| Commit message (Collapse) | Author | Age | Files | Lines |
|\
| |
| | |
Updated random functions. Fixes #2907
|
| |
| |
| | |
Can't test it, but it should work just fine. Can be extended to MinGW w/ GCC and other compilers sometime in the future
|
| |
| |
| | |
It actually doesn't use it because `rand_s` doesn't.
|
| |
| |
| | |
For Windows, `rand_s` has been available since Windows XP (see https://msdn.microsoft.com/en-us/library/sxtz2fa8.aspx). It gives a better quality random number in a larger range (the max is actually `0xffffffff`).
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
1. Introduce TerminalCmd enum and resetStyle that can be issued to issue
resetAttributes within styledEcho arguments.
2. Use templates to resolve styledEcho arguments for performance reasons.
3. Try to avoid calling trailing write "\n" and reset attributes where possible.
|
| | |
|
| | |
|
|/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Motivation
----------
Current implementation does not allow to specify any formatting. It
even cannot output log level, because `writeln()` accepts
`varags[string]` and silently ignores everything before last argument.
Modification
------------
Perform formatting in the single place during substitution. Make log
level optional as all other substitution variables. Also make verbose
logging more informative and parseable.
Result
------
Correct handling formatting and substitutions. Machine-friendly default
output of verbose logger.
|
|\
| |
| | |
Fixed copying of aggregates in JS.
|
| | |
|
|\ \
| | |
| | | |
Add nre to stdlib
|
| | |
| | |
| | |
| | | |
This reverts commit dc60a51e1545d8c73bc9415d1045864b72cfda0b.
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| |\ \
| | | |
| | | |
| | | |
| | | |
| | | | |
* 'devel' of https://github.com/Araq/Nim:
Fix #964, fix #1384
Don't inspect typedescs
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* nre-proj/master: (132 commits)
Change to options module
Reweave readme
Better handle errors
Update documentation
Change flags to inline
Improve performance
Add tests for empty or non-empty match
Fix skipping an empty match at the end
Add longer flags
Fix getinfo overflows
Use docweave
Convert readme to RST
Fix result shadowing warning
Throw an exception when replacing with a nil value
Fix potential buffer overflow
Fix zero-length matches for multibyte characters
Make splitting an empty string give 1 empty result
Change endpos to inclusive
Change endpos default from -1 to int.high
Change capture upper bounds to inclusive
...
|
|\ \ \ \
| | | | |
| | | | | |
I64 magics gone
|
| | | | | |
|
| | | | | |
|
| | | | | |
|
| | | | |
| | | | |
| | | | |
| | | | | |
Author: rku <rokups@zoho.com>
|
| | | | |
| | | | |
| | | | |
| | | | | |
Author: rku <rokups@zoho.com>
|
| |_|_|/
|/| | | |
|
|\ \ \ \
| | | | |
| | | | | |
fixed threadpool and atomics to work with Visual Studio 32 and 64 bit
|
| | | | | |
|
| |/ / / |
|
| | | | |
|
| | | | |
|
| | | | |
|
|\ \ \ \
| | | | |
| | | | | |
db: InstantRow and instantRows
|
| |/ / /
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
It is drop-in replacement for Row and fastRows, however instantRows returns a
handle, not seq[string], so no Nim string is created until [] operator is
called on the given handle. Also there is a len() proc returning number of
columns in the handle.
In some situations, when we iterate through many rows, but later we just read
few columns this solution will be quicker than converting all column to Nim
seq[string] on each iteration.
|
|\ \ \ \
| | | | |
| | | | | |
Fixes #2886 (moves compiler function into the compiler)
|
| |/ / /
| | | |
| | | | |
`-|` only used in the compiler for negative indexing
|
|/ / / |
|
| | |
| | |
| | | |
Ranges sourced from <http://www.unicode.org/Public/7.0.0/ucd/PropList.txt>_. Wikipedia also uses these ranges on its information page <http://en.wikipedia.org/wiki/Whitespace_character#Unicode>_. 0xfeff isn't included in the list, but it is a no-break space, so I guess it makes sense. 0x200b is actually a format character, but it is a zero-width space. To fit Unicode, both 0x200b and 0xfeff would be removed.
|
| |/
|/|
| |
| | |
Haven't tested this patch but should work.
|
|\ \
| | |
| | | |
Resolve base type fields in `[](TAny, string)`
|
| | | |
|
|\ \ \
| | | |
| | | | |
Use builtin overflow functions of Clang and GCC (WIP, RFC)
|