use utf8;
+{
+ v5.005 => {
+ new => [
+ ['threads' => '', {experimental => 0, stable => 0}],
+ ['<code>B::…</code>', 'backend hooks'],
+ ['<code>qr//</code>' => 'overhauled regular expression engine: precompile operator, lookahead/behind, code, conditions, localised flags'],
+ ['<code>… foreach</code>' => '<code>for(each)</code> as statement modifier, with large ranges optimised as counting loops'],
+ ['<code>…::</code>' => 'implicitly quoted package name'],
+ ['<code>ref $@</code>' => '<code>die</code> passes reference values to exception handlers'],
+ ['<code>INIT {}</code>', 'subs run just before the perl runtime begins execution'],
+ ['<code>tie @…</code>' => 'base class for <code>TIEARRAY</code> implementations, and generally improved supported of tied handlers'],
+ ['<code>substr …,…,…,$replace</code>', 'replacement string in 4th argument'],
+ ['<code>splice …,…,-$length</code>', 'negative length indicates elements to keep at the end of an array'],
+ ['<code>$/</code>', 'integer or scalar separator makes <code><></code> read records instead of lines'],
+ ],
+ release => '1998-07-22',
+ unstable => 0,
+ },
+
v5.6 => {
new => [
['<code>use warnings</code>', 'pragma to enable warnings in lexical scope'],
rhel => '2', # v5.6.0; also in red hat 7.0
solaris => '9', # v5.6.1; 2002-05 eol 2014-10
aix => '5.1', # 2001-05 eol 2006-04
+ opensuse => '7.1',
},
unicode => '3.0.1',
},
centos => '3-5', # v5.8.0 in v3 (2004-03); v5.8.8 in v5 (eol 2017-03)
ubuntu => '4.10',
aix => '5.2', # v5.8.0; v5.8.2 in 5.3 and 6.1 (eol 2017-04-30)
+ freebsd => '4-6',
+ opensuse => '8.1',
},
+ distrosum => "dinosaur platforms such as Solaris 10, AIX 5.2, RHEL 3, SLES 8",
unicode => '3.2.0',
},
['<code>given</code>', 'switch statement to smart-match with <code>when</code>/<code>default</code>', {feature => 'switch', experimental => 'smartmatch'}],
['<code>/(?<name>)/</code>', 'named capture buffers into <code>%+</code>'],
['<code>/(?1)/</code>', 'recursive regular expression patterns'],
+ ['<code>/(?|)/</code>', 'resets capture numbering for each contained branch'],
['<code>/.++/</code>', 'possessive quantifiers <code>?+</code>, <code>*+</code>, <code>++</code> to match greedily'],
['<code>s/keep\K//</code>', 'floating positive lookbehind, efficient alternative for <code>s/(keep)/$1/</code>'],
- ['<code>/\v/, /\h/</code>', 'vertical and horizontal whitespace escapes'],
+ ['<code>/p</code>', 'optionally preserve <code>${^MATCH}</code> variables (avoiding <code>$&</code> penalty until COW in v5.20)'],
+ ['<code>/\v/, /\h/</code>', 'vertical and horizontal whitespace escapes (<code>\V</code> <code>\H</code> to invert); also <code>/\R/</code> for newlines'],
['<code>my $_</code>', 'lexically scoped version of the default variable', {experimental => 'lexical_topic', dropped => v5.23.4}],
['<code>state</code>', 'persistent <code>my</code> variables', {feature => 'state'}],
],
centos => '6', # v5.10.1 (2011-07 eol 2020-11)
ubuntu => '8.10', # v5.10.1 in 10.04 LTS
aix => '7.1', # v5.10.1 (2010-09 eol 2020?)
