revbump after boost-libs update

(adam)

mysql57: get rid of unused patches

(adam)

Updated misc/stellarium

(adam)

stellarium: updated to 0.18.2

0.18.2:
The major changes of this version:
- Improvements and fixes for AstroCalc tool and plugins
- Added support Abell Catalog of Planetary Nebulae and ESO/Uppsala Survey of the ESO(B) Atlas
- Added few new scripts
- Updated code and data

(adam)

Updated meta-pkgs/boost, databases/mysql57, net/powerdns

(adam)

powerdns: updated to 4.1.3

4.1.3:
Improvements
: pdnsutil: use new domain in b2bmigrate
: Update copyright years to 2018
: Lower ���packet too short��� loglevel

Bug Fixes
: Restrict creation of OPT and TSIG RRsets
: Fix handling of user-defined axfr filters return values
: Prevent the GeoIP backend from copying NetMaskTrees around, fixes slow-downs in certain configurations
: Ensure alias answers over TCP have correct name

(adam)

mysql57: updated to 5.7.23

5.7.23:
Functionality Added or Changed

Previously, for the --ssl-mode=VERIFY_IDENTITY or --ssl-verify-server-cert option, the client checked whether the host name that it used for connecting matched the Common Name value in the certificate but not the Subject Alternative Name value. Now, if the client uses OpenSSL 1.0.2 or higher, the client checks whether the host name matches either the Subject Alternative Name value or the Common Name value in the server certificate. Thanks to Dani谷l van Eeden for a patch on which this change was based.

Bugs Fixed

Important Change; Partitioning: After creating partitioned InnoDB tables with very long names, the table_name columns in the corresponding entries in the mysql.innodb_index_stats and mysql.innodb_table_stats system tables were truncated. To fix this issue, the length of the table_name column in each of these tables has been increased from 64 to 199 characters. In both cases, this is now the same as the lengths of these columns in MySQL 8.0.

InnoDB: The ngram full-text search parser permitted comma and period characters to be tokenized as words, which caused an inconsistency between boolean and natural language mode search results. Comma and period characters are no longer tokenized.

InnoDB: An I/O error returned by an fsync() operation is now treated as a hard error.

InnoDB: A schema mismatch error reported during an import tablespace operation failed to print mismatched table flags in a readable format.

InnoDB: A DDL operation failed to wait for a FULLTEXT index optimization operation to finish.

InnoDB: An unnecessary check for read-only transactions was removed from the trx_set_rw_mode() function. Thanks to Sandeep Sethia for the patch.

InnoDB: A DDL operation that added a foreign key constraint raised an assertion when it accessed a stale memory object that belonged to the parent table.

InnoDB: A DDL operation on a table with a FULLTEXT index during full-text index cache synchronization caused an assertion failure.

InnoDB: A failing assertion occurred after initiating a memcached get operation.

InnoDB: A corrupt index ID encountered during a foreign key check raised an assertion.

InnoDB: An internal deadlock during a DDL operation resulted in a long semaphore wait followed by a server exit.

InnoDB: A DDL operation encountered a serious error due to an invalid lock upgrade.

InnoDB: On a Windows 64-bit system, invalid buffer pool configuration values caused the server to exit on startup.

Partitioning: For a partitioned table, partition update time could be incorrect after rebuilding the table or restarting the server.

Partitioning: An extraneous row lock was imposed by an update to a partitioned InnoDB table.

Replication: The log messages generated when a member fails to join the group have been improved, for example when group_replication_group_name on the member joining the group does not match the seed's group_replication_group_name this is now described in the log message.

Replication: The ER_GRP_RPL_SQL_SERVICE_FAILED_TO_RUN_SQL_QUERY error was being logged incorrectly.

Replication: The use of replication filters or binary log filters can cause issues when they are applied to tables that are updated with XA transactions. Filtering of tables could cause an XA transaction to be empty on a replication slave, and empty XA transactions are not supported. Also, with the settings master_info_repository=TABLE and relay_log_info_repository=TABLE on a replication slave, which became the defaults in MySQL 8.0, the internal state of the data engine transaction is changed following a filtered XA transaction, and can become inconsistent with the replication transaction context state.

Due to these issues, the use of replication filters or binary log filters in combination with XA transactions is not supported. This fix adds the new error ER_XA_REPLICATION_FILTERS, which is logged whenever an XA transaction is impacted by a replication filter, whether or not the transaction was empty as a result. If the transaction is not empty, the replication slave is able to continue running, but you should take steps to discontinue the use of replication filters with XA transactions in order to avoid potential issues. If the transaction is empty, the replication slave stops. In that event, the replication slave might be in an undetermined state in which the consistency of the replication process might be compromised. In particular, the gtid_executed set on a slave of the slave might be inconsistent with that on the master. To resolve this situation, isolate the master and stop all replication, then check GTID consistency across the replication topology. Undo the XA tran
saction that generated the error message, then restart replication.

Replication: When a transaction larger than the binary log transaction cache size (binlog_cache_size) was flushed to a temporary file during processing, and the flush failed due to a lack of space in the temporary directory, the flush error was not handled correctly. No message was written to the error log, and the binary log cache was not cleared after the transaction was rolled back. Now, in this situation, the server takes an appropriate action based on the binlog_error_action setting (shut down the server or halt logging), and writes a message to the error log. When the transaction is rolled back, the server checks for flush errors and clears the binary log cache if any occurred.

Replication: Using an IP address or hostname in any Group Replication related configuration on macOS was failing.

Replication: When GTIDs are in use for replication, replicated transactions that are filtered out on the slave are persisted. If binary logging is enabled on the slave, the filtered-out transaction is written to the binary log as a Gtid_log_event followed by an empty transaction containing only BEGIN and COMMIT statements. If binary logging is disabled, the GTID of the filtered-out transaction is written to the mysql.gtid_executed table. This process ensures that there are no gaps in the set of executed GTIDs, and that the filtered-out transactions are not retrieved again if the slave reconnects to the master. Previously, this process was not done for CREATE DATABASE, ALTER DATABASE, and DROP DATABASE statements, but it is now carried out for those statements as well as for others.

Replication: On a multithreaded slave, when a STOP SLAVE statement is executed on the slave, followed by a START SLAVE statement, the error log can report a different position in the binary log for the slave SQL thread when exiting, compared to the position reported for the slave SQL thread at the subsequent initialization.

For a multithreaded slave, the position reported for the SQL thread on exit is a low water mark, up to which the replication stream is consistent and has no gaps. Transactions appearing before the position are guaranteed to have committed, but transactions after the position may have committed or not. However, this low water mark was being reported before the process to stop the worker threads was actually carried out, and the low water mark was subsequently updated by a checkpoint routine during that process. The timing of the log message has now been changed so that the final low water mark is reported as the position for the SQL thread on exit.

Replication: In certain situations, such as during distributed recovery procedure, the certification info garbage collection was purging more data than it should, resulting in conflicts not being detected. The garbage collection procedure has been improved to take this case in consideration.

Replication: When the group_replication_applier channel's applier thread encountered an error, the master_log_name and end_log_pos in the error message were incorrect. In Group Replication, the events of a transaction are replicated before they are written to the binary log of the member where the transaction originated. The result is that the final master_log_name and the end_log_pos of those events are unknown at the time they are applied on the replica by group_replication_applier channel's applier thread. To avoid confusion, now any such error messages encountered by a group_replication_applier channel do not contain the binary log name and the binary log position.

A heap overflow vulnerability in the MySQL client library was fixed.

For generated columns that used the INTERVAL() function, incorrect behavior could occur.

The exec_in_background command for mysqltest is now available in MySQL 5.7.

An unencrypted connection could result from a client connection attempt specifying that an encrypted connection was required, if the server was not configured to support SSL.

A BETWEEN clause comparing negative values could lead to erroneous results.

Audit log filter rules did not permit class names to be specified as an array of strings.

(adam)

boost: updated to 1.68.0

1.68.0:
New Libraries
YAP:
An expression template library for C++14 and later, from Zach Laine.

