Daniel Kolesa
c990134bb7
path fixes on windows
2018-04-15 19:10:58 +02:00
Daniel Kolesa
3d8dbcbab2
noexcept range ctor
2018-04-15 18:57:23 +02:00
Daniel Kolesa
16518fcc83
add a range type to iterate paths by parts
2018-04-15 18:56:25 +02:00
Daniel Kolesa
de04a60251
better implementation of parent/has_parent
2018-04-15 18:00:20 +02:00
Daniel Kolesa
567a3e0b85
noexcept + remove useless methods
2018-04-15 17:40:20 +02:00
Daniel Kolesa
75e4f9b99a
for path parts, use immutable slices (no memory alloc)
2018-04-15 16:34:30 +02:00
Daniel Kolesa
85a9715ac3
implement is_relative/is_absolute
2018-04-15 15:30:30 +02:00
Daniel Kolesa
9cea53491d
only normalize during append when necessary
2018-04-15 15:17:49 +02:00
Daniel Kolesa
7cf307f60a
better, cleaner and faster path concat/append
2018-04-15 01:11:14 +02:00
Daniel Kolesa
2f42383fcd
fixes
2018-04-14 23:33:42 +02:00
Daniel Kolesa
2f180cfc0e
lots of new path functions
2018-04-14 23:21:58 +02:00
Daniel Kolesa
1d6c256b2a
correct separator replacement
2018-04-14 21:38:23 +02:00
Daniel Kolesa
9c2fdd5f67
strip trailing seps
2018-04-14 21:34:24 +02:00
Daniel Kolesa
96a3c9da30
paths are now always in a valid format
2018-04-14 21:28:01 +02:00
Daniel Kolesa
6eee000b49
drive/root/anchor should be strings
2018-04-14 18:03:43 +02:00
Daniel Kolesa
263b686ef0
drive/root/anchor/parent/name/suffix/stem/relative_to for paths
2018-04-14 17:46:44 +02:00
Daniel Kolesa
643aa7e01a
path move semantics
2018-04-14 01:42:15 +02:00
Daniel Kolesa
7e28045942
expand path structure a bit
2018-04-14 01:29:01 +02:00
Daniel Kolesa
a7b37b0d2e
initial skeleton for path module to replace std::filesystem
2018-04-13 18:57:04 +02:00
Daniel Kolesa
45161b9837
more sensible build macros on windows
2018-01-17 22:56:39 +01:00
Daniel Kolesa
827aed2bb9
minor fixes
2018-01-15 21:05:58 +01:00
Daniel Kolesa
69d5d11da1
POD => trivial, also clarify tstrlen
2018-01-13 22:34:14 +01:00
Daniel Kolesa
dab034d9e0
export dem symbols
2018-01-11 23:38:03 +01:00
Daniel Kolesa
a93ae710d0
universal internal strlen for all char types
2018-01-11 23:27:45 +01:00
Daniel Kolesa
375c014f7a
vtable translation unit for thread pool + warns
2018-01-10 02:57:29 +01:00
Daniel Kolesa
1f5e7dc193
fully expand constructors in argparse
2018-01-10 02:30:01 +01:00
Daniel Kolesa
5646f8610b
no need for a generic param
2018-01-10 00:55:50 +01:00
Daniel Kolesa
391f511131
use size_t for chunks/take
2018-01-10 00:54:31 +01:00
Daniel Kolesa
e792613792
suppress some signed conversion warnings
2018-01-10 00:40:12 +01:00
Daniel Kolesa
64843f7aa7
get rid of remaining zero-as-null-pointer-constant usages
2018-01-10 00:34:42 +01:00
Daniel Kolesa
d7d3929226
refactor expression SFINAE tests not to use pointers
2018-01-10 00:22:09 +01:00
Daniel Kolesa
b077b4fad9
respect width/precision
2018-01-08 00:46:09 +01:00
Daniel Kolesa
d47aa75bd3
fix utf-16 string printing
2018-01-08 00:44:57 +01:00
Daniel Kolesa
7becc47143
use the standard trait to check for chars
2018-01-07 22:54:02 +01:00
Daniel Kolesa
96f627f792
treat chars as ints when not %s/%c and %x/%X/etc as unsigned
2018-01-07 22:53:02 +01:00
Daniel Kolesa
36b5316ff6
documentation for all the Unicode APIs
2018-01-07 22:15:34 +01:00
Daniel Kolesa
229b715d2c
add a standard function to write a replacement character
2018-01-07 21:20:57 +01:00
Daniel Kolesa
032d687293
avoid some size checks on utf-32 range
