aboutsummaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Fix pylint errorJack Lloyd2017-12-211-0/+2
|
* Now dist script can run under python2 or python3Jack Lloyd2017-12-211-1/+1
|
* Run dist.py script through pylintJack Lloyd2017-12-212-35/+32
| | | | Python 2.7 has the io module so no need for StringIO
* Binary file IO for the UTF-8 godsJack Lloyd2017-12-211-2/+2
| | | | | When reading a binary file, unless 'b' is specified Python3 tries parsing the file as UTF-8 and gets cranky when it fails.
* Remove use of Sphinx extlinks extension [ci skip]Jack Lloyd2017-12-201-20/+0
|
* Avoid test failures when PKCSv1.5 signatures are disabled in buildJack Lloyd2017-12-204-10/+17
|
* Fix some shadow and unused parameter warningsJack Lloyd2017-12-204-9/+11
|
* Merge GH #1363 Fix various X509 path building and validation bugsJack Lloyd2017-12-20445-228/+9715
|\
| * Fix various x509 path validation bugs + path building with ambiguous DNsFabian Weissberg2017-12-20445-228/+9715
| | | | | | | | Signed-off-by: Fabian Weissberg <[email protected]>
* | Don't encode AlgorithmIdentifier parameters for ECDSA in X.509 objectsRené Korthaus2017-12-201-2/+5
|/ | | | | RFC 5758 and 4491 mandate that for DSA, ECDSA and GOST, the algorithm identifier "encoding MUST omit the parameters field".
* Correct expected PSS errorJack Lloyd2017-12-191-2/+2
|
* Merge GH #1365 Always set domain encoding correctly when loading an ECC keyJack Lloyd2017-12-192-0/+7
|\
| * Always set domain encoding correctly when loading an ECC keyRené Korthaus2017-12-192-1/+7
| | | | | | | | | | | | | | | | When loading an ECC key from file, the domain encoding was always set to explicit instead of depending on the encoded key file read. This resulted in different encodings and therefore different fingerprints of the same key when encoding the same key twice (once after generation, once after re-reading it).
* | Merge GH #1364 Expose a function returning status when verifying X509 objectsJack Lloyd2017-12-196-59/+99
|\ \
| * | Expose a function returning a status code for verifing X509 objectsJack Lloyd2017-12-196-59/+99
| |/ | | | | | | | | | | | | | | | | The versions returning bool just tell us if it could be verified but don't indicate the problem, everything got binned into "signature error" during verification. Now in the event that the params were invalid, or the signature algorithm couldn't be found, report that as a specific error. See GH #1362
* | Merge GH #1361 Cleanups in ASN.1 codeJack Lloyd2017-12-1922-184/+167
|\ \ | |/ |/|
| * Add accessors to ASN1_Attribute and AlgorithmIdentifierJack Lloyd2017-12-1917-70/+94
| |
| * Fix some other copy+paste code in BER_DecoderJack Lloyd2017-12-194-77/+58
| |
| * Simplify overloads in DER_EncoderJack Lloyd2017-12-192-37/+15
| |
* | Enable processor specific code in module policies [ci skip]Jack Lloyd2017-12-193-0/+15
|/ | | | These were added more recently and weren't enabled in the policy files.
* Avoid tying encoding of CRLs to the current system clockJack Lloyd2017-12-183-20/+68
|
* Add a simple OpenSSL vs Botan benchmark scriptJack Lloyd2017-12-182-0/+217
|
* Accept ChaCha20 as a nameJack Lloyd2017-12-181-0/+6
|
* Correct JSON output (trailing comma) and add byte-per-second valueJack Lloyd2017-12-181-3/+10
|
* Windows has a different return typeJack Lloyd2017-12-171-2/+4
|
* Merge BSD and Winsock variations togetherJack Lloyd2017-12-171-105/+61
|
* Try to debug Boost build failure - amalgamation problem?Jack Lloyd2017-12-171-1/+1
|
* Missing includeJack Lloyd2017-12-171-0/+1
|
* Expose timeouts on the freestanding OCSP request utilsJack Lloyd2017-12-173-8/+27
|
* Add an overall timeout to the HTTP requestJack Lloyd2017-12-171-0/+8
|
* Use HTTP level timeouts instead of polling on the std::futureJack Lloyd2017-12-172-25/+8
|
* Expose timeouts to the HTTP APIJack Lloyd2017-12-173-13/+32
|
* Handle timeouts in BSD socketsJack Lloyd2017-12-171-5/+61
|
* Add timeouts to Asio socketsJack Lloyd2017-12-176-39/+117
|
* PylintJack Lloyd2017-12-172-0/+3
|
* More fixes with recent pylintJack Lloyd2017-12-171-1/+2
|
* Latest pylint doesn't like these assertsJack Lloyd2017-12-171-2/+2
|
* Clean up XMSS self-benchmarkJack Lloyd2017-12-161-6/+4
| | | | Clang complained that the lambda was capturing unnecessary variables.
* Avoid using @deprecated annotation hereJack Lloyd2017-12-161-2/+4
| | | | | | Clang's documentation checker complains since it's not actually annotated as deprecated. Change comment to just indicate that it is not a public API function and will be removed in the future.
* Fix a bug in PEM decodingJack Lloyd2017-12-163-31/+79
| | | | | | | | If the label param was set to a non-empty string, the decoding would treat it as part of the label and then end up throwing a decoding error. The key fix here is adding the clear() call in decode. Private bug report by email. Add some basic tests for the PEM decoder.
* Merge GH #1357 Add ability to fingerprint public keysJack Lloyd2017-12-155-38/+79
|\
| * Add ability to fingerprint public keysJack Lloyd2017-12-145-38/+79
| |
* | Update comment [ci skip]Jack Lloyd2017-12-151-6/+8
| |
* | Fix clang macrosJack Lloyd2017-12-141-3/+3
| |
* | Address review comments: indent code and fix clang regexJack Lloyd2017-12-141-38/+36
| |
* | Detect the compiler version using the preprocessor instead of command lineJack Lloyd2017-12-144-141/+58
|/ | | | Fixes #1314
* Default to not setting a timestamp in the gzip headerJack Lloyd2017-12-142-5/+6
| | | | But allow it to be set via a constructor arg.
* Remove some unused OS feature flagsJack Lloyd2017-12-1317-20/+1
| | | | | We are unlikely to ever use gettimeofday or memset_s going forward. Also MinGW had the wrong flag for VirtualLock
* Drop IRIX supportJack Lloyd2017-12-135-13/+0
| | | | | | It turns out that GCC dropped support for IRIX in 4.8, so likely there was never even a C++11 compiler that could have been used to compile 2.x in any case.
* Simplify how make flags are setJack Lloyd2017-12-121-6/+5
|