summaryrefslogtreecommitdiff
path: root/lib
AgeCommit message (Collapse)Author
28 hoursares: handle channel being destroyed earlyHEADorigin/masterorigin/HEADmasterfor-upstreamStefan Eissing
We are destroying the ares channel already when we shutdown resolving. Querying the pollset afterwards is still happening, especially in event based processing and needs to work in the absence of a channel. Fixes #18317 Reported-by: Natris on github Closes #18318
38 hourslibssh2: fix sftp_statemachine busyloop when *block=truesunriseL
Closes #18311
38 hoursftp: store dir components as start+len instead of memdup'ingDaniel Stenberg
- Avoids allocating every path segment separately - Improved directory handling in connection reuse Closes #18312
38 hoursftp: keep the "raw" URL decoded version of the path aroundDaniel Stenberg
Instead of doing temporary decodes in three different places. Closes #18312
2 daysasyn-thrdd: silence unused parameter warningMarcel Raad
Closes https://github.com/curl/curl/pull/18310
2 daysngtcp2: use custom mem funcsDavid Zhuang
Pass curl's memory functions to the nghttp3 and ngtcp2 functions that allow them. This allows custom memory functions passed by the curl user to be used in nghttp3 and ngtcp2. Closes #18196
2 daysasyn-thrdd: fix Curl_async_pollset without socketpairDaniel Stenberg
- remove a superfluous if() - edit the #ifdef to only do the timeout checks when socketpair support is missing, as that code flow could not be reached otherways anyway CodeSonar pointed out possible issue that led me to see this. Closes #18306
2 daysbufq: simplify conditionDaniel Stenberg
'result' is always CURLE_AGAIN here Pointed out by CodeSonar Closes #18305
2 daysftp: use 'conn' instead of 'data->conn'Daniel Stenberg
In the ftp_state_use_port function, as we already use that local variable for all other conn accesses. Closes #18304
3 dayscookie: remove expired cookies before listingxfangfang
Closes #18299
5 daysopenssl: auto-pause on verify callback retryStefan Eissing
When an application install its own OpenSSL verify callback and that callback invokes `SSL_set_retry_verify()`, the transfer is automatically paused and does not progress the connect attempt any further until unpaused via `curl_easy_pause(). Added test758 to verify. Ref: #18284 Original PR by @Natris Bug: https://curl.se/mail/lib-2025-08/0012.html Closes #18288
5 dayssendf: getting less data than "max allowed" is okayDaniel Stenberg
Easily triggered when doing an FTP ranged download with the end range larger than the file size. A regression since 30de937bda0 (shipped in curl 8.9.0) Add test 2307 to verify. Reported-by: kupavcevdenis on github Fixes #18283 Closes #18295
6 daysmbedtls: check for feature macros instead of versionViktor Szakats
Drop three interim macros and mbedTLS version checks in favor of feature macros defined by mbedTLS itself. These mbedTLS features require mbedTLS 3.6.0/3.6.1 for production. Earlier versions may have partial/experimental support for them, disabled by default and (AFAICS) without documented build options to enable them. One feature guard already used the native macro instead of the interim one before this patch. (`MBEDTLS_SSL_SESSION_TICKETS`) Ref: https://github.com/curl/curl/pull/18254#issuecomment-3182668817 Closes #18271
7 dayswolfssh: move the sftp upload init out of the state machineDaniel Stenberg
Into its own function to reduce complexity. Takes complexity down from 72 to 50. Closes #18278
7 dayssocks5: split out two functions from the SOCKS5 state machineDaniel Stenberg
Complexity down from 74 to 59 Closes #18277
7 daysopenssl: save and restore OpenSSL error queue in two functionsViktor Szakats
After merging #18228, I reviewed whether the clearing of the error queue may interfere with preceding code. Turns out there may be a preceding `SSL_Connect()` call. This patch replaces the previous fix of clearing the error queue with saving and restoring it in two functions which may be called between the connect call and the `SSL_get_error()` call following it: - `ossl_log_tls12_secret()` - `Curl_ssl_setup_x509_store()` The `ERR_set_mark()`, `ERR_pop_to_mark()` functions are present in all supported OpenSSL and LibreSSL versions. Also in BoringSSL since its initial commit. OpenSSL may modify its error queue in all API calls that can fail. Thanks-to: Viktor Dukhovni Ref: https://github.com/curl/curl/issues/18190#issuecomment-3167702142 Ref: https://github.com/curl/curl/issues/18190#issuecomment-3169211739 Ref: https://github.com/curl/curl/issues/18190#issuecomment-3169988050 Follow-up to 8ec241bc990bc88c4f4f7275d81f9fb75b562a7a #18228 #18190 Ref: e8b00fcd6a0c7ff179cebb3615ccebf1f6790b69 #10432 #10389 Fixes #18190 Closes #18234
