mirror of
https://github.com/jart/cosmopolitan.git
synced 2025-07-07 11:48:30 +00:00
Write more redbean unit tests
- Fix DescribeSigset() - Introduce new unix.rmrf() API - Fix redbean sigaction() doc example code - Fix unix.sigaction() w/ more than two args - Improve redbean re module API (non-breaking) - Enhance Lua with Python string multiplication - Make third parameter of unix.socket() default to 0
This commit is contained in:
Justine Tunney
parent
c5b9902ac9
commit
1c83670229
20 changed files with 738 additions and 204 deletions
|
|
@ -357,6 +357,9 @@ LUA ENHANCEMENTS
|
|||
example, you can say `"hello %s" % {"world"}` instead of
|
||||
`string.format("hello %s", "world")`.
|
||||
|
||||
- redbean supports a string multiply operator, like Python. For
|
||||
example, you can say `"hi" * 2` instead of `string.rep("hi", 2)`.
|
||||
|
||||
- redbean supports octal (base 8) integer literals. For example
|
||||
`0644 == 420` is the case in redbean, whereas in upstream Lua
|
||||
`0644 == 644` would be the case.
|
||||
|
|
@ -1495,6 +1498,7 @@ CONSTANTS
|
|||
Logging anything at this level will result in a backtrace and
|
||||
process exit.
|
||||
|
||||
|
||||
────────────────────────────────────────────────────────────────────────────────
|
||||
LSQLITE3 MODULE
|
||||
|
||||
|
|
@ -1531,6 +1535,7 @@ LSQLITE3 MODULE
|
|||
we provide an APE build of the SQLite shell which you can use to
|
||||
administrate your redbean database. See the sqlite3.com download above.
|
||||
|
||||
|
||||
────────────────────────────────────────────────────────────────────────────────
|
||||
RE MODULE
|
||||
|
||||
|
|
@ -1540,29 +1545,144 @@ RE MODULE
|
|||
|
||||
# Example IPv4 Address Regular Expression (see also ParseIP)
|
||||
p = re.compile([[^([0-9]{1,3})\.([0-9]{1,3})\.([0-9]{1,3})\.([0-9]{1,3})$]])
|
||||
m,a,b,c,d = p:search(𝑠)
|
||||
m,a,b,c,d = assert(p:search(𝑠))
|
||||
if m then
|
||||
print("ok", tonumber(a), tonumber(b), tonumber(c), tonumber(d))
|
||||
else
|
||||
print("not ok")
|
||||
end
|
||||
|
||||
re.search(regex:str,text:str[,flags:int]) → [match[,group_1,...]]
|
||||
Shortcut for re.compile plus regex_t*:search.
|
||||
re.search(regex:str, text:str[, flags:int])
|
||||
├─→ match:str[, group1:str, ...]
|
||||
└─→ nil, re.Errno
|
||||
|
||||
re.compile(regex:str[,flags:int]) → regex_t*
|
||||
Compiles regular expression, using the POSIX extended syntax. This
|
||||
has an O(2^𝑛) cost, so it's a good idea to do this from your
|
||||
/.init.lua file. Flags may contain re.BASIC, re.ICASE, re.NOSUB,
|
||||
and/or re.NEWLINE. See also regcomp() from libc.
|
||||
Searches for regular expression match in text.
|
||||
|
||||
regex_t*:search(text:str[,flags:int]) → [match[,group_1,...]]
|
||||
Executes regular expression. This has an O(𝑛) cost. This returns
|
||||
nothing (nil) if the pattern doesn't match anything. Otherwise it
|
||||
pushes the matched substring and any parenthesis-captured values
|
||||
too. Flags may contain re.NOTBOL or re.NOTEOL to indicate whether
|
||||
or not text should be considered at the start and/or end of a
|
||||
line.
|
||||
This is a shorthand notation roughly equivalent to:
|
||||
|
||||
preg = re.compile(regex)
|
||||
patt = preg:search(re, text)
|
||||
|
||||
`flags` defaults to zero and may have any of:
|
||||
|
||||
- `re.BASIC`
|
||||
- `re.ICASE`
|
||||
- `re.NEWLINE`
|
||||
- `re.NOSUB`
|
||||
- `re.NOTBOL`
|
||||
- `re.NOTEOL`
|
||||
|
||||
This has exponential complexity. Please use re.compile() to
|
||||
compile your regular expressions once from `/.init.lua`. This
|
||||
API exists for convenience. This isn't recommended for prod.
|
||||
|
||||
This uses POSIX extended syntax by default.
|
||||
|
||||
re.compile(regex:str[, flags:int]) → re.Regex
|
||||
├─→ preg:re.Regex
|
||||
└─→ nil, re.Errno
|
||||
|
||||
Compiles regular expression.