Updated Libraries
Beast:
This version fixes a missing executor work guard in all composed operations used in the implementation. Users who are experiencing crashes related to asynchronous completion handlers are encouraged to upgrade.
For a complete list of changes, please view the official Release Notes.
Context:
78: Linker error with context-impl=ucontext: multiple definition of boost::context::detail::current_rec
Coroutine2:
14: crash while stack unwinding
18: some guidance on why to use coroutine2
20: failes to compile test in due to the error C2039
22: undefined behaviour documentation
Fiber:
170: buffered_channel::try_push has incorrect documentation
172: WIN64: 'invalid conversion from 'HANDLE' to 'std::thread::native_handle_type' errors for builds using mingw-w64
175: throw exception "Operation not permitted"
NUMA support moved to extra library (fiber-numa); enabled via property numa (numa=on)
Fusion:
Added a workaround for ambiguous call of fusion::deque constructor on GCC 4.4/c++0x
Fixed a bug with C-style array
Fixed a fusion::for_each signature to take functor by value
This may break existing code with non-copyable (non-movable) functor, but the implementation wasn't correct for a long time; the documantation was correct from the first. So, please make sure your usage.
Fixed unintentional MPL placeholder substituion bug on fusion::transform.
Moved description, how to conform Fusion sequence as MPL sequence, to mpl section.
Added notes regarding IO facility for adapted type.
Geometry:
Improvements
469 Horizontal grid shift (nadgrids) in SRS transformations (undocumented for now due to potential interface changes).
478 Box-Segment distance for spherical and geographic coordinate systems.
487 Missing combinations of geometries in distance for spherical and geographic cs
489 Additional direct geodesic problem formulas for internal use
Solved issues
470 Assertion failure with short spherical and geographic segments.
471 Invalid envelope of spherical polygon
498 Unexpected results with expand() and make_inverse() for spherical and geographic CS
Bugfixes
488 Handle non-true-references in closing_iterator and ever_circling_range_iterator
495 VxWorks 7 cross-compilation issue.
GIL:
The I/O extensions have been entirely rewritten.
The library now requires a C++11-compliant compiler.
Documentation has been reformatted and updated.
Graph:
Conditionally replace deprecated/removed C++98 std::bind1st by std::bind, std::auto_ptr by std::unique_ptr, and std::random_shuffle by std::shuffle.
Fix compiler error with release builds on VS2015
Fix the Stanford GraphBase examples
Fix friend declarations for iterator_core_access
Add missing <boost/iterator.hpp> include
Avoid an unused variable warning
Fix some typos in documentation
Fix some issues in tests and examples
Lexical Cast:
Fixes for clang-tidy warnings 12092
Log:
Improved support for VxWorks.
Save and restore ebx register on x86 PIE targets in the dump stream manipulator implementation.
Math:
Support for arbitrary precision complex valued quadrature and hence contour integration
Improve performance of polynomial addition
Math:
Continue to improve numerical integration routines, and in particular add support for contour integrals.
Improve accuracy of erfc function's rational approximations.
Multi-index Containers:
Containers of moveable but non-copyable elements can now be serialized (ticket 13478). Thanks to S辿bastien Paris for the report.
multi_index_container's default constructor is no longer explicit (ticket 13518).
Multiprecision:
Support for multiprecision complex numbers
Multiprecision:
Support added for complex multiprecision numbers.
Changed conversion to unsigned integer types to be truncating to match standard defined behaviour.
Correct bug in MPFR string formatting.
Fix undefined behaviour in cpp_dec_float conversion from long long.
Add support for Eigen interoperability.
float128.hpp: Fix Intel on Windows build.
Fix type used in temporaries when expanding expression templates containing mixed expressions.
Fix infinite loop in gmp_float to fixed-point string conversion.
Update the expression templates to issue static_asserts with better error messages when you try and do something unsupported.
Fix bug in cpp_int where incrementing to zero doesn't set the sign correctly.
Remove erroneous use of std::move, and rely on NVRO instead.
Fix up support for changes to MPIR-3.0.
Fix various conversion errors in cpp_bin_float when the exponent type is a long long, or else we're converting to an integer that is wider than we are.
Fix compatibility issue with GCC-8 caused by the introduction of std::byte.
Optional:
Added member function has_value() for compatibility with std::optional (issue 52).
Added member function map() for transforming optional<T> into optional<U> using a function of type T -> U.
Added member function flat_map() for transforming optional<T> into optional<U> using a function of type T -> optonal<U>.
Predef:
Add support for __ARM_ARCH macro.
Add detection for PTX architecture.
Add nvcc compiler detection.
Add support for detecting CUDA.
Remove reference to obsolete BOOST_ARCH_AMD64.
Program Options:
Support for multiple long names for an option, thanks to Eyal Rozenberg
Python:
Bug fixes to correct autolink support (Windows)
Rational:
Fixed undefined behavior in normalize()
System:
Add constexpr to error_code and error_condition members under C++14 and above
Signals:
Removal Notice: Boost.Signals will be removed in the next release. Boost.Signals was deprecated in version 1.54.0. Transition to Boost.Signals2 now to avoid disruption.
Stacktrace:
Fixed compilation on Solaris and other platforms that do qualify address as const in dladdr function (github 54).
Dropped dependency on Boost.LexicalCast.
Test:
Boost.test v3.8 see the Changes log for more details.
Breaking changes
The master_test_suite_t object is no more copyable
New feature:
Dataset test case can now use command line parameters
TypeIndex:
Dropped dependency on Boost.MPL.
Uuid:
Breaking change: sha1 detail namespace header redirection for backwards compatibility was removed
Added support for std::hash
Added support for move semantics on random generators
Properly handle EINTR when acquiring entropy
Use getrandom(2) instead of getentropy(3) on linux

(adam)

hwloc: fix buildlink3 on Darwin

(adam)

hwloc: fix building on Darwin

(adam)

Updated finance/py-ofxparse, textproc/py-m2r

(adam)

py-m2r: updated to 0.2.0

Version 0.2.0:
Add start-line and end-line option to mdinclude directive
Add anonymous_references option

(adam)

py-ofxparse: updated to 0.19

0.19:
Bug fixes.

(adam)

Updated textproc/py-xlsxwriter, devel/py-pathspec

(adam)

py-pathspec: updated to 0.5.7

0.5.7:
- Fix collections deprecation warning.

(adam)

py-xlsxwriter: updated to 1.0.7

Release 1.0.7:
Fix for unicode type error in Python 3.

Release 1.0.6:
Added some performance improvements.

(adam)

Updated devel/libatomic_ops, emulators/qemu

(adam)

qemu: updated to 3.0.0

3.0.0:

Incompatible changes
Option "-no-kvm-pit-reinjection" has been removed. Use "-global kvm-pit.lost_tick_policy=discard" instead.
Option "-no-kvm-irqchip" has been removed. Use "-machine kernel_irqchip=off" instead.
The "vlan" parameter of the "-net" option has been removed. Use "-netdev" with IDs or the "hubport" netdev instead.
The deprecated -drive options "geometry", "addr" and "serial" have been removed. Use "-drive if=none" and pass the options to "-device" instead.
The "xlnx-ep108" machine has been replaced by the "xlnx-zcu102" machine.
The qemu-img "convert -s snapshot_id_or_name" argument has been removed. Use "convert -l snapshot_param" instead.
PPC support for explicitly allocated RMAs has been removed (kernel support removed in 2015).
The default for VFIO's "display" option has been changed from "auto" to "off", because "auto" requires an OpenGL-compatible configuration for vGPUs that require GL support.
The deprecated xlnx-ep108 machine was removed.
Some 32-bit guests on the "virt" machine may need "-machine highmem=off" if they are unable to support LPAE and are confused by the PCI ECAM now being above the 4GB mark in the address space

New deprecated options and features
The "arch" output member of the "query-cpus-fast" command is deprecated. Use the "target" output member instead.
Option "-virtioconsole" is deprecated, use "-device virtconsole" instead.
The non-functional "-clock" option will be removed soon.
The "pc-0.10" and "pc-0.11" machines are now deprecated. Use newer machine types instead.
Option "-enable-hax" is deprecated. Use "-accel hax" instead.
The "arch" memeber in the reply to query-cpus-fast is deprecated, use "target" instead.
The "file" driver for drives is no longer appropriate for character or host devices and will only accept regular files (S_IFREG). Use host_cdrom or host_device instead.
Consult the "Deprecated Features" appendix for the full list of historically deprecated features/options.

Future incompatible changes
Three options are using different names on the command line and in configuration file. In particular:
The "acpi" configuration file section matches command-line option "acpitable";
The "boot-opts" configuration file section matches command-line option "boot";
The "smp-opts" configuration file section matches command-line option "smp".
-readconfig will standardize on the name for the command line option.
Behavior of automatic calculation of SMP topology when some SMP topology options for -smp are omitted (sockets, cores, threads) will change in the future. If guest ABI needs to be preserved on upgrades while using the SMP topology options, users should either set set all options explicitly (sockets, cores, threads), or omit all of them.
Devices "allwinner-a10", "pc87312", "ssi-sd" will be configured with explicit properties instead of implicitly. This is unlikely to affect users.
For x86, specifying a CPUID feature with both "+feature/-feature" and "feature=on/off" will cause a warning. The current behavior for this combination ("+feature/-feature" wins over "feature=on/off") will be changed so that "+feature" and "-feature" will be synonyms for "feature=on" and "feature=off" respectively).
The read-only block drivers "bochs", "cloop" and "dmg" as well as "rbd" and "vvfat" in certain read-only configurations will no longer enable read-only mode automatically. It will be necessary to specify "read-only=on" explicitly on the command line and in QMP commands for the setup to keep working; the default "read-only=off" setting will result in an error.
On s390x, using KVM with a Linux host kernel version < 3.15 has been broken since QEMU version 2.10. This will not be fixed unless a need is communicated (otherwise the code will be removed in the near future, so that you need at least Linux kernel version 3.15 on the host to run KVM on System z)

(adam)

libatomic_ops: updated to 7.6.6

7.6.6:
COPYING: sync with FSF's gpl-2.0.txt
Fix 'undefined reference to __atomic_load/store/cas_16' error (gcc-7/x64)
Fix a typo in the overview section of README
Fix comments style in configure.ac and Makefile.am
Update copyright information in README and some header files

(adam)

Updated devel/py-pyobjc

(adam)

py-pyobjc: updated to 4.2.2

Version 4.2.2:
Update metadata for Xcode 9.4
The binary release now includes wheels for both variants for the Python.org installer for python 3.6 and 3.7: 32- and 64-bit for macOS 10.6 or later, and 64-bit only for macOS 10.9 or later.
Ensure the context manager for NSAnimationContext defined in PyObjCTools.AppCategories actually works.
Fix convenience wrappers for Foundation.NSCache.
Fix convenience wrappers for Foundation.NSHashTable.

(adam)

Updated net/py-scp, devel/py-editorconfig-core

(adam)

py-editorconfig-core: updated to 0.12.2

0.12.2:
Unknown changes.

(adam)

py-scp: updated to 0.11.0

0.11.0:
Add putfo() method, allowing one to upload a file-like object
Add top-level get() and put() functions for convenience
Increase default socket time from 5 to 10 seconds

(adam)

Updated security/py-cryptography_vectors, security/py-cryptography

(adam)

py-cryptography[_vectors]: updated to 2.3.1

2.3.1:
Updated Windows, macOS, and manylinux1 wheels to be compiled with OpenSSL 1.1.0i.

(adam)

Updated devel/py-joblib

(adam)

py-joblib: updated to 0.12.2

Release 0.12.2:
Integrate loky 2.2.0 to fix regression with unpicklable arguments and functions reported by users.
Loky 2.2.0 also provides a protection against memory leaks long running applications when psutil is installed.
Joblib now includes the code for the dask backend which has been updated to properly handle nested parallelism and data scattering at the same time.
Restored some private API attribute and arguments (MemorizedResult.argument_hash and BatchedCalls.__init__'s pickle_cache) for backward compat.
Fix a deprecation warning message (for Memory's cachedir).

Release 0.12.1:
Make sure that any exception triggered when serializing jobs in the queue will be wrapped as a PicklingError as in past versions of joblib.
Fix kwonlydefaults key error in filter_args

(adam)

Updated databases/py-barman, devel/py-ruamel-yaml

(adam)

postgresql10-server: also build pgoutput plugin; bump revision

(adam)

py-ruamel-yaml: updated to 0.15.54

[0, 15, 54]:
  - fix issue where a comment could pop-up twice in the output
  - fix issue where JSON object (mapping) without spaces was not parsed
    properly
  - fix issue where comments after empty flow-style mappings were not emitted

[0, 15, 53]:
  - fix issue with flow style mapping with comments gobbled newline
  - fix issue where single '+' under YAML 1.2 was interpreted as
    integer, erroring out

[0, 15, 52]:
  - added .copy() mapping representation for round-tripping
    (CommentedMap) to fix incomplete copies of merged mappings
  - Also unmade that class a subclass of ordereddict to solve incorrect behaviour
    for {**merged-mapping} and dict(**merged-mapping)

[0, 15, 51]:
  - Fix method name dumps (were not dotted) and loads (reported by Douglas Raillard
  - Fix spurious trailing white-space caused when the comment start
    column was no longer reached and there was no actual EOL comment
    (e.g. following empty line) and doing substitutions, or when
    quotes around scalars got dropped.