8 daysmbedtls: bump minimum version required to 3.2.0Viktor Szakats
3.2.0 was released on July 11, 2022. Ref: #18161 Closes #18254
8 dayscookie: don't treat the leading slash as trailingDaniel Stenberg
If there is only a leading slash in the path, keep that. Also add an assert to make sure the path is never blank. Reported-by: Google Big Sleep Closes #18266
9 dayswebsocket: support CURLOPT_READFUNCTIONStefan Eissing
Add support for CURLOPT_READFUNCTION with WebSocket urls when *not* in connect-only mode, e.g. when using curl_multi_perform. Install the callback function and set CURLOPT_UPLOAD. Return CURL_READFUNC_PAUSE when having nothing more to send and unpause the transfer when more data is ready. This will send the read bytes in a WebSocket BINARY frame. Add support for this mode in the pytest "ws_data" client and have all tests run in 'curl_ws_send/recv' and 'peform' mode as well. Add `curl_ws_start_frame()`. Document, cover in libcurl-ws.md and explain the READFUNCTION mode for websockets. Add example `websocket-updown` for this. Closes #17683
9 daysresolving: dns error tracingStefan Eissing
* Add more tracing information to c-ares errors. * remove CURL_ASYNC_SUCCESS, rename `ares->last_status` to `ares->ares_status`. Give trace explanation for "common" errors * add ares "csv" information to tracing on failure * DoH: invoke `Curl_resolver_error()` on failure to populate error buf Closes #18247
12 daysopenssl: clear errors after a failed `d2i_X509()`Viktor Szakats
Without it, subsequent OpenSSL API calls may fail with an error caught within the OpenSSL `d2i_X509()` (decode) call. It was seen to happen when importing from the Windows certificate store (e.g. with `--ca-native`), and any one of the certificates failed while decoding, then skipped. Behind the scene (and undocumented), the failed decode call is adding an error to an internal OpenSSL error queue. This error is picked up later, at the connect phase, by another OpenSSL API call, which happens to check the error queue, without clearing it first. It made the connect fail with the error collected earlier, while decoding the malformed and discarded certificate. Fix by explicitly clearing the error queue if the decode call fails. Ref: https://docs.openssl.org/3.5/man3/d2i_X509/ `-vvvv` output before this patch: ``` [0-0] == Info: successfully imported Windows ROOT store [0-0] == Info: successfully imported Windows CA store [0-0] == Info: [SSL] SSL_connect() -> err=-1, detail=1 [0-0] == Info: TLS connect error: error:068000DD:asn1 encoding routines::illegal padding [0-0] == Info: [SSL] cf_connect() -> 35, done=0 ``` Mainline OpenSSL (as of 3.5.2) and quictls (as of 3.3.0) are affected. LibreSSL is not affected. (I did not test BoringSSL and AWS-LC) Assisted-by: Stefan Eissing Reported-by: Michał Petryka Fixes #18190 Closes #18228
13 daysschannel: add an error message for client cert not foundJay Satiro
Prior to this change curl would show the generic error message "curl: (58) Problem with the local SSL certificate". Closes https://github.com/curl/curl/pull/18124
13 dayscfilters: make Curl_conn_get_socket simplerStefan Eissing
Since it is only used for the first socket anyway, simplify the function. Closes #18219
13 daysrequest: eliminate request `getheader` bool, reverse `header` defaultStefan Eissing
Deduce that the transfer response expects headers by the protocol handler implementing `write_resp_hd` callback. This eleminates the `getheader` parameter in the `Curl_xfer_setup_*()` methods. Add an implementation to RTSP for `write_resp_hd`, joining the HTTP protocol in the only handlers having it. Reverse the default of request's `header` bit that signals that headers are expected. Default is now FALSE, set to TRUE when setting up the transfer by presence of `write_resp_hd` in the protocol handler. Closes #18218
