CHANGES:
0.10.9
#10, fix versionArgs (broken in 0.10.8)
0.10.8
Avoid compilation warnings on GHC 7.8
#9, add --numeric-version flag
Update the copyright year
Change GetOpt.usageInfo to be more like GetOpt
packaged for wip by pho.
A blazingly fast HTML combinator library for the Haskell programming
language. The Text.Blaze module is a good starting point, as well as
this tutorial: http://jaspervdj.be/blaze/tutorial.html.
packaged for wip.
A vault is a persistent store for values of arbitrary types. It's like
having first-class access to the storage space behind IORefs.
The data structure is analogous to a bank vault, where you can access
different bank boxes with different keys; hence the name.
Also provided is a locker type, representing a store for a single element.
- Fixed test failures due to the introduction of "Negative repeat count
does nothing" warning in Perl 5.21.1.
- Fixed more test failures when DBD::Firebird is installed but Firebird
isql cannot be found.
- Fixed registry file naming issues on Win32 for the SQLite engine, and
as well as the tests that failed because of it.
- Worked around Config::GitLike bug on Windows in the target test.
- Changed the exit value for an attempt to deploy to an up-to-date
database from 1 to 0. In other words, it no longer looks like an error
(Issue #147).
2.07 Mon Jul 28 02:48:49 CDT 2014
Try one more time to deal with version issues.
2.01 Sat Jul 26 01:52:46 CDT 2014
Version numbers now come from the current build rather than Makefile.PL.
Upping the version to v2.00 makes it easier to keep track of it all.
Makefile uses stringy, .pm uses v-string.
Hopefully that will keep everyone happy.
For libs_curr:
Avoid truncating dirent's with embedded newlines in taint-hack
by using single-line mode: m{ (.+) }xs
Add "scalar" option to export the first directory found only as
a scalar (vs list). Mainly useful with things like config dir's
where the first one will always be used in any case:
use FindBin::libs qw( base=etc scalar );
exports "$etc" rather than "@etc" (saves using $etc[0]
everyplace). Also works with
use FindBin::libs qw( base=etc export=config scalar );
to export $config rather than an array.
Added some examples.
Replace given block to avoid nastygrams from post-5.16 perls.
Add stub libs.pm as placeholder for version.
Finally remembered to fix typo (thanks to x.guimard).
1.11 Wed Jul 2 22:34:00 CDT 2014
R.I.P. Build.PL.
see Makefile.PL
Fix broken assignment of rel2abs from Cwd rather than
File::Spec::Functions.
Add test for abs_path or rel2abs surviving both '//' and 'cwd' --
if neither of them is available or working the rest of this is
a waste.
Note: I don't have access to windows for testing. If any of this
fails please contact me and we can figure out what is wrong.
1.9.1 Tue May 27 08:20:41 CDT 2014
Minor cleanup blows up in 5.12; works in 5.14. Given that 5.20 is
about to be on the street: the current version is pushed
up to 5.14 (i.e., if ref $^V and v5.14 le $^V then the
installed version is *_curr, otherwise it is *_5_8).
Upstream changes:
1.19 Fri Nov 22 07:15 2013
- Remove the OP tree munging optimization since it wasn't in the
end really speeding things up. (chocolateboy)
- Require Time::HiRes explicitly because CentOS cripples the base
perl install.
1.18 Mon Jun 17 18:07 2013
- Revert fixes for implicitly-lvalue getters for now since that
actually breaks user code. It seems it's not just because the
users are naughty, so more investigation required.
1.17 Mon Jun 17 07:09 2013
- For Hashes: Implement predicates that check definedness and
existance explicitly. The traditional "predicates" check
definedness. It's conceivable to also want a bool-check
type predicate. File a ticket if you need that.
- Fix bug regarding getters being implicitly lvalue by
returning the internal SV*. Instead, we now use TARG.
Upstream changes:
1.000 2014-07-16 09:55:29-04:00 America/New_York
[*** INCOMPATIBLE CHANGES ***]
- Attributes for custom accessors *MUST* be declared for them to be
set via the constructor. It is no longer sufficient for a method of
the same name to exist.
- Unknown constructor arguments are ignored rather than fatal; they are
not included in the object. Special instructions for using BUILD to
hide constructor arguments from validation are irrelevant and have
been removed.
- These changes make Class::Tiny-based classes easier to subclass by
more advanced object frameworks like Moose or Moo.
0.015 2014-07-13 23:10:47-04:00 America/New_York
[CHANGED]
- lowered minimum perl to 5.006
Add missing DEPENDS
Upstream changes:
NetworkX 1.9
Release date: 21 June 2014
Support for Python 3.1 is dropped in this release.
Highlights
Completely rewritten maximum flow and flow-based connectivity algorithms with backwards incompatible interfaces
Community graph generators
Stoer-Wagner minimum cut algorithm
Linear-time Eulerian circuit algorithm
Linear algebra package changed to use SciPy sparse matrices
Algebraic connectivity, Fiedler vector, spectral ordering algorithms
Link prediction algorithms
Goldberg-Radzik shortest path algorithm
Semiconnected graph and tree recognition algorithms
Upstream changes:
1.5.7 (2014-05-29)
Bug fixes:
Skip the traceback on "Broken pipe" signal. (Issue #275)
Do not exit when an option in setup.cfg or tox.ini is not recognized.
Check the last line even if it does not end with a newline. (Issue #286)
Always open files in universal newlines mode in Python 2. (Issue #288)
Upstream changes:
Version 0.81 (2013-04-08)
* Correct reverseName() for IPv6 addresses, so IP('::1').reverseName() returns correct.
