the cert blob
independently of verifypeer on/off.
Follow-up to fa0ccd9f1f
Spotted by Codex Security
Closes#21222
There are 69 entries, bsearch is faster than linear search for random
access.
This now also makes the matching case sensitive (zone names always in
uppercase). No docs said otherwise and all tests assumed uppercase.
Extended test 517
Closes#21266
It did not require a full-length match, so empty or prefix tokens map to
ECJPAKE would silently add that cipher to the configured list.
Follow-up to fba9afebba
Reported by Codex Security
Closes#21264
Polish the time2epoch function to become a little more readable.
Corrected the military time zones: they were going in the wrong
direction.
Add more curl_getdate() input varations to test 517
Closes#21251
Introduce a limit on the concurrent connect attempts of 6:
- document this in CURLOPT_HAPPY_EYEBALLS_TIMEOUT_MS
- close the oldest attempt before opening a new one that would
exceed the limit
- closing failed attempts early to avoid sockets use beyong their
usefulness
- add tests for limits in unit2600
These changes are externally visible as file descriptors will be
reassigned where we previously kept the old one around and started a
new socket, allocating always a new descriptor.
Closes#21252
Follow-up to bd3b2a626a
- rename it to myssh_statemachine
- remove the use of ternary operators in the switch
- fix the error handling for if 'sshp' actually ever is NULL
Closes#21250
The pieces of the state machine that were not separate functions before, now
are.
Also removed the use of ternary operators in the function calls.
Follow-up to f0bf43e209Closes#21249
Connections should only be reused when using the same "share" (and
perhaps some additional conditions), but instead of fixing this flaw,
this change completely disables connection reuse for SMB. This protocol
is about to get dropped soon anyway.
Reported-by: Osama Hamad
Closes#21238