13 daysares: destroy channel on shutdownStefan Eissing
When we cancel async resolv operations, we have kept an existing ares channel open. This seems unreliable as reported in #18216. To get reliable behaviour, always destroy the ares channel on async shutdown and create a new one on demand. Fixes #18216 Reported-by: devgs on github Closes #18217
13 dayshttp: resurrect addition from 41fe621Stefan Eissing
For some reason, these lines were unintentionally deleted in 6d5570c. Thanks to IoannisGS for noticing. Bug: https://github.com/curl/curl/pull/18179#issuecomment-3162876532 Reported-by: IoannisGS on github Closes #18215
13 dayscurl_quiche.c: add includeKarthik Das
Fixes #18211 Closes #18212
2025-08-06libcurl: reset rewind flag in curl_easy_reset()Oxan van Leeuwen
curl_easy_reset() did not reset the `rewind_read` flag. This caused any handles that previously had a CURLE_SEND_FAIL_REWIND error to get stuck with that error, failing any subsequent requests, even if they didn't have any body at all. Verified in test 3034 Fixes #18206 Closes #18207
2025-08-06multi: fix bad splay managementDaniel Stenberg
The splay tree is a tree where each easy handle can be added *once*. The expire time for that node is the closest expire time for that easy handle. Easy handles can however have more expire times queued up, so when the node is removed from the splay tree because it is the next in line to take care of, we must check if there is another expire time in the queue and then add the node back into the splay. Failing to do the later part, the calling of add_next_timeout after Curl_splaygetbest, would leave the state.expiretime on the previous time stamp, which when could make the next call to Curl_splaygetbest use the wrong time stamp and get a wrong node out, causing trouble. Reported-by: letshack9707 on hackerone Closes #18201
2025-08-06lib: xfer_setup simplifyStefan Eissing
Make variants for transfers that send/receive or do both with just the parameters they need. Split out the shutdown setting into a separate function. Only FTP bothers with that. Closes #18203
2025-08-06lib: replace conn [write]sockfd members by indexStefan Eissing
The `connectdata` members `sockfd` and `writesockfd` needed to by either CURL_SOCKET_BAD or a copy of one of `conn->sock[2]`. When equal to one, that index was used to send/recv/poll the proper socket or connection filter chain. Replace those with `send_idx` and `recv_idx` which are either -1, 0 or 1 to indicate which socket/filter to send/receive on. Closes #18179
2025-08-06splay: add another assert for detected usage problemDaniel Stenberg
Closes #18199
2025-08-06dynbuf: assert on curlx_dyn_addf use with just "%s"Daniel Stenberg
Because it is wasteful and should be done better. Closes #18194
2025-08-05vtls: set seen http version on successful ALPNStefan Eissing
When a HTTP version has been negotiated via ALPN, set the member `conn->httpversion_seen` accordingly. This allows pending transfers to reuse multiplexed http connections before the response to the first transfer has arrived. Fixes #18177 Reported-by: IoannisGS on github Closes #18181
2025-08-05setopt: add helper functions to setopt_long()Daniel Stenberg
- Consistently keep options within ranges - Reduce the maximum maxredirs value to fit a signed short - Removed comments as the place to document the options is not here Closes #18174
2025-08-05vquic: use curl_getenvDavid Zhuang
getenv isn't defined on all platforms, which prevents vquic from building. I specifically ran into this issue building on PlayStation. Closes #18170
2025-08-05openssl: output unescaped utf8 x509 issuer/subject DNsRoberto Hidalgo
Closes #18171
2025-08-05multi: change prefix for the network change bits to CURLMNWC_Daniel Stenberg
Because "CURLM_" is used for curl multi error codes and it is convenient to use the prefix to identify the number family. Closes #18176
2025-08-05wolfssl: rename ML-KEM hybrids to match IETF draftAnthony Hu
Closes #18123
2025-08-05urldata: reduce two long struct fields to unsigned shortDaniel Stenberg
Closes #18173
2025-08-05hostip: cache negative name resolvesDaniel Stenberg
Hold them for half the normal lifetime. Helps when told to transfer N URLs in quick succession that all use the same non-resolving hostname. Done by storing a DNS entry with a NULL pointer for 'addr'. Previously an attempt was made in #12406 by Björn Stenberg that was ultimately never merged. Closes #18157