|
||||
|
||||
`flags` defaults to zero and may have any of:
|
||||
|
||||
- `re.BASIC`
|
||||
- `re.ICASE`
|
||||
- `re.NEWLINE`
|
||||
- `re.NOSUB`
|
||||
|
||||
This has an O(2^𝑛) cost. Consider compiling regular
|
||||
expressions once from your `/.init.lua` file.
|
||||
|
||||
If `regex` is an untrusted user value, then `unix.setrlimit`
|
||||
should be used to impose cpu and memory quotas for security.
|
||||
|
||||
This uses POSIX extended syntax by default.
|
||||
|
||||
────────────────────────────────────────────────────────────────────────────────
|
||||
RE REGEX OBJECT
|
||||
|
||||
re.Regex:search(text:str[, flags:int])
|
||||
├─→ match:str[, group1:str, ...]
|
||||
└─→ nil, re.Errno
|
||||
|
||||
Executes precompiled regular expression.
|
||||
|
||||
Returns nothing (nil) if the pattern doesn't match anything.
|
||||
Otherwise it pushes the matched substring and any
|
||||
parenthesis-captured values too. Flags may contain re.NOTBOL
|
||||
or re.NOTEOL to indicate whether or not text should be
|
||||
considered at the start and/or end of a line.
|
||||
|
||||
`flags` defaults to zero and may have any of:
|
||||
|
||||
- `re.NOTBOL`
|
||||
- `re.NOTEOL`
|
||||
|
||||
This has an O(𝑛) cost.
|
||||
|
||||
────────────────────────────────────────────────────────────────────────────────
|
||||
RE ERRNO OBJECT
|
||||
|
||||
re.Errno:errno()
|
||||
└─→ errno:int
|
||||
|
||||
Returns regex error number.
|
||||
|
||||
re.Errno:doc()
|
||||
└─→ description:str
|
||||
|
||||
Returns English string describing error code.
|
||||
|
||||
re.Errno:__tostring()
|
||||
└─→ str
|
||||
|
||||
Delegates to re.Errno:doc()
|
||||
|
||||
────────────────────────────────────────────────────────────────────────────────
|
||||
RE ERRORS
|
||||
|
||||
re.NOMATCH
|
||||
No match
|
||||
|
||||
re.BADPAT
|
||||
Invalid regex
|
||||
|
||||
re.ECOLLATE
|
||||
Unknown collating element
|
||||
|
||||
re.ECTYPE
|
||||
Unknown character class name
|
||||
|
||||
re.EESCAPE
|
||||
Trailing backslash
|
||||
|
||||
re.ESUBREG
|
||||
Invalid back reference
|
||||
|
||||
re.EBRACK
|
||||
Missing `]`
|
||||
|
||||
re.EPAREN
|
||||
Missing `)`
|
||||
|
||||
re.EBRACE
|
||||
Missing `}`
|
||||
|
||||
re.BADBR
|
||||
Invalid contents of `{}`
|
||||
|
||||
re.ERANGE
|
||||
Invalid character range.
|
||||
|
||||
re.ESPACE
|
||||
Out of memory
|
||||
|
||||
re.BADRPT
|
||||
Repetition not preceded by valid expression
|
||||
|
||||
────────────────────────────────────────────────────────────────────────────────
|
||||
RE FLAGS
|
||||
|
||||
re.BASIC
|
||||
Use this flag if you prefer the default POSIX regex syntax. We use
|
||||
|
|
@ -1578,27 +1698,29 @@ RE MODULE
|
|||
may only be used with re.compile and re.search.
|
||||
|
||||
re.NEWLINE
|
||||
Use this flag to change the handling of NEWLINE (\x0a) characters.
|
||||
When this flag is set, (1) a NEWLINE shall not be matched by a "."
|
||||
or any form of a non-matching list, (2) a "^" shall match the
|
||||
zero-length string immediately after a NEWLINE (regardless of
|
||||
re.NOTBOL), and (3) a "$" shall match the zero-length string
|
||||
immediately before a NEWLINE (regardless of re.NOTEOL).
|
||||
Use this flag to change the handling of NEWLINE (\x0a)
|
||||
characters. When this flag is set, (1) a NEWLINE shall not be
|
||||
matched by a "." or any form of a non-matching list, (2) a "^"
|
||||
shall match the zero-length string immediately after a NEWLINE
|
||||
(regardless of re.NOTBOL), and (3) a "$" shall match the
|
||||
zero-length string immediately before a NEWLINE (regardless of
|
||||
re.NOTEOL).
|
||||
|
||||
re.NOSUB
|
||||
Causes re.search to only report success and failure. This is
|
||||
reported via the API by returning empty string for success. This
|
||||
flag may only be used with re.compile and re.search.
|
||||
reported via the API by returning empty string for success.
|
||||
This flag may only be used with re.compile and re.search.
|
||||
|
||||
re.NOTBOL
|
||||
The first character of the string pointed to by string is not the
|
||||
beginning of the line. This flag may only be used with re.search
|
||||
and regex_t*:search.