[0, 15, 50]:
  - Allow YAML() as a context manager for output, thereby making it much easier
    to generate multi-documents in a stream.
  - Fix issue with incorrect type information for load() and dump()

(adam)

py-barman: updated to 2.4

Version 2.4:
-  Add standard and retry hook scripts for backup deletion (pre/post)
-  Add standard and retry hook scripts for recovery (pre/post)
-  Add standard and retry hook scripts for WAL deletion (pre/post)
-  Add --standby-mode option to barman recover to add standby_mode = on
    in pre-generated recovery.conf
-  Add --target-action option to barman recover, allowing users to add
    shutdown, pause or promote to the pre-generated recovery.conf file
-  Improve usability of point-in-time recovery with consistency checks
    (e.g. recovery time is after end time of backup)
-  Minor documentation improvements
-  Drop support for Python 3.3

Relevant bug fixes:
-  Fix remote get_file_content method, preventing
    incremental recovery from happening
-  Unicode issues with command
-  Add --wal-method=none when pg_basebackup >= 10

Minor bug fixes:
-  Stop process manager module from ovewriting lock files content
-  Relax the rules for rsync output parsing
-  Ignore vanished files in streaming directory
-  Case insensitive slot names
-  Make DataTransferFailure.from_command_error() more resilient
-  Rename command() to barman_command()
-  Initialise synchronous standby names list if not set
-  Correct placeholders ordering
-  Force datestyle to iso for replication connections
-  Returns error if delete command does not remove the backup
-  Fix exception when calling is_power_of_two(None)
-  Downgraded sync standby names messages to debug

Version 2.3:
-  Add support to PostgreSQL 10
-  Follow naming changes in PostgreSQL 10:
    -  The switch-xlog command has been renamed to switch-wal.
    -  In commands output, the xlog word has been changed to WAL and
        location has been changed to LSN when appropriate.
-  Add the --network-compression/--no-network-compression options to
    barman recover to enable or disable network compression at run-time
-  Add --target-immediate option to recover command, in order to exit
    recovery when a consistent state is reached (end of the backup,
    available from PostgreSQL 9.4)
-  Show cluster state (master or standby) with barman status command
-  Documentation improvements
-  Bug fixes:
    -  Fix high memory usage with parallel_jobs > 1
    -  Better handling of errors using parallel copy
    -  Make barman diagnose more robust with system exceptions
    -  Let archive-wal ignore files with .tmp extension

Version 2.2:
-  Implement parallel copy for backup/recovery through the
    parallel_jobs global/server option to be overridden by the --jobs or
    -j runtime option for the backup and recover command. Parallel
    backup is available only for the rsync copy method. By default, it
    is set to 1 (for behaviour compatibility with previous versions).
-  Support custom WAL size for PostgreSQL 8.4 and newer. At backup
    time, Barman retrieves from PostgreSQL wal_segment_size and
    wal_block_size values and computes the necessary calculations.
-  Improve check command to ensure that incoming directory is empty
    when archiver=off, and streaming directory is empty when
    streaming_archiver=off
-  Add external_configuration to backup_options so that users can
    instruct Barman to ignore backup of configuration files when they
    are not inside PGDATA (default for Debian/Ubuntu installations). In
    this case, Barman does not display a warning anymore.
-  Add --get-wal and --no-get-wal options to barman recover
-  Add max_incoming_wals_queue global/server option for the check
    command so that a non blocking error is returned in case incoming
    WAL directories for both archiver and the streaming_archiver contain
    more files than the specified value.
-  Documentation improvements
-  File format changes:
    -  The format of backup.info file has changed. For this reason a
        backup taken with Barman 2.2 cannot be read by a previous
        version of Barman. But, backups taken by previous versions can
        be read by Barman 2.2.
-  Minor bug fixes:
    -  Allow replication-status to work against a standby
    -  Close any PostgreSQL connection before starting pg_basebackup
    -  Safely handle paths containing special characters
    -  Archive .partial files after promotion of streaming source
    -  Recursively create directories during recovery
    -  Improve xlog.db locking
    -  Remove tablespace_map file during recover
    -  Reconnect to PostgreSQL if connection drops

Version 2.1:
-  Add --archive and --archive-timeout options to switch-xlog command
-  Preliminary support for PostgreSQL 10
-  Minor additions:
    -  Add last archived WAL info to diagnose output
    -  Add start time and execution time to the output of delete
        command
-  Minor bug fixes:
    -  Return failure for get-wal command on inactive server
    -  Make streaming_archiver_names and streaming_backup_name options
        global
    -  Fix rsync failures due to files truncated during transfer
    -  Correctly handle compressed history files
    -  Avoid de-referencing symlinks in pg_tblspc when preparing
        recovery
    -  Fix comparison of last archiving failure
    -  Avoid failing recovery if postgresql.conf is not writable
    -  Fix output of replication-status command
    -  Exclude files from backups like pg_basebackup
    -  Exclude directories from other Postgres versions while copying
        tablespaces
    -  Make retry hook script options global

Version 2.0 - 27 Sep 2016

-  Support for pg_basebackup and base backups over the PostgreSQL
    streaming replication protocol with backup_method=postgres
    (PostgreSQL 9.1 or higher required)

-  Support for physical replication slots through the slot_name
    configuration option as well as the --create-slot and --drop-slot
    options for the receive-wal command (PostgreSQL 9.4 or higher
    required). When slot_name is specified and streaming_archiver is
    enabled, receive-wal transparently integrates with pg_receivexlog,
    and check makes sure that slots exist and are actively used

-  Support for the new backup API introduced in PostgreSQL 9.6, which
    transparently enables concurrent backups and backups from standby
    servers using the standard rsync method of backup. Concurrent backup
    was only possible for PostgreSQL 9.2 to 9.5 versions through the
    pgespresso extension. The new backup API will make pgespresso
    redundant in the future

-  If properly configured, Barman can function as a synchronous standby
    in terms of WAL streaming. By properly setting the
    streaming_archiver_name in the synchronous_standby_names priority
    list on the master, and enabling replication slot support, the
    receive-wal command can now be part of a PostgreSQL synchronous
    replication cluster, bringing RPO=0 (PostgreSQL 9.5.5 or
    higher required)

-  Introduce barman-wal-restore, a standard and robust script written
    in Python that can be used as restore_command in recovery.conf files
    of any standby server of a cluster. It supports remote parallel
    fetching of WAL files by efficiently invoking get-wal through SSH.
    Currently available as a separate project called barman-cli. The
    barman-cli package is required for remote recovery when get-wal is
    listed in recovery_options

-  Control the maximum execution time of the check command through the
    check_timeout global/server configuration option (30 seconds
    by default)

-  Limit the number of WAL segments that are processed by an
    archive-wal run, through the archiver_batch_size and
    streaming_archiver_batch_size global/server options which control
    archiving of WAL segments coming from, respectively, the standard
    archiver and receive-wal

-  Removed locking of the XLOG database during check operations

-  The show-backup command is now aware of timelines and properly
    displays which timelines can be used as recovery targets for a given
    base backup. Internally, Barman is now capable of parsing .history
    files

-  Improved the logic behind the retry mechanism when copy operations
    experience problems. This involves backup (rsync and postgres) as
    well as remote recovery (rsync)

-  Code refactoring involving remote command and physical copy
    interfaces

-  Bug fixes:
    -  Correctly handle .history files from streaming
    -  Fix replication-status on PostgreSQL 9.1
    -  Fix replication-status when sent and write locations are not
        available
    -  Fix misleading message on pg_receivexlog termination

(adam)

pytest-cov: minor polishing

(adam)

Updated databases/repmgr, math/py-numexpr

(adam)

py-numexpr: updated to 2.6.7

Changes from 2.6.6 to 2.6.7
* Thanks to Lehman Garrison for finding and fixing a bug that exhibited memory leak-like behavior. The use in numexpr.evaluate of sys._getframe combined with .f_locals from that frame object results an extra refcount on objects in the frame that calls numexpr.evaluate, and not evaluate窶冱 frame. So if the calling frame remains in scope for a long time (such as a procedural script where numexpr is called from the base frame) garbage collection would never occur.
* Imports for the numexpr.test submodule were made lazy in the numexpr module.

(adam)

repmgr: updated to 4.1.0

4.1.0:
Configuration file changes
* Default for log_level is now INFO. This produces additional informative log output, without creating excessive additional log file volume, and matches the setting assumed for examples in the documentation.
* recovery_min_apply_delay now accepts a minimum value of zero

repmgr enhancements
* repmgr: always exit with an error if an unrecognised command line option is provided. This matches the behaviour of other PostgreSQL utilities such as psql.
* repmgr: add -q/--quiet option to suppress non-error output.
* repmgr cluster show, repmgr node check and repmgr node status return non-zero exit code if node status issues detected.
* Add --csv output option for repmgr cluster event.
* repmgr witness unregister can be run on any node, by providing the ID of the witness node with --node-id.
* repmgr standby switchover will refuse to run if an exclusive backup is taking place on the current primary.

repmgrd enhancements
repmgrd: create a PID file by default. For details, see repmgrd's PID file.
repmgrd: daemonize process by default. In case, for whatever reason, the user does not wish to daemonize the process, provide --daemonize=false.

Bug fixes
repmgr standby register --wait-sync: fix behaviour when no timeout provided.
repmgr cluster cleanup: add missing help options.
Ensure witness node follows new primary after switchover.
repmgr node check and repmgr node status: fix witness node handling.

(adam)

Updated databases/unixodbc, security/sudo

(adam)

sudo: updated to 1.8.23

Sudo 1.8.23

* PAM account management modules and BSD auto approval modules are
  now run even when no password is required.

* For kernel-based time stamps, if no terminal is present, fall
  back to parent-pid style time stamps.

* The new cvtsudoers utility replaces both the "sudoers2ldif" script
  and the "visudo -x" functionality.  It can read a file in either
  sudoers or LDIF format and produce JSON, LDIF or sudoers output.
  It is also possible to filter the generated output file by user,
  group or host name.

