The latest Go release, version 1.7, arrives six months after 1.6. Most of its
changes are in the implementation of the toolchain, runtime, and libraries.
There is one minor change to the language specification. As always, the release
maintains the Go 1 promise of compatibility. We expect almost all Go programs
to continue to compile and run as before.
There is one tiny language change in this release. The section on terminating
statements clarifies that to determine whether a statement list ends in a
terminating statement, the “final non-empty statement” is considered the end,
matching the existing behavior of the gc and gccgo compiler toolchains. In
earlier releases the definition referred only to the “final statement,” leaving
the effect of trailing empty statements at the least unclear. The go/types
package has been updated to match the gc and gccgo compiler toolchains in this
respect. This change has no effect on the correctness of existing programs.
Go 1.7 adds support for macOS 10.12 Sierra. This support was backported to Go
1.6.3. Binaries built with versions of Go before 1.6.3 will not work correctly
on Sierra.
py-sqlite3. Add missing test dependency on py-nose.
Add comments with links to bug reports about test failures.
Bump PKGREVISION for dependency change.
Add 70 to PHP_VERSIONS_ACCEPTED
Upstream changes:
1.1.1 * Add IPv6 support (geoip_country_code_by_*_v6)
* PHP7 support (thanks to Adam and Dan Ackroyd)
1.1.0 * Add geoip_setup_custom_directory() to setup custom directories per request.
* Remove E_NOTICES for IPs not found in database.
* Fix a segfault with geoip_db_get_all_info() on newer libs (bug #64692).
* Add support for geoip_netspeedcell_by_name(), geoip_asnum_by_name() and geoip_domain_by_name() (bug #67121).
* Fix memory leak with custom directories (bug #67138).
* Support changing custom directory via ini_set() (bug #61607).
Version 3.8.2 released 2016-02-14
* Fix implicit cast compiler warning in _speedups.c
* simplejson is now available as wheels for OS X and Windows thanks to Travis-CI
and AppVeyor respectively! Many thanks to @aebrahim for getting this party
started.
https://github.com/simplejson/simplejson/pull/130https://github.com/simplejson/simplejson/issues/122
Version 3.8.1 released 2015-10-27
* Fix issue with iterable_as_array and indent option
https://github.com/simplejson/simplejson/issues/128
* Fix typo in keyword argument name introduced in 3.8.0
https://github.com/simplejson/simplejson/pull/123
Version 3.8.0 released 2015-07-18
* New iterable_as_array encoder option to perform lazy serialization of
any iterable objects, without having to convert to tuple or list.
Version 3.7.3 released 2015-05-31
* Fix typo introduced in 3.7.0 (behavior should be indistinguishable)
e18cc09b68 (commitcomment-11443842)
Version 3.7.2 released 2015-05-22
* Do not cache Decimal class in encoder, only reference the decimal module.
This may make reload work in more common scenarios.
Version 3.7.1 released 2015-05-18
* Fix compilation with MSVC
https://github.com/simplejson/simplejson/pull/119
Version 3.7.0 released 2015-05-18
* simplejson no longer trusts custom str/repr methods for int, long, float
subclasses. These instances are now formatted as if they were exact
instances of those types.
https://github.com/simplejson/simplejson/issues/118
Upstream changes:
0.21 2016-08-16 01:31:28Z
- no changes since 0.20
0.20 2016-06-19 02:41:02Z (TRIAL RELEASE)
- switch to plain old Exporter, removing build_* subs from the API,
Upstream changes:
0.30 2016-08-17
* Source repo moved from Subversion to Github
* Add -p to pdfstamp
* Add -g, -p, -u to pdfbklt
* Bug fixes
* Wasn't installing on Windows Perl 5.22 and up
* [rt.cpan.org] #116492: make install fails under Strawberry Perl
* [rt.cpan.org] #110855: Fix pod2man errors
* [rt.cpan.org] #86452: Parse error reading array
* [rt.cpan.org] #78351: Invalid version format (non-numeric data)
* [rt.cpan.org] #41085: ASCII85 decode broken
* [rt.cpan.org] #35871: lzw compression fixup
* [rt.cpan.org] #32210: pdfstamp patch for use strict
* [rt.cpan.org] #31353: adding first test
* fix Dict::read_stream() to write long streams to file as intended
* fix LZWDecode to handle streams longer than 4096 bytes
* close INFILE at start of release()
* Fix Dict to allow single stream filters rather than requiring an array
Upstream changes:
0.19 2016-08-18 18:52:33Z
- fix checking of SvUTF8 flag
0.18 2016-08-16 01:07:54Z
- no changes since 0.17
0.17 2016-06-27 22:45:27Z (TRIAL RELEASE)
- fixed parsing error where the name contains a single colon but the last
separator is :: (Leon Timmermans)
0.16 2016-06-21 03:41:40Z (TRIAL RELEASE)
- support binary and unicode symbol names (Leon Timmermans, Aristotle
Pagaltzis, Peter Rabbitson, Reini Urban, PR#8)
this fixes UnicodeEncodeError under py27 noticed via py-stdnum and
vatnumber (without py-suds) when checking VIES.