2025-08-05build: disable `TCP_NODELAY` for emscriptenViktor Szakats
In WebAssembly, using `TCP_NODELAY` fails with: ``` * Could not set TCP_NODELAY: Protocol not available ``` Add a new feature macro in `curl_setup.h` telling whether `TCP_NODELAY` is known to be supported at runtime, when defined at compile-time. Keep `TCP_NODELAY` guards at their current positions to ensure the necessary headers (e.g. `netinet/tcp.h` and `netinet/in.h`) define it. Reported-by: Jeroen Ooms Fixes #17974 Closes #18155
2025-08-04multi: add new information extraction methodStefan Eissing
Adds `curl_off_t curl_multi_get_offt(CURLM *multi_handle, CURLMinfo_offt info)` to the multi interface with enums: * CURLMINFO_XFERS_CURRENT: current number of transfers * CURLMINFO_XFERS_RUNNING: number of running transfers * CURLMINFO_XFERS_PENDING: number of pending transfers * CURLMINFO_XFERS_DONE: number of finished transfers to read * CURLMINFO_XFERS_ADDED: total number of transfers added, ever Add documentation for functions and info enums. Add use in the curl command line tool to replace two static variables counting the same "from the outside". refs #17870 Closes #17992
2025-08-04lib: replace `getsock()` logic with pollsetsStefan Eissing
`getsock()` calls operated on a global limit that could not be configure beyond 16 sockets. This is no longer adequate with the new happy eyeballing strategy. Instead, do the following: - make `struct easy_pollset` dynamic. Starting with a minimal room for two sockets, the very common case, allow it to grow on demand. - replace all protocol handler getsock() calls with pollsets and a CURLcode to return failures - add CURLcode return for all connection filter `adjust_pollset()` callbacks, since they too can now fail. - use appropriately in multi.c and multi_ev.c - fix unit2600 to trigger pollset growth Closes #18164
2025-08-04setopt: split out cookielist() and cookiefile()Daniel Stenberg
into their own sub functions Closes #18162
2025-08-04hostip: do DNS cache pruning in millisecondsDaniel Stenberg
Instead of using integer seconds. Also: if the cache contains over 30,000 entries after first pruning, it makes anoter round and removes all entries that are older than half the age of the oldest entry until it goes below 30,000. Closes #18160
2025-08-03splay: rename KEY_NOTUSED TO SPLAY_SUBNODEDaniel Stenberg
- explains its purpose better - make it global static const - added an assert for a condition that should never happen (that we also catch run-time) Closes #18152
2025-08-01urlapi: allow more path characters "raw" when asked to URL encodeDaniel Stenberg
Setting the path component to contain the letters: ! $ & ' ( ) { } [ ] * + , ; = : @ now leaves them un-encoded when CURLU_URLENCODE is used. Amended test 1560 to verify. Reported-by: Jeroen Ooms Fixes #17977 Closes #18024
2025-08-01CURLOPT: bump `CURLPROTO_*` macros to `long`Viktor Szakats
This patch bumps the size of these macros from `int` to `long`, while keeping their actual values the same. It may cause incompatibilities in user code, requiring the bump of holder variables and/or adding casts: - CURLPROTO_HTTP - CURLPROTO_HTTPS - CURLPROTO_FTP - CURLPROTO_FTPS - CURLPROTO_SCP - CURLPROTO_SFTP - CURLPROTO_TELNET - CURLPROTO_LDAP - CURLPROTO_LDAPS - CURLPROTO_DICT - CURLPROTO_FILE - CURLPROTO_TFTP - CURLPROTO_IMAP - CURLPROTO_IMAPS - CURLPROTO_POP3 - CURLPROTO_POP3S - CURLPROTO_SMTP - CURLPROTO_SMTPS - CURLPROTO_RTSP - CURLPROTO_RTMP - CURLPROTO_RTMPT - CURLPROTO_RTMPE - CURLPROTO_RTMPTE - CURLPROTO_RTMPS - CURLPROTO_RTMPTS - CURLPROTO_GOPHER - CURLPROTO_SMB - CURLPROTO_SMBS - CURLPROTO_MQTT - CURLPROTO_GOPHERS - CURLPROTO_ALL Also: - keep existing cast within the documentation to make sure it applies to older curl versions as well. Closes #18136
2025-08-01openssl: check SSL_write() length on retriesStefan Eissing
When an SSL_write() blocks we need to retry it with the same length as before or stupid OpenSSL freaks out. Remember it, limit any longer sends and fail shorter ones. Fixes #18121 Reported-by: adamse on github Closes #18132
generated by cgit v1.2.3 (git 2.46.0) at 2025-08-20 19:49:03 +0000