* The file, ldap and sss sudoers backends now share a common set
  of formatting functions for "sudo -l" output, which is also used
  by the cvtsudoers utility.

* The /run directory is now used in preference to /var/run if it
  exists.

* More accurate descriptions of the --with-rundir and --with-vardir
  configure options.

* The setpassent() and setgroupent() functions are now used on systems
  that support them to keep the passwd and group database open.
  Sudo performs a lot of passwd and group lookups so it can be
  beneficial to avoid opening and closing the files each time.

* The new case_insensitive_user and case_insensitive_group sudoers
  options can be used to control whether sudo does case-sensitive
  matching of users and groups in sudoers.  Case insensitive
  matching is now the default.

* Fixed a bug on some systems where sudo could hang on command
  exit when I/O logging was enabled.

* Fixed the build-time process start time test on Linux when the
  test is run from within a container.

* When determining which temporary directory to use, sudoedit now
  checks the directory for writability before using it.  Previously,
  sudoedit only performed an existence check.

* Sudo now includes an optional set of Monty Python-inspired insults.

* Fixed the execution of scripts with an associated digest (checksum)
  in sudoers on FreeBSD systems.  FreeBSD does not have a proper
  /dev/fd directory mounted by default and its fexecve(2) is not
  fully POSIX compliant when executing scripts.

* Chinese (Taiwan) translation for sudo from translationproject.org.

(adam)

unixodbc: updated to 2.3.7

2.3.7:
Fix for pkg-config file update on no linux platforms
Add W entry for GUI work
Various fixes for SQLBrowseConnect/W, SQLGetConnectAttr/W,and SQLSetConnectAttr/W
Fix buffer overflows in SQLConnect/W and refine behaviour of SQLGet/WritePrivateProfileString
SQLBrowseConnect/W allow disconnecting a started browse session after error
Add --with-stats-ftok-name configure option to allow the selection of a file name used to generate the IPC id when collecting stats. Default is the system odbc.ini file
Improve diag record handling with the behavior of Windows DM and export SQLCancelHandle
Bug fix when SQLGetPrivateProfileString() is called to get a list of sections or a list of keys
Connection pooling: Fix liveness check for Unicode drivers

(adam)

Updated www/py-django-admin-sortable2, devel/py-configargparse

(adam)

py-configargparse: updated to 0.13.0

0.13.0:
Unknown changes.

(adam)

py-django-admin-sortable2: updated to 0.6.21

0.6.21:
Unknown changes.

0.6.20:
Fix SortableAdminMixin to work in combination with other mixins like that from django-import-export.
Added jQuery compatibility layer for Django-2.1.

(adam)

Updated www/py-beautifulsoup4, devel/py-faker, devel/py-dash, math/py-simpleeval

(adam)

py-simpleeval: updated to 0.9.6

0.9.6:
Unknown changes.

(adam)

py-dash: updated to 4.7.3

v4.7.3:
Bug Fixes
Fix bug in spread where arguments were not being passed to wrapped function properly.

(adam)

py-faker: updated to 0.9.0

0.9.0:
.random_sample() now returns a list of unique elements instead of a set.
.random_sample_unique() is removed in favor of .random_sample().
Added random_choices(), random_elements() and random_letters().
Added faker.utils.distribution.choices_distribution_unique().
words(), password(), uri_path and pystr() now use the new the random_choices() method.

0.8.18:
Change blood group from 0 (zero) to O (capital letter O). Some locales do use 'zero', but O is more common and it is the medical standard.
Fix alpha-2 country code for Haiti.
Fix abbreviation for Nunavut.
Standardized postcode in address providers. Now all locales are guaranteed to have a postcode method and may have a localized alias for it (eg: zipcode).
Fix typo in pt_BR Person perovider.
Fix timezone handling.
Use tzinfo when provided in date_of_birth.

(adam)

py-beautifulsoup4: updated to 4.6.3

4.6.3:
* Exactly the same as 4.6.2. Re-released to make the README file
  render properly on PyPI.

4.6.2:
* Fix an exception when a custom formatter was asked to format a void
  element

(adam)

Updated emulators/qemu, www/py-scrapy

(adam)

qemu: updated to 2.12.1

2.12.1:
Bug fix release.

(adam)

py-scrapy: updated to 1.5.1

Scrapy 1.5.1:
This is a maintenance release with important bug fixes, but no new features:
* O(N^2) gzip decompression issue which affected Python 3 and PyPy
  is fixed
* skipping of TLS validation errors is improved
* Ctrl-C handling is fixed in Python 3.5+
* testing fixes
* documentation improvements

(adam)

Updated net/py-zmq

(adam)

py-zmq: updated to 17.1.2

17.1.2:
Fix possible hang when working with asyncio
Remove some outdated workarounds for old Cython versions
Fix some compilation with custom compilers
Remove unneeded link of libstdc++ on PyPy

(adam)

Updated security/py-cryptodome, devel/py-nose2

(adam)

py-nose2: updated to 0.8.0

0.8.0:
Fixed
- For junitxml plugin use test module in place of classname if no classname exists
Features
- Add code to enable plugins to documentation
Dropped support for python 3.3

(adam)

py-cryptodome: updated to 3.6.5

3.6.5:
Fixed incorrect AES encryption/decryption with AES acceleration on x86 due to gcc窶冱 optimization and strict aliasing rules.
More prime number candidates than necessary where discarded as composite due to the limited way D values were searched in the Lucas test.
Fixed ResouceWarnings and DeprecationWarnings.
Workaround for Python 3.7.0 bug on Windows

(adam)

lldb: add missing patch

(adam)

Updated devel/py-astroid, devel/py-blessings

(adam)

py-blessings: updated to 1.7

1.7:
Drop support for Python 2.6 and 3.3, which are end-of-lifed.
Switch from 2to3 to the six library.

(adam)

py-astroid: updated to 2.0.4

2.0.4:
Make sure that assign nodes can find ``yield`` statements in their values

2.0.3:
The environment markers for PyPy were invalid.

(adam)

Updated databases/postgresqlNN

(adam)

postgresql: updated to 10.5, 9.6.10, 9.5.14, 9.4.19, 9.3.24

The PostgreSQL Global Development Group has released an update to all supported versions of our database system, including 10.5, 9.6.10, 9.5.14, 9.4.19, 9.3.24. This release fixes two security issues as well as bugs reported over the last three months.

SECURITY ISSUES:
CVE-2018-10915: CERTAIN HOST CONNECTION PARAMETERS DEFEAT CLIENT-SIDE SECURITY DEFENSES
CVE-2018-10925: MEMORY DISCLOSURE AND MISSING AUTHORIZATION IN INSERT ... ON CONFLICT DO UPDATE

BUG FIXES AND IMPROVEMENTS
Several fixes related to VACUUM, including an issue that could lead to data corruption in certain system catalog tables
Several fixes for replaying write-ahead logs, including a case where a just-promoted standby server would not restart if it crashed before its first post-recovery checkpoint
Several performance improvements for replaying write-ahead logs
Several fixes for logical replication and logical decoding, including ensuring logical WAL senders are reporting the streaming state correctly
Allow replication slots to be dropped in single-user mode
Fix to have variance and similar aggregate functions return accurate results when executed using parallel query
Fix SQL-standard FETCH FIRST syntax to allow parameters ($n), as the standard expects
Fix to ensure that a process doing a parallel index scan will respond to signals, such as one to abort a query
Fix EXPLAIN's accounting for resource usage, particularly buffer accesses, in parallel workers
Several fixes for the query planner including improving the cost estimates for hash-joins and choosing to use indexes for mergejoins on composite type columns
Fix performance regression related to POSIX semaphores for multi-CPU systems running Linux or FreeBSD
Fix for GIN indexes that could lead to an assertion failure after a pg_upgrade from a version before PostgreSQL 9.4
Fix for SHOW ALL to display superuser configuration settings to roles that are allowed to read all settings
Fix issue where COPY FROM .. WITH HEADER would drop a line after every 4,294,967,296 lines processed
Several fixes for XML support, including using the document node as the context for XPath queries as defined in the SQL standard, which affects the xpath and xpath_exists functions, as well as XMLTABLE
Fix libpq for certain cases where hostaddr is used
Several ecpg fixes for Windows
Fix password prompting in Windows client programs so that echo is properly disabled
Several pg_dump fixes, including correctly outputting REPLICA IDENTITY properties for constraint indexes
Make pg_upgrade check that the old server was shut down cleanly

(adam)

Updated math/py-numpy, math/py-pandas

(adam)

py-pandas: updated to 0.23.4

v0.23.4:
This is a minor bug-fix release in the 0.23.x series and includes some regression fixes, bug fixes, and performance improvements. We recommend that all users upgrade to this version.

(adam)

py-numpy: updated to 1.15.0

NumPy 1.15.0 is a release with an unusual number of cleanups, many deprecations
of old functions, and improvements to many existing functions. Please read the
detailed descriptions below to see if you are affected.

For testing, we have switched to pytest as a replacement for the no longer
maintained nose framework. The old nose based interface remains for downstream
projects who may still be using it.

The Python versions supported by this release are 2.7, 3.4-3.7. The wheels are
linked with OpenBLAS v0.3.0, which should fix some of the linalg problems
reported for NumPy 1.14.

Highlights:
- NumPy has switched to pytest for testing.
- A new numpy.printoptions context manager.
- Many improvements to the histogram functions.
- Support for unicode field names in python 2.7.
- Improved support for PyPy.
- Fixes and improvements to numpy.einsum.

(adam)

Updated www/py-uwsgi, databases/py-ldap3

(adam)

py-ldap3: updated to 2.5.1

2.5.1:
connection.result is populated when exception raised with raise_exceptions=True
fixed objectSid in mocking strategies
fixed circular reference in exception history
added objectSid validator
byte values are properly searched in MOCK strategies
exception history refactored
connections in context manager don窶冲 bind anymore when auto_bind is set to AUTO_BIND_NONE
Cython compatible build
more detailed exception message in Mock strategy
exceptions flow refactored in reusable strategy
pwdlastset accept any positive integer
fixed an exception while logging packet with pyasn1 decoder
fixed importing abc from collections for Python 3.8

(adam)

py-uwsgi: updated to 2.0.17.1

Changes 2.0.17.1:
Fixed memory leak in HTTPS_CLIENT_CERTIFICATE
TLSv1 is now disabled by default (you can re-enable it with ssl-enable-tlsv1 at your own risk)
Improved daemons throttle system
Add ���secs��� log formatting variable
Improved snprintf() usage to be OpenBSD-friendly
Improved glibc crypt/crypt_r management
Fixed websocket pong timeout check
Add the ���License��� classifier to setup.py
Add support for php user.ini
Official support for Python 3.7

(adam)

fuse-emulator: updated to 1.5.6

Fuse 1.5.6 released.

