regex.3

Newlib 嵌入式 C库 标准实现代码

.Va rm_eo
set to -1.
If a subexpression participated in the match several times,
the reported substring is the last one it matched.
(Note, as an example in particular, that when the RE
.Ql "(b*)+"
matches
.Ql bbb ,
the parenthesized subexpression matches each of the three
.So Li b Sc Ns s
and then
an infinite number of empty strings following the last
.Ql b ,
so the reported substring is one of the empties.)
.Pp
If
.Dv REG_STARTEND
is specified,
.Fa pmatch
must point to at least one
.Ft regmatch_t
(even if
.Fa nmatch
is 0 or
.Dv REG_NOSUB
was specified),
to hold the input offsets for
.Dv REG_STARTEND .
Use for output is still entirely controlled by
.Fa nmatch ;
if
.Fa nmatch
is 0 or
.Dv REG_NOSUB
was specified,
the value of
.Fa pmatch Ns [0]
will not be changed by a successful
.Fn regexec .
.Pp
.Fn Regerror
maps a non-zero
.Fa errcode
from either
.Fn regcomp
or
.Fn regexec
to a human-readable, printable message.
If
.Fa preg
is
.No non\- Ns Dv NULL ,
the error code should have arisen from use of
the
.Ft regex_t
pointed to by
.Fa preg ,
and if the error code came from
.Fn regcomp ,
it should have been the result from the most recent
.Fn regcomp
using that
.Ft regex_t .
.No ( Fn Regerror
may be able to supply a more detailed message using information
from the
.Ft regex_t . )
.Fn Regerror
places the NUL-terminated message into the buffer pointed to by
.Fa errbuf ,
limiting the length (including the NUL) to at most
.Fa errbuf_size
bytes.
If the whole message won't fit,
as much of it as will fit before the terminating NUL is supplied.
In any case,
the returned value is the size of buffer needed to hold the whole
message (including terminating NUL).
If
.Fa errbuf_size
is 0,
.Fa errbuf
is ignored but the return value is still correct.
.Pp
If the
.Fa errcode
given to
.Fn regerror
is first ORed with
.Dv REG_ITOA ,
the
.Dq message
that results is the printable name of the error code,
e.g.\&
.Dq Dv REG_NOMATCH ,
rather than an explanation thereof.
If
.Fa errcode
is
.Dv REG_ATOI ,
then
.Fa preg
shall be
.No non\- Ns Dv NULL
and the
.Va re_endp
member of the structure it points to
must point to the printable name of an error code;
in this case, the result in
.Fa errbuf
is the decimal digits of
the numeric value of the error code
(0 if the name is not recognized).
.Dv REG_ITOA
and
.Dv REG_ATOI
are intended primarily as debugging facilities;
they are extensions,
compatible with but not specified by
.St -p1003.2 ,
and should be used with
caution in software intended to be portable to other systems.
Be warned also that they are considered experimental and changes are possible.
.Pp
.Fn Regfree
frees any dynamically-allocated storage associated with the compiled RE
pointed to by
.Fa preg .
The remaining
.Ft regex_t
is no longer a valid compiled RE
and the effect of supplying it to
.Fn regexec
or
.Fn regerror
is undefined.
.Pp
None of these functions references global variables except for tables
of constants;
all are safe for use from multiple threads if the arguments are safe.
.Sh IMPLEMENTATION CHOICES
There are a number of decisions that
.St -p1003.2
leaves up to the implementor,
either by explicitly saying
.Dq undefined
or by virtue of them being
forbidden by the RE grammar.
This implementation treats them as follows.
.Pp
See
.Xr re_format 7
for a discussion of the definition of case-independent matching.
.Pp
There is no particular limit on the length of REs,
except insofar as memory is limited.
Memory usage is approximately linear in RE size, and largely insensitive
to RE complexity, except for bounded repetitions.
See
.Sx BUGS
for one short RE using them
that will run almost any system out of memory.
.Pp
A backslashed character other than one specifically given a magic meaning
by
.St -p1003.2
(such magic meanings occur only in obsolete
.Bq Dq basic
REs)
is taken as an ordinary character.
.Pp
Any unmatched
.Ql [\&
is a
.Dv REG_EBRACK
error.
.Pp
Equivalence classes cannot begin or end bracket-expression ranges.
The endpoint of one range cannot begin another.
.Pp
.Dv RE_DUP_MAX ,
the limit on repetition counts in bounded repetitions, is 255.
.Pp
A repetition operator
.Ql ( ?\& ,
.Ql *\& ,
.Ql +\& ,
or bounds)
cannot follow another
repetition operator.
A repetition operator cannot begin an expression or subexpression
or follow
.Ql ^\&
or
.Ql |\& .
.Pp
.Ql |\&
cannot appear first or last in a (sub)expression or after another
.Ql |\& ,
i.e. an operand of
.Ql |\&
cannot be an empty subexpression.
An empty parenthesized subexpression,
.Ql "()" ,
is legal and matches an
empty (sub)string.
An empty string is not a legal RE.
.Pp
A
.Ql {\&
followed by a digit is considered the beginning of bounds for a
bounded repetition, which must then follow the syntax for bounds.
A
.Ql {\&
.Em not
followed by a digit is considered an ordinary character.
.Pp
.Ql ^\&
and
.Ql $\&
beginning and ending subexpressions in obsolete
.Pq Dq basic
REs are anchors, not ordinary characters.
.Sh SEE ALSO
.Xr grep 1 ,
.Xr re_format 7
.Pp
.St -p1003.2 ,
sections 2.8 (Regular Expression Notation)
and
B.5 (C Binding for Regular Expression Matching).
.Sh DIAGNOSTICS
Non-zero error codes from
.Fn regcomp
and
.Fn regexec
include the following:
.Pp
.Bl -tag -width REG_ECOLLATE -compact
.It Dv REG_NOMATCH
.Fn regexec
failed to match
.It Dv REG_BADPAT
invalid regular expression
.It Dv REG_ECOLLATE
invalid collating element
.It Dv REG_ECTYPE
invalid character class
.It Dv REG_EESCAPE
.Ql \e
applied to unescapable character
.It Dv REG_ESUBREG
invalid backreference number
.It Dv REG_EBRACK
brackets
.Ql "[ ]"
not balanced
.It Dv REG_EPAREN
parentheses
.Ql "( )"
not balanced
.It Dv REG_EBRACE
braces
.Ql "{ }"
not balanced
.It Dv REG_BADBR
invalid repetition count(s) in
.Ql "{ }"
.It Dv REG_ERANGE
invalid character range in
.Ql "[ ]"
.It Dv REG_ESPACE
ran out of memory
.It Dv REG_BADRPT
.Ql ?\& ,
.Ql *\& ,
or
.Ql +\&
operand invalid
.It Dv REG_EMPTY
empty (sub)expression
.It Dv REG_ASSERT
can't happen - you found a bug
.It Dv REG_INVARG
invalid argument, e.g. negative-length string
.El
.Sh HISTORY
Originally written by
.An Henry Spencer .
Altered for inclusion in the
.Bx 4.4
distribution.
.Sh BUGS
This is an alpha release with known defects.
Please report problems.
.Pp
The back-reference code is subtle and doubts linger about its correctness
in complex cases.
.Pp
.Fn Regexec
performance is poor.
This will improve with later releases.
.Fa Nmatch
exceeding 0 is expensive;
.Fa nmatch
exceeding 1 is worse.
.Fn Regexec
is largely insensitive to RE complexity
.Em except
that back
references are massively expensive.
RE length does matter; in particular, there is a strong speed bonus
for keeping RE length under about 30 characters,
with most special characters counting roughly double.
.Pp
.Fn Regcomp
implements bounded repetitions by macro expansion,
which is costly in time and space if counts are large
or bounded repetitions are nested.
An RE like, say,
.Ql "((((a{1,100}){1,100}){1,100}){1,100}){1,100}"
will (eventually) run almost any existing machine out of swap space.
.Pp
There are suspected problems with response to obscure error conditions.
Notably,
certain kinds of internal overflow,
produced only by truly enormous REs or by multiply nested bounded repetitions,
are probably not handled well.
.Pp
Due to a mistake in
.St -p1003.2 ,
things like
.Ql "a)b"
are legal REs because
.Ql )\&
is
a special character only in the presence of a previous unmatched
.Ql (\& .
This can't be fixed until the spec is fixed.
.Pp
The standard's definition of back references is vague.
For example, does
.Ql "a\e(\e(b\e)*\e2\e)*d"
match
.Ql "abbbd" ?
Until the standard is clarified,
behavior in such cases should not be relied on.
.Pp
The implementation of word-boundary matching is a bit of a kludge,
and bugs may lurk in combinations of word-boundary matching and anchoring.