diff options
author | John Peacock <jpeacock@rowman.com> | 2003-01-05 16:28:41 -0500 |
---|---|---|
committer | hv <hv@crypt.org> | 2003-02-10 00:26:50 +0000 |
commit | 129318bdc5341dc6c9c199fa27cbfe9b42b96328 (patch) | |
tree | 96243e4460b9ffaabdeede0bf28f992389fe9266 /lib/version.pm | |
parent | c9d8ec30e843d646cf43a9517acd0a6c4a17a510 (diff) | |
download | perl-129318bdc5341dc6c9c199fa27cbfe9b42b96328.tar.gz |
version objects final(?) patch
Message-ID: <3E18E9D9.2040908@rowman.com>
p4raw-id: //depot/perl@18682
Diffstat (limited to 'lib/version.pm')
-rw-r--r-- | lib/version.pm | 220 |
1 files changed, 161 insertions, 59 deletions
diff --git a/lib/version.pm b/lib/version.pm index 5fd3b31692..15cf81b547 100644 --- a/lib/version.pm +++ b/lib/version.pm @@ -1,4 +1,4 @@ -#!/usr/local/bin/perl -w +#!/usr/bin/perl -w package version; use 5.005_03; @@ -9,7 +9,7 @@ use vars qw(@ISA $VERSION $CLASS); @ISA = qw(DynaLoader); -$VERSION = (qw$Revision: 2.3 $)[1]/10; +$VERSION = (qw$Revision: 2.7 $)[1]/10; $CLASS = 'version'; @@ -31,7 +31,7 @@ version - Perl extension for Version Objects $version = new version "12.2.1"; # must be quoted! print $version; # 12.2.1 print $version->numify; # 12.002001 - if ( $version > 12.2 ) # true + if ( $version gt "v12.2" ) # true $vstring = new version qw(v1.2); # must be quoted! print $vstring; # 1.2 @@ -39,7 +39,7 @@ version - Perl extension for Version Objects $betaver = new version "1.2_3"; # must be quoted! print $betaver; # 1.2_3 - $perlver = new version "5.005_03"; # must be quoted! + $perlver = new version 5.005_03; # must not be quoted! print $perlver; # 5.5.30 =head1 DESCRIPTION @@ -56,10 +56,29 @@ single underscore. This corresponds to what Perl itself uses for a version, as well as extending the "version as number" that is discussed in the various editions of the Camel book. -However, in order to be compatible with earlier Perl version styles, -any use of versions of the form 5.006001 will be translated as 5.6.1, -In other words, a version with a single decimal place will be parsed -as implicitly having three places between subversion. +There are actually two distinct ways to initialize versions: + +=over 4 + +=item * Numeric Versions - any initial parameter which "looks like +a number", see L<Numeric Versions>. + +=item * V-String Versions - any initial parameter which contains more +than one decimal point, contains an embedded underscore, or has a +leading 'v' see L<V-String Versions>. + +=back + +Both of these methods will produce similar version objects, in that +the default stringification will always be in a reduced form, i.e.: + + $v = new version 1.002003; # 1.2.3 + $v2 = new version "1.2.3"; # 1.2.3 + $v3 = new version v1.2.3; # 1.2.3 for Perl > v5.8.0 + $v4 = new version 1.2.3; # 1.2.3 for Perl > v5.8.0 + +Please see L<Quoting> for more details on how Perl will parse various +input values. Any value passed to the new() operator will be parsed only so far as it contains a numeric, decimal, or underscore character. So, for example: @@ -67,28 +86,113 @@ contains a numeric, decimal, or underscore character. So, for example: $v1 = new version "99 and 94/100 percent pure"; # $v1 == 99.0 $v2 = new version "something"; # $v2 == "" and $v2->numify == 0 -NOTE: it is strongly recommended that version objects only be created -with numeric values based on the different types of versions in this -documentation, see L<"Types of Versions Objects">. That way, there is -no confusion about what constitutes the version. +However, see L<New Operator> for one case where non-numeric text is +acceptable when initializing version objects. + +=head2 Numeric Versions + +These correspond to historical versions of Perl itself prior to v5.6.0, +as well as all other modules which follow the Camel rules for the +$VERSION scalar. A numeric version is initialized with what looks like +a floating point number. Leading zeros B<are> significant and trailing +zeros are implied so that a minimum of three places is maintained +between subversions. What this means is that any subversion (digits +to the right of the decimal place) that contains less than three digits +will have trailing zeros added to make up the difference. For example: + + $v = new version 1.2; # 1.200 + $v = new version 1.02; # 1.20 + $v = new version 1.002; # 1.2 + $v = new version 1.0023; # 1.2.300 + $v = new version 1.00203; # 1.2.30 + $v = new version 1.002_03; # 1.2.30 See L<Quoting> + $v = new version 1.002003; # 1.2.3 + +All of the preceeding examples except the second to last are true +whether or not the input value is quoted. The important feature is that +the input value