* Emulation core improvements:
  * Z80 flags register is now correct after SCF and CCF

* Miscellaneous improvements:
  * Factor out common paths code between Linux and generic UNIX
  * More improvements disabling phantom typist after finishing loading
    TAP or standard ROM TZX files
  * Saving and loading binary data no longer increments tstate count
    or triggers breakpoints
  * "Variant" Alkatraz loaders (e.g. Gauntlet 3 and Shadow Dancer),
    "Variant" Search Loader programs (e.g. Lotus Esprit Turbo
    Challenge and Space Crusade) and Dinaload loaders (e.g. Astro Marine
    Corps) are now accelerated
  * Stop RZX playback/recording on machine reset/change
  * Various minor bugfixes.

(adam)

Updated devel/py-uvloop, devel/py-setuptools_scm, devel/cmake

(adam)

cmake: update to 3.12.1

Changes in 3.12.1 since 3.12.0:
- CSharp: Set CMAKE_CSharp_COMPILER_LOADED variable when language is enabled
- UseSWIG: Use CSharp language only if it is enabled
- Help: Add explicit <PackageName>_ROOT variable documentation
- Tests: Add case showing CMP0048 warning on injected project command
- project: Do not issue CMP0048 warnings on injected call
- CPack: Restore support for 0-valued version components
- FindCUDA/select_compute_arch: Restore two-component CUDA_VERSION
- GoogleTest: Ensure policy settings allow use of IN_LIST
- FindTCL: Add support for version 8.7
- FindPython*: fix erroneous behavior on multiple 'find_package' calls
- UseSWIG: restore legacy behavior for SWIG_MODULE_<name>_EXTRA_FLAGS

(adam)

py-setuptools_scm: updated to 3.1.0

v3.1.0:
fix - correct the invocation in version_from_scm and deprecate it as its exposed by accident
fix - handle git file listing on empty repositories
fix - deprecate ScmVersion.extra

(adam)

py-uvloop: updated to 0.11.2

v0.11.2:
Fix a memory leak related to contextvars support.

(adam)

Updated devel/creduce

(adam)

creduce: updated to 2.8.0

2.8.0:
No news is good news.

(adam)

Updated lang/llvm, lang/clang, lang/clang-static-analyzer, lang/libunwind, devel/lldb, devel/polly, devel/include-what-you-use, devel/py-llvmlite

(adam)

py-llvmlite: updated to 0.24.0

v0.24.0

This release adds support for Python 3.7 and fixes some build issues. It also
contains an updated SVML patch for the llvmdev package that works around some
vectorization issues. It also adds a selective LLVM 6.0.1 llvmdev build for the
`ppc64le` architecture.

* Fix up broken patch selector
* Add long description from readme
* LLVM 6.0.1 build based on RC and fixes for PPC64LE
* Recipe fixes for Conda Build 3
* Workaround for incorrect vectorization factor computed for SVML functions
* fix build on OpenBSD.
* Python 3.7 compat: Properly escape repl in re.sub

(adam)

include-what-you-use: updated to 0.10

0.10:
Use LLVM 6.0 for clang_6.0 branch

(adam)

llvm: updated to 6.0.1

6.0.1:

Non-comprehensive list of changes in this release

Support for retpolines was added to help mitigate 窶彙ranch target injection窶� (variant 2) of the 窶彜pectre窶� speculative side channels described by Project Zero and the Spectre paper.
The Redirects argument of llvm::sys::ExecuteAndWait and llvm::sys::ExecuteNoWait was changed to an ArrayRef of optional StringRef窶�s to make it safer and more convenient to use.
The backend name was added to the Target Registry to allow run-time information to be fed back into TableGen. Out-of-tree targets will need to add the name used in the def X : Target definition to the call to RegisterTarget.
The Debugify pass was added to opt to facilitate testing of debug info preservation. This pass attaches synthetic DILocations and DIVariables to the instructions in a Module. The CheckDebugify pass determines how much of the metadata is lost.
Significantly improved quality of CodeView debug info for Windows.
Preliminary support for Sanitizers and sibling features on X86(_64) NetBSD (ASan, UBsan, TSan, MSan, SafeStack, libFuzzer).

Changes to the LLVM IR
----------------------
The fast-math-flags (FMF) have been updated. Previously, the 窶惑ast窶� flag indicated that floating-point reassociation was allowed and all other flags were set too. The 窶惑ast窶� flag still exists, but there is a new flag called 窶腕eassoc窶� to indicate specifically that reassociation is allowed. A new bit called 窶和fn窶� was also added to selectively allow approximations for common mathlib functions like square-root. The new flags provide more flexibility to enable/disable specific floating-point optimizations. Making the optimizer respond appropriately to these flags is an ongoing effort.

Changes to the AArch64 Target
-----------------------------
Enabled the new GlobalISel instruction selection framework by default at -O0.

Changes to the ARM Target
-------------------------
Support for enabling SjLj exception handling on platforms where it isn窶冲 the default.

Changes to the Hexagon Target
-----------------------------
The Hexagon backend now supports V65 ISA.
The -mhvx option now takes an optional value that specifies the ISA version of the HVX coprocessor. The available values are v60, v62 and v65. By default, the value is set to be the same as the CPU version.
The compiler option -mhvx-double is deprecated and will be removed in the next release of the compiler. Programmers should use the -mhvx-length option to specify the desired vector length: -mhvx-length=64b for 64-byte vectors and -mhvx-length=128b for 128-byte vectors. While the current default vector length is 64 bytes, users should always specify the length explicitly, since the default value may change in the future.
The target feature hvx-double is deprecated and will be removed in the next release. LLVM IR generators should use target features hvx-length64b and hvx-length128b to indicate the vector length. The length should always be specified when HVX code generation is enabled.

Changes to the MIPS Target
--------------------------
Fixed numerous bugs:

fpowi on MIPS64 giving incorrect results when used with a negative integer.
Usage of the asm 窶歪窶� constraint with the wrong datatype causing an assert/crash.
Fixed a conversion bug when using the DSP ASE.
Fixed an inconsistency where objects were not marked as using the microMIPS as when the micromips function attribute or the 窶�.set micromips窶� directive was used.
Reordered the MIPSR6 specific hazard scheduler pass to after the delay slot filler, fixing a class of rare edge case bugs where the delay slot filler would violate ISA restrictions.
Fixed a crash when using a type of unknown size with gp relative addressing.
Corrected the j macro for microMIPS.
Corrected the encoding of movep for microMIPS32r6.
Fixed an issue with the usage of insert instructions having an invalid set of operands.
Fixed an issue where TLS symbols were not marked as such.
Enabled the usage of register scavenging with MSA, due to its shorter offsets for loads and stores.
Corrected the ELF headers when using the DSP ASE.

New features:

The long branch pass now generates some R6 specific instructions when targeting MIPSR6.
The delay slot filler now performs more branch conversions if delay slots cannot be filled.
The MIPS MT ASE is now fully supported.
Added support for the lapc pseudo instruction.
Improved the selection of multiple instructions (dext, nmadd, nmsub).
Further improved microMIPS codesize reduction.

Deprecation notices:

microMIPS64R6 support was been deprecated since 5.0, and has now been completely removed.

Changes to the SystemZ Target
-----------------------------
During this release the SystemZ target has:

Added support for 128-bit atomic operations.
Added support for the 窶徙窶� constraint for inline asm statements.

Changes to the X86 Target
-------------------------
During this release the X86 target has:

Added support for enabling SjLj exception handling on platforms where it isn窶冲 the default.
Added intrinsics for Intel Extensions: VAES, GFNI, VPCLMULQDQ, AVX512VBMI2, AVX512BITALG, AVX512VNNI.
Added support for Intel Icelake CPU.
Fixed some X87 codegen bugs.
Added instruction scheduling information for Intel Sandy Bridge, Ivy Bridge, Haswell, Broadwell, and Skylake CPUs.
Improved scheduler model for AMD Jaguar CPUs.
Improved llvm-mc窶冱 disassembler for some EVEX encoded instructions.
Add support for i8 and i16 vector signed/unsigned min/max horizontal reductions.
Improved codegen for memory comparisons
Improved codegen for i32 vector multiplies
Improved codegen for scalar integer absolute values
Improved codegen for vector integer rotations (XOP and AVX512)
Improved codegen of data being transferred between GPRs and K-registers.
Improved codegen for vector truncations.
Improved folding of address computations into gather/scatter instructions.
Gained initial support recognizing variable shuffles from vector element extracts and inserts.
Improved documentation for SSE/AVX intrinsics in intrin.h header files.
Gained support for emitting retpolines, including automatic insertion of the necessary thunks or using external thunks.

(adam)

Updated www/py-pylint-django, devel/py-setuptools_scm, time/py-pendulum, www/py-test-django, www/py-cheroot, www/py-cherrypy

(adam)

py-cherrypy: updated to 17.0.0

v17.0.0
CherryPy now allows namespace packages for its dependencies. Environments that cannot handle namespace packgaes like py2exe will need to add such support or pin to older CherryPy versions.

(adam)

py-cheroot: updated to 6.4.0

v6.4.0:
Factor out parts of :py:meth:cheroot.server.HTTPServer.start into :py:meth:prepare() <cheroot.server.HTTPServer.prepare> and :py:meth:serve() <cheroot.server.HTTPServer.serve>

(adam)

py-test-django: updated to 3.3.3

3.3.3:

Bug fixes
Fixed registration of ignore_template_errors() marker, which is required with pytest --strict
Fixed another regression with unittest

Docs
Use sphinx_rtf_theme
Minor fixes.

(adam)

py-pendulum: updated to 2.0.3

2.0.3:
Fixed handling of pytz timezones.
Fixed some formatter's tokens handling.
Fixed errors on some systems when retrieving timezone from localtime files.
Fixed diff methods.
Fixed closest()/farthest() methods.