+ opensuse => '11.0',
},
unicode => '5.0.0',
},
distro => {
solaris => '11', # also v5.8.4; 2010-11 eol 2024-11
ubuntu => '11.10',
+ freebsd => '7',
+ opensuse => '11.3',
},
},
v5.14 => {
new => [
['<code>s///r</code>', 'non-destructive substitution'],
- ['<code>/(?^)/</code>', 'construct to reset to default modifiers'],
['<code>/(?{ m() })/</code>', 'regular expressions can be nested in <code>/(?{})/</code> and <code>/(??{})/</code>', {experimental => 0, stable => v5.20}],
+ ['<code>/dalu</code>', 'regexp modifiers to restrict character classes: either <strong>d</strong>efault, <strong>a</strong>scii, <strong>l</strong>ocale, or <strong>u</strong>nicode semantics.'],
[q"<code>use re '/flags'</code>", 'customise default modifiers'],
+ ['<code>/(?^)/</code>', 'construct to reset to default modifiers'],
['<code>each $ref</code> e.a.', 'array and hash container functions accept references', {experimental => 'postderef', dropped => v5.23.1}],
['<code>FH->method</code>', 'filehandle method calls load IO::File on demand (eg. <code>STDOUT->flush</code>)'],
+ ['<code>\o{}</code>', 'escape sequence for octal values beyond \777'],
],
release => '2011-05-14',
distro => {
- debian => 'wheezy',
+ debian => '7', # wheezy 2013-05 eol 2018-05
ubuntu => '12.04',
+ opensuse => '12.1',
},
+ distrosum => "stable servers such as Debian 7, Ubuntu 12.04, CentOS 7",
unicode => '6.0+#8',
},
distro => {
rhel => '7', # v5.16.3
centos => '7', # v5.16.3 (2014-07 eol 2024-06)
+ freebsd => '9',
+ opensuse => '12.2',
},
unicode => '6.1',
},
new => [
['<code>${^LAST_FH}</code>', 'last read filehandle (used by <code>$.</code>)'],
['<code>/(?[ a + b ])/</code>', 'regex set operations (character substraction <code>-</code>, unions <code>&</code>)', {experimental => 'regex_sets'}],
- ['<code>my sub</code>', 'lexical subroutines (also <code>state</code>, <code>our</code>); buggy before v5.22', {feature => 'lexical_subs', experimental => 'lexical_subs'}],
+ ['<code>my sub</code>', 'lexical subroutines (also <code>state</code>, <code>our</code>); buggy before v5.22', {experimental => 'lexical_subs', stable => v5.26}],
['<code>next $expression</code>', 'loop controls allow runtime expressions'],
[q"<code>no warnings 'experimental::…'</code>", 'mechanism for experimental features, as of now required for <em>smartmatch</em>'],
],
release => '2013-05-18',
distro => {
- ubuntu => '14.04',
+ ubuntu => '14.04 LTS', # trusty
+ opensuse => '13.1',
},
unicode => '6.2',
},
unicode => '6.3',
release => '2014-05-27',
distro => {
- debian => 'jessie',
- ubuntu => '14.10',
+ debian => '8', # jessie 2015-04 eol 2020-05
+ ubuntu => '14.10', # utopic
aix => '7.2',
+ opensuse => '13.2', # 2014-11 eol 2017-01
},
+ distrosum => "stable servers such as Debian 8 (oldstable), Ubuntu 14.10, FreeBSD 10, openSUSE 13.2",
},
v5.22 => {
['<code><<>></code>', 'safe <code>readline</code> ignoring open flags in arguments'],
['<code>/()/n</code>', 'flag to disable numbered capturing, turning <code>()</code> into <code>(?:)</code>'],
['<code>/\b{}/</code>', 'boundary types: <em>gcb</em> (grapheme cluster), <em>sb</em> (sentence), <em>wb</em> (word)'],
- ['<code>&.</code>', '<code>& | ^ ~</code> consistently numeric, dotted operators for strings', {experimental => 'bitwise'}],
+ ['<code>&.</code>', '<code>& | ^ ~</code> consistently numeric, dotted operators for strings', {feature => 'bitwise', experimental => 'bitwise', stable => v5.28}],
[q"<code>use re 'strict'</code>", 'apply stricter syntax rules to regular expression patterns', {experimental => 're_strict'}],
['<code>0x.beep+0</code>', q"hexadecimal floating point notation with binary power; <code>printf '%a'</code> to display"],
+ ['<code><s>??</s></code>', 'single match shorthand (deprecated since v5.14) requires the operator <code><em>m</em>?PATTERN?</code>'],
],
unicode => '7.0',
release => '2015-06-01',