contains only a single decimal. + +=head2 V-String Versions + +These are the newest form of versions, and correspond to Perl's own +version style beginning with v5.6.0. Starting with Perl v5.10.0, +this is likely to be the preferred form. This method requires that +the input parameter be quoted, although Perl > v5.9.0 can use bare +v-strings as a special form of quoting. + +Unlike L<Numeric Versions>, V-String Versions must either have more than +a single decimal point, e.g. "5.6.1" B<or> must be prefaced by a "v" +like this "v5.6" (much like v-string notation). In fact, with the +newest Perl v-strings themselves can be used to initialize version +objects. Also unlike L<Numeric Versions>, leading zeros are B<not> +significant, and trailing zeros must be explicitely specified (i.e. +will not be automatically added). In addition, the subversions are +not enforced to be three decimal places. + +So, for example: + + $v = new version "v1.2"; # 1.2 + $v = new version "v1.002"; # 1.2 + $v = new version "1.2.3"; # 1.2.3 + $v = new version "v1.2.3"; # 1.2.3 + $v = new version "v1.0003"; # 1.3 + +In additional to conventional versions, V-String Versions can be +used to create L<Beta Versions>. + +In general, V-String Versions permit the greatest amount of freedom +to specify a version, whereas Numeric Versions enforce a certain +uniformity. See also L<New Operator> for an additional method of +initializing version objects. =head2 Object Methods Overloading has been used with version objects to provide a natural interface for their use. All mathematical operations are forbidden, -since they don't make any sense for versions. For the subsequent -examples, the following two objects will be used: +since they don't make any sense for versions. + +=over 4 + +=item * New Operator - Like all OO interfaces, the new() operator is +used to initialize version objects. One way to increment versions +when programming is to use the CVS variable $Revision, which is +automatically incremented by CVS every time the file is committed to +the repository. + +=back + +In order to facilitate this feature, the following +code can be employed: + + $VERSION = new version qw$Revision: 2.7 $; + +and the version object will be created as if the following code +were used: + + $VERSION = new version "v2.6"; + +In other words, the version will be automatically parsed out of the +string, and it will be quoted to preserve the meaning CVS normally +carries for versions. + +For the subsequent examples, the following two objects will be used: $ver = new version "1.2.3"; # see "Quoting" below $beta = new version "1.2_3"; # see "Beta versions" below +=over 4 + =item * Stringification - Any time a version object is used as a string, a stringified representation is returned in reduced form (no extraneous zeros): +=back + print $ver->stringify; # prints 1.2.3 print $ver; # same thing +=over 4 + =item * Numification - although all mathematical operations on version objects are forbidden by default, it is possible to retrieve a number which roughly corresponds to the version object through the use of the @@ -96,15 +200,26 @@ $obj->numify method. For formatting purposes, when displaying a number which corresponds a version object, all sub versions are assumed to have three decimal places. So for example: +=back + print $ver->numify; # prints 1.002003 +=over 4 + =item * Comparison operators - Both cmp and <=> operators perform the same comparison between terms (upgrading to a version object automatically). Perl automatically generates all of the other comparison -operators based on those two. For example, the following relations hold: +operators based on those two. In addition to the obvious equalities +listed below, appending a single trailing 0 term does not change the +value of a version for comparison purposes. In other words "v1.2" and +"v1.2.0" are identical versions. + +=back + +For example, the following relations hold: - As Number As String Truth Value - --------- ------------ ----------- + As Number As String Truth Value + --------- ------------ ----------- $ver > 1.0 $ver gt "1.0" true $ver < 2.5 $ver lt true $ver != 1.3 $ver ne "1.3" true @@ -129,10 +244,24 @@ notation and stick with it, to reduce confusion. See also L<"Quoting">. =head2 Quoting Because of the nature of the Perl parsing and tokenizing routines, -you should always quote the parameter to the new() operator/method. The -exact notation is vitally important to correctly determine the version -that is requested. You don't B<have> to quote the version parameter, -but you should be aware of what Perl is likely to do in those cases. +certain