Changelog
Recent changes (2014/2015):
Plug-in system to support for WSSE (Web-Services Security extensions)
WSSE UsernameToken, UsernameDigestToken and BinaryTokenSignature support
Pythonic XML Security Library basic implementation (canonicalization, SHA1 hashing and RSA signing / verification using X509 digital certificates)
Improved SOAP Fault details
Several fixes (basic python3 support, CDATA, )
Ongoing efforts:
Unit Tests update & clean up (removing old tests, better framework, fixing non-deterministic results, etc.)
WSDL advanced support (unifying nested elements structure dialects)
Python3 support for WSSE XMLSec (M2Crypto alternatives?)
Source code refactory to improve readability and maintainability
changed depency on py-suds to documented fallback py-simplesoap
in order to support py3.
changes from 1.3 to 1.4
-----------------------
* add modules for the following number formats:
- NIF (Numéro d'Immatriculation Fiscale, French tax identification number)
provided by Dimitri Papadopoulos
- NIR (French personal identification number) provided by Dimitri
Papadopoulos
- SIRET (a French company establishment identification number) provided by
Yoann Aubineau
- NHS (United Kingdom National Health Service patient identifier)
- T.C. Kimlik No. (Turkish personal identification number)
* add an implementation of the Damm algorithm
* ensure data files are properly closed
* documentation improvements
* extend test suite
* a number of minor bug fixes and improvements
changes from 1.2 to 1.3
-----------------------
* fix ISO 6346 check digit calculation (thanks Ronald Paloschi)
* fix problem with check_vies_approx() (thanks Lionel Elie Mamane)
* fix problem with SOAP client when using suds
* include documentation that was previously on the Trac wiki
changes from 1.1 to 1.2
-----------------------
* add modules for the following number formats:
- Austrian Company Register Numbers
- CNPJ (Cadastro Nacional da Pessoa Jurídica, Brazillian company identifier)
- UID (Unternehmens-Identifikationsnummer, Swiss business identifier)
- VAT, MWST, TVA, IVA, TPV (Mehrwertsteuernummer, the Swiss VAT number)
- CUSIP number (financial security identification number)
- Wertpapierkennnummer (German securities identification code)
- Isikukood (Estonian Personcal ID number)
- Finnish Association Identifier
- Y-tunnus (Finnish business identifier)
- SEDOL number (Stock Exchange Daily Official List number)
- IMO number (International Maritime Organization number)
- ISIN (International Securities Identification Number)
- RFC (Registro Federal de Contribuyentes, Mexican tax number)
- PESEL (Polish national identification number)
- REGON (Rejestr Gospodarki Narodowej, Polish register of economic units)
- ИНН (Идентификационный номер налогоплательщика, Russian tax identifier)
* add an alternate VIES check (check_vies_approx()) that includes a proof
(certificate) that the check was performed
* fall back to pysimplesoap if suds is unavailable
* test and code quality improvements
changes from 1.0 to 1.1
-----------------------
* add number formats based on the implementation in the vatnumber module:
- NIPT (Numri i Identifikimit për Personin e Tatueshëm, Albanian VAT number)
- CUIT (Código Único de Identificación Tributaria, Argentinian tax number)
- RUT (Rol Único Tributario, Chilean national tax number)
- NIT (Número De Identificación Tributaria, Colombian identity code)
- COE (Codice operatore economico, San Marino national tax number)
* add modules for the following number formats:
- Cedula (Dominican Republic national identification number)
- RNC (Registro Nacional del Contribuyente, Dominican Republic tax number)
- Kennitala (Icelandic personal and organisation identity code, provided by
Tuomas Toivonen)
- VSK number (Virðisaukaskattsnúmer, Icelandic VAT number, provided by
Tuomas Toivonen)
- ISO 9362 (Business identifier codes, provided by Tony Bajan)
- MVA (Merverdiavgift, Norwegian VAT number, provided by Tuomas Toivonen)
- Orgnr (Organisasjonsnummer, Norwegian organisation number, provided by
Tuomas Toivonen)
- Orgnr (Organisationsnummer, Swedish company number, provided by Tomas
Thor Jonsson)
* add Croatia to list of EU VAT numbers
* update getcnloc script
* various small fixes and additional tests for existing modules
Version 1.18 August 2016
* configure.ac : patch by Greg Troxel <gdt@lexort.com>
CPPFLAGS was set with +=, which is a bash extension and fails on
POSIX-conforming shells. This resulted in sqlite3 support not being
enabled on NetBSD, even though the sqlite3 library was found and
linked in.