2018-01-07 21:14:59 +01:00
Daniel Kolesa
3db863c68d
allow encode to work on arbitrary ranges
2018-01-07 19:38:47 +01:00
Daniel Kolesa
818fd1e8e8
add utf::unit_bits and encode/iter_u taking number of bits (8/16/32)
2018-01-07 19:27:08 +01:00
Daniel Kolesa
798fcec6c8
remove the iter_u* wrappers, leave iter_u
2018-01-07 19:16:13 +01:00
Daniel Kolesa
fb91f77eb0
remove the encode_u* wrappers (just use generic type-basedd ver)
2018-01-07 19:08:08 +01:00
Daniel Kolesa
4a2e5cd557
completely unify encode funcs
2018-01-07 18:44:12 +01:00
Daniel Kolesa
541fa43cbb
reduce the encode_u* to 1 sink version per variant
2018-01-07 18:22:30 +01:00
Daniel Kolesa
24d1b5ec25
various traits and constants for unicode types
2018-01-07 17:13:53 +01:00
Daniel Kolesa
be803bac7b
remove sink-based decode (covered by encode_u32)
2018-01-07 16:37:45 +01:00
Daniel Kolesa
640a9714f0
universal templated encode, iter for all types
2018-01-07 02:17:05 +01:00
Daniel Kolesa
4343bb408d
implement compare/case_compare for all slice types
2018-01-07 01:15:17 +01:00
Daniel Kolesa
92cfe0aaf5
iter_codes -> iter_u32, iter_bytes -> iter_u8
2018-01-06 02:10:38 +01:00
Daniel Kolesa
d94e75cd18
add by-utf8-byte iteration for all slice types
2018-01-06 02:00:56 +01:00
Daniel Kolesa
4a992d64b5
move back to sane string range comparison operator defs
2018-01-06 01:17:18 +01:00
Daniel Kolesa
7912e699d5
remove char_traits usage
2018-01-06 01:08:19 +01:00
Daniel Kolesa
fa5ae71202
perform validity check when decoding utf-32 into itself
2018-01-06 00:52:50 +01:00
Daniel Kolesa
ed82fa0233
unified length handling for all encodings
2018-01-06 00:42:14 +01:00
Daniel Kolesa
aeb5023b30
relax the rules of zero-argument utf::length
2018-01-06 00:27:04 +01:00
Daniel Kolesa
e5162233d4
add missing inline
2018-01-06 00:17:47 +01:00
Daniel Kolesa
51d7a62bee
eliminate -Wweak-vtables warnings
2018-01-05 22:48:38 +01:00
Daniel Kolesa
770ed476ca
sink-writing decode
2018-01-05 22:16:18 +01:00
Daniel Kolesa
58ccfbe276
direct encoding funcs to u8/u16/uw from any other UTF
2018-01-05 21:49:00 +01:00
Daniel Kolesa
daea42666e
add range-based advancing encode funcs
2018-01-05 21:16:46 +01:00
Daniel Kolesa
61ed0ce71f
printing of utf-16 and wide strings in format
2018-01-05 20:02:14 +01:00
Daniel Kolesa
200919d96f
add funcs to deal with decoding/encoding of wchar_t values/sequences
2018-01-05 19:26:30 +01:00
Daniel Kolesa
be25d42660
refactor unicode impl
2018-01-05 18:55:34 +01:00
Daniel Kolesa
44854072f7
prevent usage on potential broken platforms/toolchains
2018-01-05 03:05:17 +01:00
Daniel Kolesa
d74736d8f4
add utf-16 decoding/encoding support
2018-01-05 02:18:36 +01:00
Daniel Kolesa
723c06c612
various warning fixes with -Weverything
2018-01-03 17:13:38 +01:00
Daniel Kolesa
fd4b26046c
noexcept the ctype funcs
2018-01-03 02:12:23 +01:00
Daniel Kolesa
ad149ff0f6
unicode-aware case-insensitive string compares
2018-01-03 01:22:07 +01:00
Daniel Kolesa
a0337c401e
implement all the unicode ctype funcs, generate the tables
2018-01-02 22:23:18 +01:00
Daniel Kolesa
7c2bfa45df
overload more Unicode stuff for UTF-32 slices
2018-01-02 00:30:58 +01:00
Daniel Kolesa
dd2515de6c
add a utility func to construct a container using a range
2018-01-01 21:06:25 +01:00
Daniel Kolesa
b75f5f4881
implement utf-32 string printing in format
2018-01-01 20:59:39 +01:00
Daniel Kolesa
c35f7377bf
encode characters into utf-8 in format with our API
2018-01-01 03:11:00 +01:00
Daniel Kolesa
0857edfef4
add a function to encode utf-32 to utf-8
2018-01-01 01:02:49 +01:00
Daniel Kolesa
278b6a6269
define string ranges over wchar/char16/char32
2017-12-31 23:42:46 +01:00
Daniel Kolesa
c4f67b08b9
rename codepoint to decode
2017-12-31 20:06:36 +01:00
Daniel Kolesa
b350eced7e
move utf::length wrappper to header
2017-12-31 19:18:08 +01:00
Daniel Kolesa
48a7b45115
remove unneeded forward decl
2017-12-31 19:17:02 +01:00
Daniel Kolesa
1a07db8bac
merge utf impl bits into one place
2017-12-31 19:16:16 +01:00
Daniel Kolesa
d3cdbe2fcf
expose unicode stuff through string slices
2017-12-31 19:12:51 +01:00
Daniel Kolesa
fb2f9e3b0e
add support for printing unicode and wide characters
2017-12-31 15:32:33 +01:00
Daniel Kolesa
b2ee5c1bd0
add iter_codes to iterate a UTF-8 string by code points
2017-12-31 14:50:48 +01:00
Daniel Kolesa
d6a13d8f97
expose multibyte-to-codepoint conversion
2017-12-31 03:26:15 +01:00
Daniel Kolesa
3c75d7db98
add some initial code for upcoming unicode support
2017-12-31 03:01:25 +01:00
Daniel Kolesa
f7929a1b45
ditch char_traits in string stuff
2017-12-15 23:32:06 +01:00
Daniel Kolesa
84fc2bc9c4
do not rely on global locale when converting in format
2017-11-13 17:15:30 +01:00
Daniel Kolesa
e5a21382af
much better integer formatting (less stack space, better precision handling)
2017-11-13 02:13:42 +01:00
Daniel Kolesa
baf0dd4ca6
MB_CUR_MAX is not a constant
2017-11-12 20:25:58 +01:00
Daniel Kolesa
db28b66892
workaround the awful bullshit when formatting numbers under locale
...
Because of the C++ locale APIs and libstdc++ being worthless trash,
we need to resort to this kind of nonsense in order to avoid
gibberish when dealing with grouping and decimal separators.
Libc++ gets this right (comes up with ASCII style representations
when requesting locale facets dealing with char type) but for some
dumb reason libstdc++ comes up with representations that are
garbage even when using a UTF-8 locale, so I guess we'll deal
with it this way for the time being...
That said, all of this code is probably broken on systems that
don't use Unicode and honestly I don't care.
2017-11-12 20:14:26 +01:00
Daniel Kolesa
84572dfd01
use C locale in I/O and format by default
...
Should probably get rid of state in locale-aware APIs, too.
2017-11-12 19:15:43 +01:00
Daniel Kolesa
f882292f2a
get rid of gcc specific workarounds
2017-11-10 20:19:25 +01:00
Daniel Kolesa
1893f4b941
gcc warning fix
2017-11-10 20:11:58 +01:00
Daniel Kolesa
ec9ddb2aad
yield current task in single-threaded coroutine scheduler on spawn
...
This schedules tasks more aggressively, letting side tasks always run.
2017-11-05 22:10:22 +01:00
Daniel Kolesa
67525af4e5
gcc/libstdc++ 7.x fixes
2017-11-03 12:56:34 +01:00
Daniel Kolesa
146da105b9
forward the writer out of get_line
2017-06-19 17:00:53 +02:00
Daniel Kolesa
2ce488002a
use <> for includes everywhere
2017-06-19 16:59:36 +02:00
Daniel Kolesa
3034918496
move glob matching impl into a source file
2017-06-16 18:55:08 +02:00
Daniel Kolesa
bee56e6a52
add glob example
2017-06-11 18:51:29 +02:00
Daniel Kolesa
cca5952f93
fix the ** pattern not including itself's dir when used in dir position
2017-06-11 18:43:04 +02:00
Daniel Kolesa
511ea88161
proper handling of [!...] glob patterns
2017-06-11 18:17:44 +02:00
Daniel Kolesa