initialization values B<must> be quoted in order to correctly +parse as the intended version, and additionally, some initial values +B<must not> be quoted to obtain the intended version. + +Except for L<Beta versions>, any version initialized with something +that looks like a number (a single decimal place) will be parsed in +the same way whether or not the term is quoted. In order to be +compatible with earlier Perl version styles, any use of versions of +the form 5.006001 will be translated as 5.6.1. In other words, a +version with a single decimal place will be parsed as implicitly +having three places between subversions. + +The complicating factor is that in bare numbers (i.e. unquoted), the +underscore is a legal numeric character and is automatically stripped +by the Perl tokenizer before the version code is called. However, if +a number containing a single decimal and an underscore is quoted, i.e. +not bare, that is considered a L<Beta Version> and the underscore is +significant. If you use a mathematic formula that resolves to a floating point number, you are dependent on Perl's conversion routines to yield the version you @@ -144,13 +273,6 @@ but other operations are not likely to be what you intend. For example: $V2 = new version 100/9; # Integer overflow in decimal number print $V2; # yields 11_1285418553 -You B<can> use a bare number, if you only have a major and minor version, -since this should never in practice yield a floating point notation -error. For example: - - $VERSION = new version 10.2; # almost certainly ok - $VERSION = new version "10.2"; # guaranteed ok - Perl 5.9.0 and beyond will be able to automatically quote v-strings (which may become the recommended notation), but that is not possible in earlier versions of Perl. In other words: @@ -161,14 +283,17 @@ earlier versions of Perl. In other words: =head2 Types of Versions Objects -There are three basic types of Version Objects: +There are two types of Version Objects: + +=over 4 =item * Ordinary versions - These are the versions that normal modules will use. Can contain as many subversions as required. In particular, those using RCS/CVS can use one of the following: - $VERSION = new version (qw$Revision: 2.3 $)[1]; # all Perls - $VERSION = new version qw$Revision: 2.3 $[1]; # Perl >= 5.6.0 +=back + + $VERSION = new version qw$Revision: 2.7 $; and the current RCS Revision for that file will be inserted automatically. If the file has been moved to a branch, the @@ -177,12 +302,16 @@ have only two. This allows you to automatically increment your module version by using the Revision number from the primary file in a distribution, see L<ExtUtils::MakeMaker/"VERSION_FROM">. +=over 4 + =item * Beta versions - For module authors using CPAN, the convention has been to note unstable releases with an underscore in the version string, see L<CPAN>. Beta releases will test as being newer than the more recent stable release, and less than the next stable release. For example: +=back + $betaver = new version "12.3_1"; # must quote obeys the relationship @@ -196,38 +325,11 @@ As a matter of fact, if is also true that where the subversion is identical but the beta release is less than the non-beta release. -=item * Perl-style versions - an exceptional case is versions that -were only used by Perl releases prior to 5.6.0. If a version -string contains an underscore immediately followed by a zero followed -by a non-zero number, the version is processed according to the rules -described in L<perldelta/Improved Perl version numbering system> -released with Perl 5.6.0. As an example: - - $perlver = new version "5.005_03"; - -is interpreted, not as a beta release, but as the version 5.5.30, NOTE -that the major and minor versions are unchanged but the subversion is -multiplied by 10, since the above was implicitly read as 5.005.030. -There are modules currently on CPAN which may fall under of this rule, so -module authors are urged to pay close attention to what version they are -specifying. - =head2 Replacement UNIVERSAL::VERSION In addition to the version objects, this modules also replaces the core UNIVERSAL::VERSION function with one that uses version objects for its -comparisons. So, for example, with all existing versions of Perl, -something like the following pseudocode would fail: - - package vertest; - $VERSION = 0.45; - - package main; - use vertest 0.5; - -even though those versions are meant to be read as 0.045 and 0.005 -respectively. The UNIVERSAL::VERSION replacement function included -with this module changes that behavior so that it will B<not> fail. +comparisons. =head1 EXPORT |