f003557674
typedParams will now handle parameters with pragmas
2020-05-23 20:12:56 +03:00
a695d9e7bd
Add shims/macros.getCustomPragmaFixed
2020-05-20 13:42:29 +03:00
10e39dcfe5
Merge pull request #39 from rockcavera/master
...
fixes for `-d:swapBytesBuiltin --cc:vcc`
2020-05-20 10:23:28 +07:00
2218717cf6
fixes for `-d:swapBytesBuiltin --cc:vcc`
...
renamed the `builtin_bswapT` procedures of the vcc compiler to `swapBytesBuiltin`, to then work correctly with `-d:swapBytesBuiltin --cc:vcc`
2020-05-19 21:58:46 -03:00
d0f5be4971
Fix a compilation error when using a 32-bit varint parser
2020-05-12 22:59:46 +03:00
c500d3dda1
Some renames to cater to the latest FastStreams API
2020-05-12 15:02:06 +03:00
5fa6bb2742
Add a note about compiler crash with raises pragma (results)
2020-05-12 19:37:15 +09:00
21bcf97e26
Remove `ranges` from readme
...
deprecated
2020-05-08 21:15:01 +02:00
f125f6e8d1
space
2020-05-08 18:54:36 +03:00
2d9226464d
heterogenous `or`
...
Useful for translating `error` results
2020-05-08 18:54:36 +03:00
638e7acc8e
deprecate ranges ( #31 )
...
Without lifetime tracking, ranges pose a number of issues with regards
to memory safety - use openArray instead which is limited but safe.
In its present form, ranges take a copy of whatever is passed to them
which obscures the copies that they take, leading to misleading and slow
code.
If openArray is not applicable, simply do a seq copy - it's fine and
really doesn't matter in 95% of all cases, then profile and use other
techniques where needed.
2020-05-07 07:11:43 +02:00
720f1a254d
byteutils: compile-time to/fromBytes ( #35 )
2020-05-06 15:51:07 +02:00
8065e36c5a
results: fix non-void [] overload
2020-04-23 06:46:10 +02:00
df7f2e7ed5
Fix a typo [skip ci]
2020-04-22 16:39:37 +03:00
1eed7a72b7
Make any accidental "run-time" usage of declval an error
2020-04-22 16:34:11 +03:00
4acc3866ed
Adds results.Opt
...
It turned out that Nim's Option type is broken for `not nil` and
`requiresInit` types. It will report an udesired warning due to
the way `none` is defined. The `Result` type doesn't suffer from
this problem, so I've transitioned some APIs in NBC to use the
new `results.Opt[T]` type (alias of `Result[T, void]`).
Perhaps we should renamed `Result` to `Res` to make the naming
more consistent and to allow the code to more easily fit in the
80 character per-line budget.
2020-04-22 15:43:26 +03:00
9a0cded592
Add objects.declval
2020-04-22 15:41:42 +03:00
ff755bbf75
change get/[] to always raise Defect and tryGet to do eh bridge mode ( #30 )
...
* also sprinkle mixin randomly across the codebase
2020-04-16 18:23:12 +02:00
8528ce28b4
Writing to FastStreams can have side-effects
2020-04-14 19:36:15 +03:00
805ef4f1b2
result: cleanups
...
* fix defect raising without `$`
* doc updates
* better `?` that doesn't cause raises effect (it should, because of
FieldError but...)
2020-04-11 16:22:37 +02:00
5512e89d4c
a few more cleanups
...
* add raises to endians2, bitops2 since they sometimes call C functions
2020-04-10 22:40:00 +03:00
87ccf8e66b
hex byte: raise instead of assert on hex-to-byte odd lengths
2020-04-10 22:40:00 +03:00
3e1fa974a4
Use a simpler recursion in recordFields as it's no longer an iterator
2020-04-09 20:39:02 +03:00
b06a5b6e32
result -> results ( #27 )
2020-04-07 11:43:07 +02:00
55c2ec8977
result: more performance notes
2020-04-06 15:53:53 +03:00
4d20b25c02
result: fix and/or template evaluation, add perf notes
2020-04-06 15:53:53 +03:00
9414202d53
result: `and` can deal with heterogenous values
2020-04-04 00:28:46 +03:00
86739f99c4
Use defect for normal errors, document exception bridge mode ( #20 )
...
Co-authored-by: Zahary Karadjov <zahary@gmail.com>
2020-04-02 11:39:11 +02:00
4201f46750
Update README.md
2020-04-01 14:22:21 +02:00
fe074766f1
Update README.md
2020-04-01 14:22:04 +02:00
7287fffebe
result: add `expect` helper ( #23 )
...
* result: add `expect` helper
* result: use explicit conversion in mapConvert
2020-04-01 14:16:20 +02:00
d622c07a08
result: cleanups ( #22 )
...
* fix ResultError type
* add/fix documentation
* clean up raises (in preparation for better Defect handling)
* fix toException mixin
* work around compiler bug with more explicit types
* fix capture exception type
* fix result type on `?`
2020-03-30 22:49:13 +02:00
4477f45c40
Merge branch 'basetype'
2020-03-30 19:56:37 +02:00
09e55cd375
stew/objects: add baseType()
2020-03-29 17:34:45 +02:00
989c33a82b
Take advantage of fixes in Nim 1.2
2020-03-25 17:52:09 +02:00
76beeb769e
fix fromBytes for zero length arrays
2020-03-07 23:29:46 +01:00
ca8fb7ebd0
add slice assignment for openarray
2020-03-05 12:11:25 +02:00
598fe151f8
non-cast implementation of toBytes, add fromBytes
2020-03-05 12:11:25 +02:00
50562b515a
Add bitseqs.bitsLen(openarray)
2020-01-29 02:42:50 +02:00
f5be0ab63a
add nim-result
2020-01-23 17:35:30 +01:00
1edeabb453
Renames ( #17 )
...
* renames and warning fixes
* gitignore: nimcache
2019-12-20 13:23:30 +01:00
898f4f47e3
Merge pull request #13 from status-im/bitsetting
...
set, clear, toggle and change bits
2019-12-20 12:55:41 +01:00
a36df61d2b
set, clear, toggle and change bits
...
* use standard bit manipulation naming
* fix signed overflow on large shifts
* add toggle
* add smoke tests
* branch-free changeBit
2019-12-20 12:49:20 +01:00
d8c2a64055
Merge pull request #16 from status-im/comparebyte
...
byteutils: lexicographical less than
2019-12-20 12:48:10 +01:00
9c18a1cc55
byteutils: lexicographical less than
2019-12-19 13:29:38 +01:00
1c1195dba3
Copied base32, 58, 64 from libp2p
2019-12-13 15:31:19 +02:00
e9d75c05f6
fix 1.0 test compat
2019-12-10 15:50:02 +01:00
cf4df3dc50
add toBytes ( #14 )
2019-12-10 15:39:57 +01:00
1c4293b3e7
A new fix for newLit
...
Upstream fix submitted here:
https://github.com/nim-lang/Nim/pull/12542
2019-10-28 14:30:49 +02:00
2c9ca5dabb
recordFields returns more accurate information in nested case objects
2019-10-23 09:36:16 +09:00
Zahary Karadjov
andri lim
rockcavera
Giovanni Petrantoni
Jacek Sieka
Ștefan Talpalaru
Yuriy Glukhov