Age | Commit message (Collapse) | Author |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The order of operators are different from the initial test, so this increases
the code coverage in wee-calc.c.
|
|
|
|
|
|
"revscr:"
|
|
"modifier:name,data,string" (issue #60)
|
|
|
|
|
|
|
|
|
|
"calc:..." (issue #997)
|
|
It is possible to trigger a segmentation fault while processing
an evaluation of repeating string. On a Linux 64 bit system,
enter this (or adjust arguments for 32 bit accordingly):
/eval -n ${repeat:1073741824,----}
It will overflow an integer calculation because int instead of
size_t is used. Proper check of int limitations fixes this issue.
I haven't changed this specific piece of code to size_t because it
would crash in other parts of the code tree instead. For now, int
is a limitating factor when it comes to strings (and should be
enough for sane use cases).
Signed-off-by: Tobias Stoeckmann <tobias@stoeckmann.org>
|
|
#1394)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
(closes #1322)
|
|
The value -1 means it's a standard split, but empty items are kept, and
separators are not removed at beginning/end of string.
|
|
|
|
The following special sequences are not supported in regular expressions on
FreeBSD:
- "\w": replaced with "[a-zA-Z0-9_]"
- "\S": replaced with "[^ ]" (it should be "[^ \t\n\r\f\v]", but in practice
only spaces could be a problem when we use this sequence).
|
|
Tests for size on screen are made on U+2EE9 (CJK Radical Simplified Yellow)
instead of U+24B62 (CJK Unified Ideograph-24B62) which returns a length of 1 on
FreeBSD and 2 on Linux.
|
|
|
|
|
|
|
|
string_{encode,decode}_base64
|
|
|
|
|
|
|
|
|
|
"repeat:count,string" (closes #958)
|
|
|
|
|
|
|
|
|
|
|