* fixed some minor problems reported by Greg Troxel (doc, NEWS)
* in local/network.cpp commented out // removed_obs.clear();
* fixed a formatting bug in outlaying terms (due to c++11)
* fixed a bug in HTML parser test
* simplified class LocalPoint
* fixed a bug in sql export reported by Dominik Hladik
* added @documentencoding UTF-8 into gama.texi
* Dario Teixeira: migrate syntax extension from OCaml-pcre to OCaml-re.
* Rudi Grinberg: migrate core library from OCaml-pcre to OCaml-re.
* Rudi Grinberg: add unit tests for regular expressions.
* Evgenii Lepikhin: add support for types UUID and JSONB.
* Philippe Wang: add support for type CITEXT.
.. changelog::
:version: 1.0.4
:released: Thu Mar 10 2016
.. change::
:tags: feature, test
The default test runner is now py.test. Running "python setup.py test"
will make use of py.test instead of nose. nose still works as a test
runner as well, however.
.. change::
:tags: bug, lexer
:pullreq: github:19
Major improvements to lexing of intricate Python sections which may
contain complex backslash sequences, as well as support for the bitwise
operator (e.g. pipe symbol) inside of expression sections distinct
from the Mako "filter" operator, provided the operator is enclosed
within parentheses or brackets. Pull request courtesy Daniel Martin.
.. change::
:tags: feature
:pullreq: bitbucket:16
Added new method :meth:`.Template.list_defs`. Pull request courtesy
Jonathan Vanasco.
.. changelog::
:version: 1.0.3
:released: Tue Oct 27 2015
.. change::
:tags: bug, babel
:pullreq: bitbucket:21
Fixed an issue where the Babel plugin would not handle a translation
symbol that contained non-ascii characters. Pull request courtesy
Roman Imankulov.
.. changelog::
:version: 1.0.2
:released: Wed Aug 26 2015
.. change::
:tags: bug, installation
🎟️ 249
The "universal wheel" marker is removed from setup.cfg, because
our setup.py currently makes use of conditional dependencies.
In 🎫`249`, the discussion is ongoing on how to correct our
setup.cfg / setup.py fully so that we can handle the per-version
dependency changes while still maintaining optimal wheel settings,
so this issue is not yet fully resolved.
.. change::
:tags: bug, py3k
🎟️ 250
Repair some calls within the ast module that no longer work on Python3.5;
additionally replace the use of ``inspect.getargspec()`` under
Python 3 (seems to be called from the TG plugin) to avoid deprecation
warnings.
.. change::
:tags: bug
:pullreq: bitbucket:18
Update the Lingua translation extraction plugin to correctly
handle templates mixing Python control statements (such as if,
for and while) with template fragments. Pull request courtesy
Laurent Daverio.
.. change::
:tags: feature
🎟️ 236
Added ``STOP_RENDERING`` keyword for returning/exiting from a
template early, which is a synonym for an empty string ``""``.
Previously, the docs suggested a bare
``return``, but this could cause ``None`` to appear in the
rendered template result.
.. seealso::
:ref:`syntax_exiting_early`