distro => {
- ubuntu => '16.04',
+ ubuntu => '16.04 LTS', # xenial
},
},
],
unicode => '8.0',
release => '2016-05-09',
+ distro => {
+ debian => '9', # stretch 2017-06
+ ubuntu => '17.04', # zesty
+ freebsd => '10',
+ },
},
- v5.25.8 => {
+ v5.26 => {
new => [
['<code><<~EOT</code>', 'indented here-docs, strips same whitespace before delimiter in each line'],
- ['<code>@{^CAPTURE}</code>', q"array of last match's captures, so <code>@{^CAPTURE}[0]</code> is <code>$1</code>"],
+ ['<code>@{^CAPTURE}</code>', q"array of last match's captures, so <code>${^CAPTURE}[0]</code> is <code>$1</code>"],
+ ['<code>//xx</code>', 'extended modifier to also ignore whitespace in bracketed character classes'],
],
unicode => '9.0', # also Script_Extensions/scx in "\p{script}"
- release => '2016-12-20',
+ release => '2017-05-30',
+ distro => {
+ ubuntu => '17.10', # artful
+ },
+ },
+
+ v5.28 => {
+ new => [
+ ['<code>delete %hash{…}</code>', 'hash slices can be deleted with key+value pairs'],
+ ['<code>/(*…)/</code>', 'alphabetic synonyms for assertions, e.g. <code>(*atomic:…)</code> for <code>(?>…)</code> and <code>(*nlb:…)</code> for <code>(?<!…)</code>', {experimental => 'alpha_assertions', stable => v5.31.6}],
+ ['<code>/(*script_run:)/</code>', 'enforces all characters to be from the same script', {experimental => 'script_run', stable => v5.31.6}],
+ ['<code>state @a</code>', 'persistent lexical array or hash variables'],
+ ['<code>perl -i -pe die</code>', 'safe in-place editing: files are replaced only after successful completion'],
+ ['<code>${^SAFE_LOCALES}</code>', 'locales are thread-safe on supported systems, indicated by this variable'],
+ ],
+ unicode => '10.0',
+ release => '2018-06-22',
+ distro => {
+ },
+ },
+
+ v5.30 => {
+ new => [
+ ['<code>/(?<=var+)</code>', 'variable length lookbehind assertions', {experimental => 'vlb'}],
+ ['<code>m(\p{nv=/.*/})</code>', 'match unicode properties by regular expressions', {experimental => 'uniprop_wildcards'}],
+ ['<code><s>my $state if 0</s></code>', 'workaround for <code>state</code> (deprecated since v5.10!) is now prohibited'],
+ [q"<code>qr'\N'</code>", 'Delimiters must be graphemes; unescaped { illegal; \N in single quotes'],
+ ],
+ unicode => '12.1',
+ release => '2019-05-22',
+ },
+
+ v5.32 => {
+ new => [
+ ['<code>isa</code>', 'infix operator to check class instance'],
+ ['<code>$min < $_ <= $max</code>', 'chained comparison repeats inner part as <code>$min < $_ and $_ <= $max</code>'],
+ ['<code>/\p{Name=$var}/</code>', 'match Unicode Name property like \N{} but with interpolation and subpatterns'],
+ [q"<code>open F, '+>>', undef</code>", 'respect append mode on temporary files with mixed access'],
+ ["<code>no feature 'indirect'</code>", 'disable indirect object notation such as <code>new Class</code> instead of <code>Class->new</code>'],
+ ['streamzip', 'program distributed with core IO::Compress::Base to compress stdin into a zip container'],
+ ],
+ unicode => '13.0',
+ release => '2020-06-20',
+ details => 'https://www.effectiveperlprogramming.com/2020/01/perl-v5-32-new-features/',
+ },
+
+ v5.34 => {
+ new => [
+ ['<code>try {} catch</code>', 'exception handling similar to eval blocks', {experimental => 'try'}],
+ ['<code>/{,<i>n</i>}/</code>', 'empty lower bound quantifier is accepted as shorthand for 0'],
+ ['<code>\x{ … }</code>', 'insignificant space within curly braces, also for <code>\b{}</code>, <code>\g{}</code>, <code>\k{}</code>, <code>\N{}</code>, <code>\o{}</code> as well as <code>/{m,n}/</code> quantifiers'],
+ ['<code>0o0</code>', 'octal prefix <code>0o</code> alternative to <code>0…</code> and <code>oct</code>'],
+ ['<code>re::optimization(qr//)</code>', 'debug regular expression optimization information discovered at compile time'],
+ ],
+ release => '2021-05-20',
},
}
git.shiar.nl / sheet.git / blobdiff