(adam)

py-setuptools_scm: updated to 3.0.6

v3.0.6
fix 295 - correctly handle selfinstall from tarballs

v3.0.5
fix 292 - match leading 'V' character as well
https://www.python.org/dev/peps/pep-0440/#preceding-v-character

v3.0.4
rerelease of 3.0.3 after fixing the release process

v3.0.3 (pulled from pypi due to a packaging issue)
fix 286 - duo an oversight a helper functio nwas returning a generator instead of a list

v3.0.2
fix a regression from tag parsing - support for multi-dashed prefixes

v3.0.1
fix a regression in setuptools_scm.git.parse - reorder arguments so the positional invocation from before works as expected

v3.0.0
introduce pre-commit and use black
print the origin module to help testing
switch to src layout (breaking change)
no longer alias tag and parsed_version in order to support understanding a version parse failure
require parse results to be ScmVersion or None (breaking change)
fix 266 by requiring the prefix word to be a word again (breaking change as the bug allowed arbitrary prefixes while the original feature only allowed words")
introduce a internal config object to allow the configruation fo tag parsing and prefixes (thanks to @punkadiddle for introducing it and passing it trough)

(adam)

py-pylint-django: updated to 2.0

Version 2.0:
Requires pylint >= 2.0 which doesn窶冲 support Python 2 anymore!
Add modelform-uses-unicode check to flag dangerous use of the exclude attribute in ModelForm.Meta

(adam)

Updated devel/py-setuptools, devel/py-pip, devel/py-pbr

(adam)

py-pbr: updated to 4.2.0

4.2.0
* Deprecate 'test' integration
* Deprecate 'build\_sphinx' integration
* Add Sphinx extension
* doc: Add documentation for missing '[pbr]' options
* doc: Fix formatting of packagers guide
* Switch to stestr

(adam)

py-pip: updated to 18.0

18.0:

Process
- Switch to a Calendar based versioning scheme.
- Formally document our deprecation process as a minimum of 6 months of deprecation
  warnings.
- Adopt and document NEWS fragment writing style.
- Switch to releasing a new, non bug fix version of pip every 3 months.

Deprecations and Removals
- Remove the legacy format from pip list.
- Dropped support for Python 3.3.
- Remove support for cleaning up #egg fragment postfixes.
- Remove the shim for the old get-pip.py location.

  For the past 2 years, it's only been redirecting users to use the newer
  https://bootstrap.pypa.io/get-pip.py location.

Features
- Introduce a new --prefer-binary flag, to prefer older wheels over newer source packages.
- Improve autocompletion function on file name completion after options
  which have ``<file>``, ``<dir>`` or ``<path>`` as metavar.
- Add support for installing PEP 518 build dependencies from source.
- Improve status message when upgrade is skipped due to only-if-needed strategy.

Bug Fixes
- Update pip's self-check logic to not use a virtualenv specific file and honor cache-dir.
- Remove compiled pyo files for wheel packages.
- Speed up printing of newly installed package versions.
- Restrict install time dependency warnings to directly-dependant packages.

  Warning about the entire package set has resulted in users getting confused as
  to why pip is printing these warnings.
- Improve handling of PEP 518 build requirements: support environment markers and extras.
- Remove username/password from log message when using index with basic auth.
- Remove trailing os.sep from PATH directories to avoid false negatives.
- Fix "pip wheel pip" being blocked by the "don't use pip to modify itself" check.
- Disable pip's version check (and upgrade message) when installed by a different package manager.

  This works better with Linux distributions where pip's upgrade message may
  result in users running pip in a manner that modifies files that should be
  managed by the OS's package manager.
- Check for file existence and unlink first when clobbering existing files during a wheel install.
- Improve error message to be more specific when no files are found as listed in as listed in PKG-INFO.
- Always read ``pyproject.toml`` as UTF-8. This fixes Unicode handling on Windows and Python 2.
- Fix a crash that occurs when PATH not set, while generating script location warning.
- Disallow packages with ``pyproject.toml`` files that have an empty build-system table.

Vendored Libraries
- Update CacheControl to 0.12.5.
- Update certifi to 2018.4.16.
- Update distro to 1.3.0.
- Update idna to 2.7.
- Update ipaddress to 1.0.22.
- Update pkg_resources to 39.2.0 (via setuptools).
- Update progress to 1.4.
- Update pytoml to 0.1.16.
- Update requests to 2.19.1.
- Update urllib3 to 1.23.

Improved Documentation
- Document how to use pip with a proxy server.
- Document that the output of pip show is in RFC-compliant mail header format.

(adam)

py-setuptools: updated to 40.0.0

v40.0.0
Drop support for Python 3.3.
In package_index, fixed handling of encoded entities in URLs.
In pkg_resources VendorImporter, avoid removing packages imported from the root.
Minor doc fixes after actually using the new release process.
Removed section on non-package data files.
Fix developer's guide.
Fix PEP 518 configuration: set build requirements in pyproject.toml to ["wheel"].

(adam)

Updated devel/py-cython, www/py-django2

(adam)

py-django2: updated to 2.1

2.1:
Model 窶忻iew窶� permission

django.contrib.admin
ModelAdmin.search_fields now accepts any lookup such as field__exact.
jQuery is upgraded from version 2.2.3 to 3.3.1.
The new ModelAdmin.delete_queryset() method allows customizing the deletion process of the 窶彭elete selected objects窶� action.
You can now override the default admin site.
The new ModelAdmin.sortable_by attribute and ModelAdmin.get_sortable_by() method allow limiting the columns that can be sorted in the change list page.
The admin_order_field attribute for elements in ModelAdmin.list_display may now be a query expression.
The new ModelAdmin.get_deleted_objects() method allows customizing the deletion process of the delete view and the 窶彭elete selected窶� action.
The actions.html, change_list_results.html, date_hierarchy.html, pagination.html, prepopulated_fields_js.html, search_form.html, and submit_line.html templates can now be overridden per app or per model (besides overridden globally).
The admin change list and change form object tools can now be overridden per app, per model, or globally with change_list_object_tools.html and change_form_object_tools.html templates.
InlineModelAdmin.has_add_permission() is now passed the parent object as the second positional argument, obj.
Admin actions may now specify permissions to limit their availability to certain users.

django.contrib.auth
createsuperuser now gives a prompt to allow bypassing the AUTH_PASSWORD_VALIDATORS checks.
UserCreationForm and UserChangeForm no longer need to be rewritten for a custom user model.

django.contrib.gis
The new GEOSGeometry.buffer_with_style() method is a version of buffer() that allows customizing the style of the buffer.
OpenLayersWidget is now based on OpenLayers 4.6.5 (previously 3.20.1).

django.contrib.sessions
Added the SESSION_COOKIE_SAMESITE setting to set the SameSite cookie flag on session cookies.

Cache
The local-memory cache backend now uses a least-recently-used (LRU) culling strategy rather than a pseudo-random one.
The new touch() method of the low-level cache API updates the timeout of cache keys.

CSRF
Added the CSRF_COOKIE_SAMESITE setting to set the SameSite cookie flag on CSRF cookies.

Forms
The widget for ImageField now renders with the HTML attribute accept="image/*".

Internationalization
Added the get_supported_language_variant() function.
Untranslated strings for territorial language variants now use the translations of the generic language. For example, untranslated pt_BR strings use pt translations.

Management Commands
The new inspectdb --include-views option allows creating models for database views.
The BaseCommand class now uses a custom help formatter so that the standard options like --verbosity or --settings appear last in the help output, giving a more prominent position to subclassed command窶冱 options.

Migrations
Added support for serialization of functools.partialmethod objects.
To support frozen environments, migrations may be loaded from .pyc files.

Models
Models can now use __init_subclass__() from PEP 487.
A BinaryField may now be set to editable=True if you wish to include it in model forms.
A number of new text database functions are added: Chr, Left, LPad, LTrim, Ord, Repeat, Replace, Right, RPad, RTrim, and Trim.
The new TruncWeek function truncates DateField and DateTimeField to the Monday of a week.
Query expressions can now be negated using a minus sign.
QuerySet.order_by() and distinct(*fields) now support using field transforms.
BooleanField can now be null=True. This is encouraged instead of NullBooleanField, which will likely be deprecated in the future.
The new QuerySet.explain() method displays the database窶冱 execution plan of a queryset窶冱 query.
QuerySet.raw() now supports prefetch_related().

Requests and Responses
Added HttpRequest.get_full_path_info().
Added the samesite argument to HttpResponse.set_cookie() to allow setting the SameSite cookie flag.
The new as_attachment argument for FileResponse sets the Content-Disposition header to make the browser ask if the user wants to download the file. FileResponse also tries to set the Content-Type and Content-Length headers where appropriate.

Templates
The new json_script filter safely outputs a Python object as JSON, wrapped in a <script> tag, ready for use with JavaScript.

(adam)

py-paver: alternate bin/paver

(adam)

py-cython: updated to 0.28.5

0.28.5:

Bugs fixed

* The discouraged usage of GCC's attribute optimize("Os") was replaced by the
  similar attribute cold to reduce the code impact of the module init functions.

* A reference leak in Py2.x was fixed when comparing str to unicode for equality.

(adam)

Added devel/py-cogapp, lang/py-paver

(adam)

py-paver: added version 1.3.4

Paver is a Python-based build/distribution/deployment scripting tool along the
lines of Make or Rake. What makes Paver unique is its integration with commonly
used Python libraries. Common tasks that were easy before remain easy. More
importantly, dealing with your applications specific needs and requirements is
also easy.

(adam)

py-cogapp: added version 2.5.1

Cog is a file generation tool. It lets you use pieces of Python code as
generators in your source files to generate whatever text you need.

(adam)

Updated lang/python35, lang/python34

(adam)

python34: updated to 3.4.9

3.4.9:
Bug fix release for Python 3.4.

(adam)

python35: updated to 3.5.6

3.5.6:
Bug fix release for Python 3.5.

(adam)

Updated devel/py-iso3166, net/py-celery

(adam)

py-celery: updated to 4.2.1

4.2.1:
- **Result Backend**: Fix deserialization of exceptions that are present in the producer codebase but not in the consumer codebase.
- **Message Protocol Compatibility**: Fix error caused by an invalid (None) timelimit value in the message headers when migrating messages from 3.x to 4.x.
- **Result Backend**: Fix serialization of exception arguments when exception arguments are not JSON serializable by default.
- **Worker**: Fixed multiple issues with rate limited tasks
  Maintain scheduling order.
  Fix possible scheduling of a :class:celery.worker.request.Request with the wrong :class:kombu.utils.limits.TokenBucket which could cause tasks' rate limit to behave incorrectly.
  Fix possible duplicated execution of tasks that were rate limited or if ETA/Countdown was provided for them.
- **Worker**: Defensively handle invalid timelimit header values in requests.

(adam)

py-iso3166: updated to 0.9

0.9:
* Updated entries
  - New entry for Kosovo (XK / XKX)
  - Swaziland changed to Eswatini (effective 2018-07-16)

(adam)

Updated security/py-gssapi, devel/py-greenlet, textproc/py-html-sanitizer, textproc/py-regex

(adam)

py-regex: updated to 2018.07.11

2018.07.11:
Unknown changes.

(adam)

py-html-sanitizer: updated to 1.6.1

1.6:
Fixed another edge case where a tag which is allowed to be empty was erroneously removed if it contained not only whitespace but also a <br> tag.

1.5:
Fixed a few edge whitespace normalization edge cases and a bug where removing an empty tag removed all whitespace.
Added black for automatically formatting the Python code.
By default, links with target="_blank" get an additional rel="noopener" attribute (Article by Mathias Bynens). If you're overriding the list of allowed attributes for anchor tags you must add rel to your list.

(adam)

py-greenlet: updated to 0.4.14

0.4.14:
- Support for C-SKY architecture
- Fixed support for ppc64 ABI
- Fixed support for Python 3.7

(adam)

py-gssapi: updated to 1.5.1

1.5.1:
Remove warning about collections.abc usage for Python 3.7+

(adam)

Updated devel/py-txaio, www/py-django-mptt

(adam)

py-django-mptt: updated to 0.9.1

0.9.1:
Bug fixes.

(adam)

py-txaio: updated to 18.7.1

18.7.1
- move to calver
- deprecate Python 3.3 support and CI testing

(adam)

Updated archivers/py-lz4, graphics/py-altgraph, devel/py-macholib

(adam)

py-macholib: updated to 1.10

macholib 1.10
* Add support for LC_NOTE and LC_BUILD_VERSION

(adam)

py-altgraph: updated to 0.16.1

0.16.1:
Explicitly mark Python 3.7 as supported in wheel metadata.

0.16:
Add LICENSE file

(adam)

py-lz4: updated to 2.1.0

v2.1.0:
Enhance support for block decompression with unknown size of decompressed data

(adam)

Updated devel/py-serpent, textproc/py-vobject

(adam)

py-vobject: updated to 0.9.6.1

0.9.6.1:
Bug fixes.

(adam)

py-serpent: updated to 1.27

release 1.27:
fixed serialization of unicode elements in dicts and sets (this was a regression on Python 2.x)

release 1.26:
support for enum34 backport library for enums in older python versions.

(adam)

Updated textproc/py-sphinx-rtd-theme, textproc/py-sphinx, textproc/py-openpyxl

(adam)

py-openpyxl: updated to 2.5.5

2.5.5:

Bugfixes
Files with Mac epoch are read incorrectly
Cannot copy merged cells
Cannot access ws.active_cell

Pull Requests
Introduce read-support for images

(adam)

py-sphinx: do not depend on py-sphinx-rtd-theme, but the other way around

(adam)

py-sphinx-rtd-theme: updated to 0.4.1

v0.4.1:

Fixes
Line height adjustments for Liberation Mono

Other Changes
Add Sphinx as a dependency

(adam)

Updated devel/py-async_generator, devel/py-pylint

(adam)

py-pylint: updated to 2.1.1

What's New in Pylint 2.1.1?
* fix pylint crash due to misplaced-format-function not correctly handling class attribute.

(adam)

py-async_generator: updated to 1.10

Async_Generator 1.10:

Features
- Add support for PEP 525-style finalization hooks via
  set_asyncgen_hooks() and get_asyncgen_hooks() functions. On
  Python 3.6+, these are aliases for the versions in sys; on
  Python 3.5, they're work-alike implementations. And,
  @async_generator generators now call these hooks at the
  appropriate times.

Fixes
- Package now properly includes license files.

(adam)

Updated textproc/py-lxml, devel/py-uvloop, devel/waf, www/py-selenium

(adam)

py-selenium: updated to 3.14.0

Selenium 3.14.0
* Fix doc of URL-related ExpectedCondition
* Added ExpectedCondition invisibility_of_element
* Swap out httplib for urllib3
* Be consistent with webdriver init kwarg service_log_path

(adam)

waf: updated to 2.0.10

NEW IN WAF 2.0.10
* Add a task semaphore system
* Fix --help when no wscript is supplied
* Fix Fortran processing with generated Fortran files

(adam)

py-uvloop: updated to 0.11.1

v0.11.1:

Bug Fixes
Fix server to shutdown when alive connections exist
Fix a few bugs and crashes in UDP layer
Fix FD leakage if spawning a subprocess fails
Fix libuv process handles leak when uv_spawn() fails

(adam)

py-lxml: updated to 4.2.4

4.2.4:
Features added
* Allow using pkg-config for build configuration.

Bugs fixed
* Crash when moving an element to another document with
  Element.insert().

(adam)

Updated security/py-asn1, security/py-asn1-modules

(adam)

py-asn1-modules: updated to 0.2.2

Revision 0.2.2:
- Copyright notice extended to the year 2018
- Migrated references from SourceForge
- rfc2986 module added

(adam)

py-asn1: updated to 0.4.4

Revision 0.4.4:
- Fixed native encoder type map to include all ASN.1 types
  rather than just ambiguous ones
- Fixed crash in .prettyPrint of Sequence and Set occurring
  at OPTIONAL components

(adam)

Updated emulators/atari800, net/lftp

(adam)

lftp: updated to 4.8.4

Version 4.8.4:
* fixed a security vulnerability with "file:" file names.
* fixed mirror --flat.
* http: extract links from <source> tags.
* fixed upload of zero-length files over ftps.
* fixed assert on "mput -d".
* fixed a core dump.
* avoid multiple backup files of DHT cache.
* translations updated (uk, zh_CN).

(adam)

atari800: updated to 4.0.0

Version 4.0.0:

This release brings four years of hard work of fellow Atari800 developers
on improvements and bugfixes of our favorite Atari emulator.

Ports to Atari ST/TT/Falcon, Android, Raspberry Pi and Sega Dreamcast
have been improved (some of them greatly).

CPU, SIO, ANTIC, GTIA, POKEY and PIA emulation has been corrected.

Atari800 now includes Altirra BIOS so it is now possible to run *some*
programs even without installing the original ROM files.

The total number of changes is so huge that major version bump was necessary.

Contrary to usual NEWS entries here follows a very incomplete list. It might
be updated in later releases. In the meantime please read DOC/ChangeLog
and the commit history in git for complete list of changes.

This release contains breaking changes in color handling. Users updating from
an earlier version should reset their color settings, or else the display
might be unreadable. To reset the color settings, do one of these:
* Select one of the presets available in the menu option "Display settings"->
  "Color preset"; or
* Run atari800 with the -color-preset command-line option, e.g.:
  atari800 -colors-preset standard; or
* Delete the emulator's config file (.atari800.cfg). Caution: you'll lose
  all Atari800 settings!

New features:
-------------
* Support for loading of CAS images with "fsk" chunks - images of
  copy-protected tapes can now be loaded, with SIO patch being disabled.
* Bit3 Full View 80 Column card emulation.
* New cartridge type 68: "Atrax 128 KB cartridge".
  Previously existing cartridge type 17: "Atrax 128 KB cartridge" was based
  on a misconception - real Atrax cartridges have their address and data
  lines intermixed, so type 17 could not be used with actual ROM dumps. So,
  type 17 has been renamed to "Decoded Atrax 128 KB cartridge", and new type
  68 has been added, whcih now can be used for Atrax ROM chip dumps.
* New cartridge types supported:
  - aDawliah 32 KB cartridge
  - aDawliah 64 KB cartridge

* new command line switch "-volume" (for 16bit sound output) that can set
  the output volume of the Atari 800 emulator with value from 0 to 100.

Changes:
--------
* The Sound Settings option "Fragment size" has been renamed to less cryptic
  "Hardware buffer size".
* Total emulator volume is lower now because the output has been shifted
  in order to fix a possible annoying humming sound in silence on some
  receivers (TV sets).

Fixes:
------
* Fixed computation of gamma adjustment - now it is applied to each of the
  three RGB channels separately.
* On systems that support synchronized sound: Fixed the emulator crashing
  when Dual POKEY was enabled while High Fidelity POKEY was turned off.
* Improve screen update routines in the Dreamcast port. They don't use
  DIRTYRECT anymore but are faster than the old routines when the whole
  screen is dirty.

(adam)

Updated graphics/py-blockdiag, devel/py-dash

(adam)

py-dash: updated to 4.7.1

v4.7.1:
New Features
- Modify to_dict to first try to convert using dict() before falling back to using pydash.helpers.iterator().

v4.7.0:
Misc
- Internal code optimizations.

(adam)

py-blockdiag: updated to 1.5.4

1.5.4:
- Fix Python 3.7 compatibility

(adam)

Updated fonts/py-fonttools, devel/py-test-xdist, devel/py-test, devel/py-test-asyncio, devel/py-pluggy, devel/py-test-relaxed

(adam)

py-test-relaxed: updated to 1.1.4

1.1.4:
- :support:- backported Add missing universal wheel indicator in setup
  metadata.

1.1.3:
- :bug:- Fix the @raises helper decorator so it actually raises an
  exception when the requested exception is not raised by the decorated
  function. That's definitely not a confusing sentence.

(adam)

py-test: bump py-pluggy DEPENDS

(adam)

py-pluggy: updated to 0.7.1

pluggy 0.7.1:

Deprecations and Removals
- Deprecate the implprefix kwarg to PluginManager and instead
  expect users to start using explicit HookimplMarker everywhere.

Features
- Add .plugin member to PluginValidationError to access failing plugin during post-mortem.
- Add per implementation warnings support for hookspecs allowing for both
  deprecation and future warnings of legacy and (future) experimental hooks
  respectively.

Bug Fixes
- Fix a bug where _HookCaller.call_historic() would call the proc
  arg even when the default is None resulting in a TypeError.
- Fix problem when handling VersionConflict errors when loading setuptools plugins.

Improved Documentation
- Document how exceptions are handled and how the hook call loop
  terminates immediately on the first error which is then delivered
  to any surrounding wrappers.
- Docs rework including a much better introduction and comprehensive example
  set for new users. A big thanks goes out to @obestwalter for the great work!

Trivial/Internal Changes
- Break up the main monolithic package modules into separate modules by concern
- Automate setuptools wheels building and PyPi upload using TravisCI.
- Reorganize tests more appropriately by modules relating to each
  internal component/feature. This is in an effort to avoid (future)
  duplication and better separation of concerns in the test set.
- Add HookImpl.__repr__() for better debugging.
- Start using towncrier and a custom tox environment to prepare releases!

pluggy 0.7.0 (Unreleased)
* We discovered a deployment issue so this version was never released to PyPI, only the tag exists.

