rishi/scryer-prolog
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
3,764 Commits 2 Branches 18 Tags
902712695c33154679bf9c3e6ecc954d69b9d37f
Commit Graph
2923 Commits
Author SHA1 Message Date
Mark Thom 0060c8988a use -1 as eof_code for binary streams (#555) 2020-05-22 15:58:07 -06:00
Matthieu Wipliez cc9d6587a3 Implementation of UTF-8 bytes to chars 2020-05-22 23:33:12 +02:00
Mark Thom f9aad34d28 Merge branch 'master' of https://github.com/mthom/rusty-wam 2020-05-22 14:50:51 -06:00
Mark Thom b2d720b853 correct get_byte/1 not emitted -1 upon discovery of end_of_stream position (#555) 2020-05-22 14:50:40 -06:00
Markus Triska 48c0b0ab3c ENHANCED: Faster format/3 for binary streams. 
This speeds up web servers considerably when sending binary files.
 2020-05-22 17:30:29 +02:00
Markus Triska cbbed310c5 ADDED: format/3, writing formatted output to a stream 
Both binary and text streams are supported.
 2020-05-22 17:23:51 +02:00
Mark Thom 9e9c3b6342 modify int_pow so that a power of -1 is valid (#548) 2020-05-21 22:33:22 -06:00
Markus Triska 1108c99688 document that lists of integers can be specified if encoding(octet) is used 
This seems to be a good compromise: The API now strongly encourages
lists of characters, which are ideally suited to represent text, and
also allows lists of bytes if the encoding(octet) option is used.
 2020-05-21 13:41:57 +02:00
Markus Triska 93da831ef0 correct mode indication for ed25519_verify/4 
Noted by @notoria in #545. Many thanks!
 2020-05-21 01:27:06 +02:00
Markus Triska 72bf45912e "PrivateKey" --> "KeyPair" 2020-05-21 00:14:15 +02:00
Markus Triska 5d94276c4f remove several mentions of list of bytes in predicate descriptions 
This is to focus more on the intended core representation of text: In
Prolog, text is ideally represented as a list of characters, and this
is what we want to encourage, especially given Scryer's compact
representation for strings.

For the time being, library(crypto) still also supports lists of bytes in
several predicates. This will likely be removed at some point in the
future. Please use lists of characters to make your code future-proof.
 2020-05-21 00:07:42 +02:00
Markus Triska b43f27030e require PKCS#8 v2 format for better security 
Notably, this format requires that the public key also be present.
This format is what ed25519_new_keypair/1 generates, and it is
strongly encouraged for higher security.
 2020-05-20 23:55:21 +02:00
Markus Triska 2845f55157 ADDED: ed25519_keypair_public_key/2, relating a key pair to its public key 2020-05-20 23:43:28 +02:00
Markus Triska e254d84710 clarify format of public key 2020-05-20 23:17:42 +02:00
Markus Triska 0ef7b5488d ADDED: ed25519_new_keypair/1 to dynamically create a new Ed25519 key pair 2020-05-20 23:11:56 +02:00
Markus Triska 56b04f8df8 use LessSafeKey to simplify the implementation of authenticated encryption 
The nonce is explicitly specified, and the application programmer
must (and always had to) ensure that it is unique for a given key.
 2020-05-20 22:32:39 +02:00
Markus Triska 2d036c6b25 "codes" --> "bytes" 
Support for lists of bytes may be dropped from library(crypto). Use
lists of characters to make your code future-proof. Lists of
characters will always be supported due to their compactness,
and because Prolog applications should move towards characters.
 2020-05-20 22:32:39 +02:00
Markus Triska ad3be5c848 ADDED: Public key signatures and signature verification with Ed25519 2020-05-20 22:32:39 +02:00
Mark Thom a0544cc345 revert printing of characters 2020-05-19 17:34:55 -06:00
Markus Triska 21b7348c6a import member/2 from library(lists) 2020-05-19 20:13:06 +02:00
Mark Thom ad6fc89a24 Merge pull request #533 from triska/master 
ADDED: Support for SHA-3 algorithms in crypto_data_hash/3
 2020-05-19 14:23:12 -03:00
Markus Triska 0bd9831eec ADDED: phrase_from_file/3 in library(pio) 
This allows us to specify type(binary), and read from binary files
with DCGs.
 2020-05-19 18:52:18 +02:00
Markus Triska dec2ef70c7 better error handling in crypto_data_hkdf/4 
Noted by @notoria in #533. Many thanks!
 2020-05-19 18:22:25 +02:00
Markus Triska d19d8ea770 crypto_data_hkdf/4: do not crash for length > usize::max_value() 
For now, we fail silently in such cases.