|
||||
The first character of the string pointed to by string is not
|
||||
the beginning of the line. This flag may only be used with
|
||||
re.search and re.Regex:search.
|
||||
|
||||
re.NOTEOL
|
||||
The last character of the string pointed to by string is not the
|
||||
end of the line. This flag may only be used with re.search and
|
||||
regex_t*:search.
|
||||
The last character of the string pointed to by string is not
|
||||
the end of the line. This flag may only be used with re.search
|
||||
and re.Regex:search.
|
||||
|
||||
|
||||
────────────────────────────────────────────────────────────────────────────────
|
||||
MAXMIND MODULE
|
||||
|
|
@ -1621,6 +1743,7 @@ MAXMIND MODULE
|
|||
|
||||
For further details, please see maxmind.lua in redbean-demo.com.
|
||||
|
||||
|
||||
────────────────────────────────────────────────────────────────────────────────
|
||||
ARGON2 MODULE
|
||||
|
||||
|
|
@ -1688,6 +1811,7 @@ ARGON2 MODULE
|
|||
"password")
|
||||
true
|
||||
|
||||
|
||||
────────────────────────────────────────────────────────────────────────────────
|
||||
UNIX MODULE
|
||||
|
||||
|
|
@ -2276,6 +2400,18 @@ UNIX MODULE
|
|||
thereby assisting with simple absolute filename checks in addition
|
||||
to enabling one to exceed the traditional 260 character limit.
|
||||
|
||||
unix.rmrf(path:str)
|
||||
├─→ true
|
||||
└─→ nil, unix.Errno
|
||||
|
||||
Recursively removes filesystem path.
|
||||
|
||||
Like unix.makedirs() this function isn't actually a system call but
|
||||
rather is a Libc convenience wrapper. It's intended to be equivalent
|
||||
to using the UNIX shell's `rm -rf path` command.
|
||||
|
||||
`path` is the file or directory path you wish to destroy.
|
||||
|
||||
unix.fcntl(fd:int, cmd:int, ...)
|
||||
├─→ ...
|
||||
└─→ nil, unix.Errno
|
||||
|
|
@ -2532,7 +2668,7 @@ UNIX MODULE
|
|||
|
||||
`whence` can be one of:
|
||||
|
||||
- `SEEK_SET`: Sets the file position to `offset`
|
||||
- `SEEK_SET`: Sets the file position to `offset` [default]
|
||||
- `SEEK_CUR`: Sets the file position to `position + offset`
|
||||
- `SEEK_END`: Sets the file position to `filesize + offset`
|
||||
|
||||
|
|
@ -2580,14 +2716,14 @@ UNIX MODULE
|
|||
- `SOCK_CLOEXEC`
|
||||
- `SOCK_NONBLOCK`
|
||||
|
||||
`protocol` defaults to `IPPROTO_TCP` for AF_INET` and `0` for
|
||||
`AF_UNIX`. It can also be:
|
||||
`protocol` may be any of:
|
||||
|
||||
- `IPPROTO_IP`
|
||||
- `IPPROTO_ICMP`
|
||||
- `0` to let kernel choose [default]
|
||||
- `IPPROTO_TCP`
|
||||
- `IPPROTO_UDP`
|
||||
- `IPPROTO_RAW`
|
||||
- `IPPROTO_IP`
|
||||
- `IPPROTO_ICMP`
|
||||
|
||||
unix.socketpair([family:int[, type:int[, protocol:int]]])
|
||||
├─→ fd1:int, fd2:int
|
||||
|
|
@ -2964,18 +3100,19 @@ UNIX MODULE
|
|||
|
||||
Example:
|
||||
|
||||
assert(unix.sigaction(unix.SIGUSR1, function(sig)
|
||||
gotsigusr1 = true
|
||||
end))
|
||||
gotsigusr1 = false
|
||||
assert(unix.raise(unix.SIGUSR1))
|
||||
ok, err = unix.sigsuspend()
|
||||
assert(err:errno == unix.EINTR)
|
||||
if gotsigusr1
|
||||
print('hooray the signal was delivered')
|
||||
else
|
||||
print('oh no some other signal was handled')
|
||||
function OnSigUsr1(sig)
|
||||
gotsigusr1 = true
|
||||
end
|
||||
gotsigusr1 = false
|
||||
oldmask = assert(unix.sigprocmask(unix.SIG_BLOCK, unix.Sigset(unix.SIGUSR1)))
|
||||
assert(unix.sigaction(unix.SIGUSR1, OnSigUsr1))
|
||||
assert(unix.raise(unix.SIGUSR1))
|
||||
assert(not gotsigusr1)
|
||||
ok, err = unix.sigsuspend(oldmask)
|
||||
assert(not ok)
|
||||
assert(err:errno() == unix.EINTR)
|
||||
assert(gotsigusr1)
|
||||
assert(unix.sigprocmask(unix.SIG_SETMASK, oldmask))
|
||||
|
||||
It's a good idea to not do too much work in a signal handler.
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue