Shivaram Lingamneni
8f778c57a3
test more string cases
2020-06-04 18:40:14 -04:00
Shivaram Lingamneni
4391b1ba5a
restrict ASCII mode to printable characters only
2019-12-22 09:57:49 -05:00
Shivaram Lingamneni
781bb6b051
more systematic bad-character check in permissive mode
2019-12-22 09:31:51 -05:00
Shivaram Lingamneni
2d4dbeba1c
disallow some bad characters in foldPermissive
2019-12-22 09:19:28 -05:00
Shivaram Lingamneni
91d6888b7e
fix #693
2019-12-18 12:26:09 -05:00
Shivaram Lingamneni
baa71ba2be
remove unnecessary special-casing for ASCII
2019-10-23 02:18:45 -04:00
Shivaram Lingamneni
332f0d8d91
fix #644
2019-10-16 12:51:40 -04:00
Shivaram Lingamneni
938ce7e435
fix #331
2019-07-14 16:17:37 -04:00
Shivaram Lingamneni
24c46235a3
remove m -> rn skeleton mapping again
2019-06-24 06:23:10 -04:00
Shivaram Lingamneni
b075ea6eb9
use regular Skeleton instead of SkeletonTweaked
...
dan: "I probably wouldn't expect those mappings to cause much
trouble at all for legitimate users at the end of the day"
shivaram: [concurs]
2019-06-24 00:16:42 -04:00
Shivaram Lingamneni
6e9a31a574
use confusables.SkeletonTweaked to fix the test
2019-06-19 05:24:09 -04:00
Shivaram Lingamneni
f17777995b
add a new test
2019-06-19 04:53:15 -04:00
Shivaram Lingamneni
be4d098945
fix an edge case in skeletonization
...
'm' skeletonizes to 'rn' (but is exempted by the isBoring check),
but the fullwidth 'ｍ' does not skeletonize to anything. The root cause
of this is the (still unexplained) patchiness of the skeleton mapping
for fullwidth -> standard-width Latin characters; the fix is to perform
width mapping first, before either skeletonization or isBoring.
2019-06-18 02:34:16 -04:00
Daniel Oaks
8cd5db1194
Restrict idents as other servers do
2019-02-03 18:49:42 +10:00
Shivaram Lingamneni
c34d9e0b72
more lenient casefolding for skeletons
2019-02-03 02:45:02 -05:00
Shivaram Lingamneni
b9b2553a2f
use the TR39 skeleton algorithm to prevent confusables ( #178 )
2019-01-30 21:48:53 -05:00
Shivaram Lingamneni
92eca4d795
add more test cases
2018-12-06 06:46:11 -05:00
Shivaram Lingamneni
40e63dbbe8
fix #306
...
Fix spurious bidi rule violations in casefolding channel names
by stripping the # before starting the casefolding.
2018-12-05 22:39:01 -05:00
Daniel Oaks
ac91a3e484
strings: Follow latest advice on PRECIS regarding string stabilizing
2017-09-06 14:38:55 +10:00
Euan Kemp
449ef4cea1
strings: disallow ':' in nicks
...
This matches the behavior of inspircd at the very least.
Previously, the comment above that section claimed ':' should be
disallowed, but the code didn't do so.
I also simplified the code a little bit and added tests.
2017-07-25 23:28:29 -07:00