* Add network mask awareness to v46map()
* Fix Python 3 errors in IPSet class
* Make IPSet base class be object when MutableSet isn't available, fixing
errors in Python 2.5
Version 0.80 (2013-03-26)
------------
* Drop support of Python older than 2.4
* Python 3 does not need 2to3 conversion anymore (same code base)
* Fix adding of non-adjacent networks:
192.168.0.0/24 + 192.168.255.0/24 made 192.168.0.0/23
* Fix adding networks that don't create a valid subnet:
192.168.1.0/24 + 192.168.2.0/24 made 192.168.1.0/23
* Fix adding with an IPv6 address where .int() was < 32 bits made IPy believe it
was an IPv4 address:
::ffff:0/112 + ::1:0:0/112 made 255.255.0.0/111
* Add support of IPSets
* Add support for subtracting a network range
* Prevent IPv4 and IPv6 ranges from saying they contain each other
* Add a .v46map() method to convert mapped address ranges
such as IP('::ffff:192.168.1.1'); RFC 4291
* Change sort order to more natural:
IPv4 before IPv6; less-specific prefixes first (/0 before /32)
Version 0.76 (2013-03-19)
-------------------------
* ip == other and ip != other doesn't fail with an exception anymore if other
is not a IP object
* Add IP.get_mac() method: get the 802.3 MAC address from IPv6 RFC 2464
address.
* Fix IP('::/0')[0]: return an IPv6 instead of an IPv4 address
Upstream changes:
0.016 2014-06-21 04:04:55Z
- fix prereq problem with last release - now no longer testing the
example scripts for user installs
0.015 2014-06-20 03:36:33Z
- adjusted packaging and tests to become perl-5.6 friendly,
including only using core or dual-lifed prerequisites
0.014 2014-03-01 22:08:26Z
- fix test that was failing when FOO or BAR environment variables
were set (RT#93447)
0.013 2013-12-14 21:26:06Z
- update configure_requires checking in Makefile.PL
Upstream changes:
0.05 2014-03-26
- Improved the DESCRIPTION section, particularly the first paragraph.
- Added a bunch more Exporter modules to the SEE ALSO section.
0.04 2014-03-26
- Added Constant::Exporter::Lazy to SEE ALSO
- Had forgotten to put README and Changes in MANIFEST. Sigh.
0.03 2014-03-25
- Now maintained by NEILB
- Fixed typo reported by dsteinbrunner in RT#87096
- Now use strict and warnings
- Added github repo to metadata and pod
- Added README
- Specified min perl version 5.6.0 in metadata
- Set PREREQ_PM, TEST_REQUIRES, CONFIGURE_REQUIRES in Makefile.PL
- Added a load of modules to SEE ALSO.
Update DEPENDS
Upstream changes:
2.023003 2014-07-21 21:22:27-05:00 America/Chicago
- Fix ToJSON to use accessors if it needs to (Thanks Kevin Benson!)
- Fix silly typo in ::Explain (Good catch Jonathan W. Taylor!)
2.023002 2014-06-28 15:04:15-05:00 America/Chicago
- Remove silly layer of subtesting (thanks to new features in Test::Roo)
- Remove silly sub wrapper in Explain internals (should moderately increase
performance)
2.023001 2014-06-14 12:39:08-05:00 America/Chicago
- Add handy SYNOPSIS to ::DateMethods1 (thanks for asking rjbs!)
2.023000 2014-05-26 19:33:01-05:00 America/Chicago
- Add ::Shortcut::Explain RS helper
2.022000 2014-05-03 10:39:30-05:00 CST6CDT
- Add ::Shortcut::ResultsExist RS helper (Olaf Alders)
- Add abstract to ::DateMethods1::Announcement (Gregor Herrmann)
2.021001 2014-04-06 11:43:36-05:00 America/Chicago
- Fix ::RemoveColumns to work with a specified `columns` (Anthony DeRobertis)
(Fixes GH#27) Also fixes RT#91977/GH#24)
2.021000 2014-04-01 20:12:40-05:00 America/Chicago
- Create Shortcut::Page and Shortcut::LimitedPage ResultSet helpers (wreis)
2.020001 2014-03-05 10:33:46CST-0600 America/Chicago
- Make ::DateMethods1 tests paralellizable (thanks Alexander Hartmaier!)
- fix ::Helper::ResultSet::DateMethods1 for Oracle (thanks Alexander Hartmaier!)
- fix ABSTRACT on ::Helper::ResultSet::DateMethods1
2.020000 2014-03-04 08:31:39-06:00 America/Chicago
- Add ::Helper::ResultSet::DateMethods1
- Add abstract to ::Schema::LintContents
2.019004 2014-02-14 07:53:19 America/Chicago
- Make IgnoreWantarray's search die in void context
2.019003 2014-02-07 22:21:47-06:00 America/Chicago
- Fix ->copy on rows with proxied ResultSet methods (thanks moltar for the
test!) (NOTE: This fix is what requires upgrading to DBIC 0.08260)
2.019002 2014-01-12 09:40:41 America/Chicago
- Pick SQL for random row selection in a cleaner way
- Stop using Class::MOP::load_class (RT#91035)
- Really stop using RT
2.019001 2013-11-23 10:19:28 America/Chicago
- Fix typo in ::CorrelateRelationship (Getty)
DBIx::Introspector is a module factored out of the DBIx::Class
database detection code. Most code that needs to detect which
database it is connected to assumes that there is a one-to-one
mapping from database drivers to database engines. Unfortunately
reality is rarely that simple. For instance, DBD::ODBC is typically
used to connect to SQL Server, but ODBC can be used to connect
to PostgreSQL, MySQL, and Oracle. Additionally, while ODBC is the
most common way to connect to SQL Server, it is not the only option,
as DBD::ADO can also be used.
