NAME

Convert::Scalar - convert between different representations of perl scalars

SYNOPSIS

use Convert::Scalar;

DESCRIPTION

This module exports various internal perl methods that change the internal representation or state of a perl scalar. All of these work in-place, that is, they modify their scalar argument. No functions are exported by default.

The following export tags exist:

:utf8   all functions with utf8 in their name
:taint  all functions with taint in their name
:refcnt all functions with refcnt in their name
:ok     all *ok-functions.
utf8 scalar[, mode]

Returns true when the given scalar is marked as utf8, false otherwise. If the optional mode argument is given, also forces the interpretation of the string to utf8 (mode true) or plain bytes (mode false). The actual (byte-) content is not changed. The return value always reflects the state before any modification is done.

This function is useful when you "import" utf8-data into perl, or when some external function (e.g. storing/retrieving from a database) removes the utf8-flag.

utf8_on scalar

Similar to utf8 scalar, 1, but additionally returns the scalar (the argument is still modified in-place).

utf8_off scalar

Similar to utf8 scalar, 0, but additionally returns the scalar (the argument is still modified in-place).

utf8_valid scalar [Perl 5.7]

Returns true if the bytes inside the scalar form a valid utf8 string, false otherwise (the check is independent of the actual encoding perl thinks the string is in).

utf8_upgrade scalar

Convert the string content of the scalar in-place to its UTF8-encoded form (and also returns it).

utf8_downgrade scalar[, fail_ok=0]

Attempt to convert the string content of the scalar from UTF8-encoded to ISO-8859-1. This may not be possible if the string contains characters that cannot be represented in a single byte; if this is the case, it leaves the scalar unchanged and either returns false or, if fail_ok is not true (the default), croaks.

utf8_encode scalar

Convert the string value of the scalar to UTF8-encoded, but then turn off the SvUTF8 flag so that it looks like bytes to perl again. (Might be removed in future versions).

utf8_length scalar

Returns the number of characters in the string, counting wide UTF8 characters as a single character, independent of wether the scalar is marked as containing bytes or mulitbyte characters.

$old = readonly scalar[, $new]

Returns whether the scalar is currently readonly, and sets or clears the readonly status if a new status is given.

readonly_on scalar

Sets the readonly flag on the scalar.

readonly_off scalar

Clears the readonly flag on the scalar.

unmagic scalar, type

Remove the specified magic from the scalar (DANGEROUS!).

weaken scalar

Weaken a reference. (See also WeakRef).

taint scalar

Taint the scalar.

tainted scalar

returns true when the scalar is tainted, false otherwise.

untaint scalar

Remove the tainted flag from the specified scalar.

length = len scalar

Returns SvLEN (scalar), that is, the actual number of bytes allocated to the string value, or undef, is the scalar has no string value.

scalar = grow scalar, newlen

Sets the memory area used for the scalar to the given length, if the current length is less than the new value. This does not affect the contents of the scalar, but is only useful to "pre-allocate" memory space if you know the scalar will grow. The return value is the modified scalar (the scalar is modified in-place).

scalar = extend scalar, addlen=64

Reserves enough space in the scalar so that addlen bytes can be appended without reallocating it. The actual contents of the scalar will not be affected. The modified scalar will also be returned.

This function is meant to make append workloads efficient - if you append a short string to a scalar many times (millions of times), then perl will have to reallocate and copy the scalar basically every time.

If you instead use extend $scalar, length $shortstring, then Convert::Scalar will use a "size to next power of two, roughly" algorithm, so as the scalar grows, perl will have to resize and copy it less and less often.

nread = extend_read fh, scalar, addlen=64

Calls extend scalar, addlen to ensure some space is available, then do the equivalent of sysread to the end, to try to fill the extra space. Returns how many bytes have been read, 0 on EOF or undef> on eror, just like sysread.

This function is useful to implement many protocols where you read some data, see if it is enough to decode, and if not, read some more, where the naive or easy way of doing this would result in bad performance.

nread = read_all fh, scalar, length

Tries to read length bytes into scalar. Unlike read or sysread, it will try to read more bytes if not all bytes could be read in one go (this is often called xread in C).

Returns the total nunmber of bytes read (normally length, unless an error or EOF occured), 0 on EOF and undef on errors.

nwritten = write_all fh, scalar

Like readall, but for writes - the equivalent of the xwrite function often seen in C.

refcnt scalar[, newrefcnt]

Returns the current reference count of the given scalar and optionally sets it to the given reference count.

refcnt_inc scalar

Increments the reference count of the given scalar inplace.

refcnt_dec scalar

Decrements the reference count of the given scalar inplace. Use weaken instead if you understand what this function is fore. Better yet: don't use this module in this case.

refcnt_rv scalar[, newrefcnt]

Works like refcnt, but dereferences the given reference first. This is useful to find the reference count of arrays or hashes, which cannot be passed directly. Remember that taking a reference of some object increases it's reference count, so the reference count used by the *_rv-functions tend to be one higher.

refcnt_inc_rv scalar

Works like refcnt_inc, but dereferences the given reference first.

refcnt_dec_rv scalar

Works like refcnt_dec, but dereferences the given reference first.

ok scalar
uok scalar
rok scalar
pok scalar
nok scalar
niok scalar

Calls SvOK, SvUOK, SvROK, SvPOK, SvNOK or SvNIOK on the given scalar, respectively.

CANDIDATES FOR FUTURE RELEASES

The following API functions (perlapi) are considered for future inclusion in this module If you want them, write me.

sv_upgrade
sv_pvn_force
sv_pvutf8n_force
the sv2xx family

AUTHOR

Marc Lehmann <schmorp@schmorp.de>
http://home.schmorp.de/