Noted by @notoria in #533. Many thanks!
 2020-05-19 17:31:36 +02:00
Markus Triska 34f7752c0f use more fixnums in cryptographic routines 2020-05-19 17:03:02 +02:00
Markus Triska da4d061067 correct option processing in crypto_data_decrypt/6 2020-05-19 17:02:33 +02:00
Markus Triska 85155439be use Fixnums for bytes in hashing. 
Suggested by @notoria in #533. Many thanks!
 2020-05-19 16:43:44 +02:00
Markus Triska dd32e69061 use matching to select the hashing algorithm 
Suggested by @notoria in #533. Many thanks!
 2020-05-19 16:37:50 +02:00
Mark Thom 2eac902c33 correct fixnum overflow on negation (#528) 2020-05-19 10:45:54 -06:00
Markus Triska 9f5322d309 ADDED: Support for BLAKE2 algorithms in crypto_data_hash/3. 2020-05-19 12:44:10 +02:00
Markus Triska f5c2f6f9e9 ADDED: Support for SHA-3 algorithms in crypto_data_hash/3 2020-05-19 12:10:46 +02:00
Mark Thom e0e3b180e7 accomodate \0\ in partial strings, print null as \0\ (#267, #526), update prolog parser, version bump 2020-05-18 23:45:34 -06:00
Mark Thom 041dc039d4 add Addr::Lis as case in PartialStringTail (#530) 2020-05-18 12:27:26 -06:00
Mark Thom ac755b0ada Merge pull request #527 from triska/master 
Add type checks to increase robustness of library(crypto)
 2020-05-18 11:18:58 -03:00
Mark Thom c14259060c fix partial_string_tail panic (#530) 2020-05-18 11:14:27 -06:00
Markus Triska a423eb5323 stronger validation of input lists for cryptographic routines 
Example:

    ?- crypto_data_hkdf(Var, 32, Bs, []).
    caught: error(instantiation_error,must_be/2)

Reported by @notoria in #527. Many thanks!
 2020-05-18 13:28:52 +02:00
Markus Triska fd732550d8 crypto_data_hkdf/4: Fail if the length is too long. 
Due to the way the counter is constructed in the HKDF specification,
the requested output length can be at most 255 times the size of the
digest algorithm's output.

Reported by @notoria in #527. Many thanks!
 2020-05-18 13:21:38 +02:00
Markus Triska fac7ba70c8 crypto_password_hash/3: fail if the number of iterations is too high 
Discussed in #527.
 2020-05-18 13:11:39 +02:00
Markus Triska e9f8b35918 centralize reasoning about encoding 2020-05-18 11:29:35 +02:00
Markus Triska 23034dd4f5 raise instantiation errors for variable encoding 
Reported by notoria in #527.

Note that from a declarative perspective, it would indeed be valid to
give answers for both available encodings.
 2020-05-18 11:21:38 +02:00
Markus Triska 70ad44adfd type check for length argument in crypto_data_hkdf/4 
Reported by @notoria in #527.
 2020-05-18 11:12:59 +02:00
Markus Triska 95ca8a2630 throw instantiation error if the list of options contains a variable (#523) 
Many thanks to @notoria for the test case!
 2020-05-18 09:58:09 +02:00
Mark Thom eecd809d9a fix atom_codes (#521) 2020-05-17 16:34:52 -06:00
Mark Thom 5720b7b94c eliminate Addr::CharCode (#519), fix atom_code panic (#521), fix failure involving lists of character codes (#520), bump version number 2020-05-17 15:19:27 -06:00
Markus Triska 3855d7ea02 type test for salt in crypto_password_hash/3 2020-05-17 22:01:22 +02:00
Mark Thom 9f3351469d solve overflow of left arithmetic shift (#518) 2020-05-17 12:20:22 -06:00
Mark Thom b3cff6555f propagate syntax errors from read_term when they're not UnexpectedEOF (#507) 2020-05-16 23:10:38 -06:00
Mark Thom 69706ecab0 fix sign error in shl on fixnum with n >= 63 (#499) 2020-05-16 23:00:37 -06:00
Mark Thom 05d3b97eae correct for second argument when (is)/2 is a tail call 2020-05-16 20:22:17 -06:00
Mark Thom a38fb41e37 Merge branch 'master' of https://github.com/mthom/rusty-wam 2020-05-16 14:51:19 -06:00