(adam)

py-test-asyncio: updated to 0.9.0

0.9.0:
Python 3.7 support.
Remove event_loop_process_pool fixture and pytest.mark.asyncio_process_pool marker.

(adam)

py-test: updated to 3.7.1

pytest 3.7.1:

Bug Fixes
- Raise immediately if approx() is given an expected value of a type it doesn't understand (e.g. strings, nested dicts, etc.).
- Correctly represent the dimensions of an numpy array when calling repr() on approx().
- Display the absolute path if cache_dir is not relative to the rootdir instead of failing.
- Fix compatibility problem with plugins and the warning code issued by fixture functions when they are called directly.
- Fix infinite recursion in pytest.approx with arrays in numpy<1.13.
- Pin pathlib2 to >=2.2.0 as we require __fspath__ support.
- Fix TypeError when the assertion message is bytes in python 3.

pytest 3.7.0:

Deprecations and Removals
- pytest_namespace has been deprecated.
  See the documentation for pytest_namespace hook for suggestions on how to deal
  with this in plugins which use this functionality.
- Calling a fixture function directly, as opposed to request them in a test function, now issues a RemovedInPytest4Warning. It will be changed into an error in pytest 4.0.
  This is a great source of confusion to new users, which will often call the fixture functions and request them from test functions interchangeably, which breaks the fixture resolution model.

Features
- New package fixture scope: fixtures are finalized when the last test of a *package* finishes. This feature is considered **experimental**, so use it sparingly.
- Node.add_marker now supports an append=True/False parameter to determine whether the mark comes last (default) or first.
- Fixture caplog now has a messages property, providing convenient access to the format-interpolated log messages without the extra data provided by the formatter/handler.
- New --trace option to enter the debugger at the start of a test.
- Introduce pytester.copy_example as helper to do acceptance tests against examples from the project.

Bug Fixes
- Fix a bug where fixtures overridden by direct parameters (for example parametrization) were being instantiated even if they were not being used by a test.
- Fix ApproxNumpy initialisation argument mixup, abs and rel tolerances were flipped causing strange comparsion results.
  Add tests to check abs and rel tolerances for np.array and test for expecting nan with np.array()
- Fix truncated locals output in verbose mode.

Improved Documentation
- Correct the usage documentation of --last-failed-no-failures by adding the missing --last-failed argument in the presented examples, because they are misleading and lead to think that the missing argument is not needed.

Trivial/Internal Changes
- Now a README.md file is created in .pytest_cache to make it clear why the directory exists.

(adam)

py-test-xdist: updated to 1.22.5

pytest-xdist 1.22.5:
Bug Fixes
- Revert change that dropped support for pytest<3.4 and require six.
  This change caused problems in some installations, and was a mistaken
  in the first place as we should not change version requirements
  in bug-fix releases unless they fix an actual bug.

pytest-xdist 1.22.4:
Bug Fixes
- Remove last references to obsolete py.code.
  Remove some unnecessary references to py.builtin.
- Workaround cpu detection on Travis CI.

(adam)

py-fonttools: updated to 3.29.0

3.29.0:
- [feaLib] In the OTL table builder, when the name table is excluded
  from the list of tables to be build, skip compiling featureNames blocks,
  as the records referenced in FeatureParams table don't exist.
- [otBase] Try ExtensionLookup if other offset-overflow methods fail.
- [feaLib] Added support for explicit subtable; break statements in
  PairPos lookups; previously these were ignored.
- [cffLib.specializer] Make sure the stack depth does not exceed maxstack - 1,
  so that a subroutinizer can insert subroutine calls.
- [otTables] Added support for fixing offset overflow errors occurring inside
  MarkBasePos subtables.
- [subset] Write the default output file extension based on --flavor option,
  or the value of TTFont.sfntVersion.
- [unicodedata] Updated Blocks, Scripts and ScriptExtensions for Unicode 11.
- [xmlWriter] Added context manager to XMLWriter class to autoclose file
  descriptor on exit.
- [psCharStrings] Optimize the charstring's bytecode by encoding as integers
  all float values that have no decimal portion.
- [ttFont] Fixed missing import of TTLibError exception.
- [feaLib] Allow any languages other than dflt under DFLT script.

(adam)

Updated devel/py-ipython5, textproc/json-c

(adam)

json-c: updated to 0.13.1

0.13.1:
* Bump the major version of the .so library generated up to 4.0 to avoid
  conflicts because some downstream packagers of json-c had already done
  their own bump to ".so.3" for a much older 0.12 release.
* Add const size_t json_c_object_sizeof()
* Avoid invalid free (and thus a segfault) when ref_count gets < 0
* fix handling of custom double formats that include a ".0"
* Avoid uninitialized variable warnings in json_object_object_foreach
* Issue 396: fix build for certain uClibc based systems.
* Add a top level fuzz directory for fuzzers run by OSS-Fuzz

0.13:
This release, being three and a half years after the 0.12 branch (f84d9c),
  has quite a number of changes included.  The following is a sampling of
  the most significant ones.

Since the 0.12 release, 250 issues and pull requests have been closed.
See issues_closed_for_0.13.md for a complete list.

Deprecated and removed features:
--------------------------------
* All internal use of bits.h has been eliminated.  The file will be removed.
        Do not use: hexdigit(), error_ptr(), error_descrition() and it_error()
* lh_abort() is deprecated.  It will be removed.

Behavior changes:
-----------------
* Tighten the number parsing algorithm to raise errors instead of truncating
    the results.  For example 12.3.4 or 2015-01-15, which now return null.
        See commit 99d8fc

* Use size_t for array length and size.  Platforms where sizeof(size_t) != sizeof(int) may not be backwards compatible
        See commits 45c56b, 92e9a5 and others.

* Check for failue when allocating memory, returning NULL and errno=ENOMEM.
        See commit 2149a04.

* Change json_object_object_add() return type from void to int, and will return -1 on failures, instead of exiting. (Note: this is not an ABI change)

New features:
-------------
* We're aiming to follow RFC 7159 now.

* Add a couple of additional option to json_object_to_json_string_ext:
        JSON_C_TO_STRING_PRETTY_TAB
        JSON_C_TO_STRING_NOSLASHESCAPE

* Add a json_object_object_add_ex() function to allow for performance
        improvements when certain constraints are known to be true.

* Make serialization format of doubles configurable, in two different ways:
        Call json_object_set_serializer with json_object_double_to_json_string and a custom
        format on each double object, or
        Call json_c_set_serialization_double_format() to set a global or thread-wide format.

* Add utility function for comparing json_objects - json_object_equal()

* Add a way to copy entire object trees: json_object_deep_copy()
* Add json_object_set_<type> function to modify the value of existing json_object's
without the need to recreate them.  Also add a json_object_int_inc function to
adjust an int's value.
* Add support for JSON pointer, RFC 6901.  See json_pointer.h
* Add a json_util_get_last_err() function to retrieve the string describing the
cause of errors, instead of printing to stderr.
* Add perllike hash function for strings, and json_global_set_string_hash() 8f8d03d
* Add a json_c_visit() function to provide a way to iterate over a tree of json-c objects.

Notable bug fixes and other improvements:
-----------------------------------------
* Make reference increment and decrement atomic to allow passing json objects between threads.
* Fix json_object_object_foreach to avoid uninitialized variable warnings.
* Improve performance by removing unneeded data items from hashtable code and reducing duplicate hash computation.
* Improve performance by storing small strings inside json_object
* Improve performance of json_object_to_json_string by removing variadic printf. commit 9ff0f49
* Issue 371: fix parsing of "-Infinity", and avoid needlessly copying the input when doing so.
* Fix stack buffer overflow in json_object_double_to_json_string_format() - commit 2c2deb87
* Fix various potential null ptr deref and int32 overflows
* Issue 332: fix a long-standing bug in array_list_put_idx() where it would attempt to free previously free'd entries due to not checking the current array length.
* Issue 195: use uselocale() instead of setlocale() in json_tokener to behave better in threaded environments.
* Issue 275: fix out of bounds read when handling unicode surrogate pairs.
* Ensure doubles that happen to be a whole number are emitted with ".0" - commit ca7a19
* for Visual Studio, use a snprintf/vsnprintf wrapper that ensures the string is terminated.
* Fix double to int cast overflow in json_object_get_int64.
* Clamp double to int32 when narrowing in json_object_get_int.
* Use strtoll() to parse ints - instead of sscanf
* Miscellaneous smaller changes, including removing unused variables, fixing warning
about uninitialized variables adding const qualifiers, reformatting code, etc...

(adam)

py-ipython5: updated to 5.8.0

IPython 5.8.0
* Update inspecting function/methods for future-proofing.

(adam)

Updated finance/py-braintree, www/py-beautifulsoup4, devel/py-test-flake8, www/py-MechanicalSoup

(adam)

py-MechanicalSoup: updated to 0.10.0

Version 0.10

Main changes:
Added StatefulBrowser.refresh() to reload the current page with the same request.
StatefulBrowser.follow_link, StatefulBrowser.submit_selected() and the new StatefulBrowser.download_link now sets the Referer: HTTP header to the page from which the link is followed.
Added method StatefulBrowser.download_link, which will download the contents of a link to a file without changing the state of the browser.
The selector argument of Browser.select_form can now be a bs4.element.Tag in addition to a CSS selector.
Browser.submit and StatefulBrowser.submit_selected accept a larger number of keyword arguments. Arguments are forwarded to requests.Session.request.

Internal changes:
StatefulBrowser.choose_submit will now ignore input elements that are missing a name-attribute instead of raising a KeyError.
Private methods Browser._build_request and Browser._prepare_request have been replaced by a single method Browser._request.

(adam)

py-test-flake8: updated to 1.0.2

1.0.2
- Test on Python 3.7
- Escape a regex tring with r""

(adam)

py-beautifulsoup4: updated to 4.6.1

4.6.1:

* Stop data loss when encountering an empty numeric entity, and
  possibly in other cases.

* Preserve XML namespaces introduced inside an XML document, not just
  the ones introduced at the top level.

* Added a new formatter, "html5", which represents void elements
  as "<element>" rather than "<element/>".

* Fixed a problem where the html.parser tree builder interpreted
  a string like "&foo " as the character entity "&foo;"

* Correctly handle invalid HTML numeric character entities
  which reference code points that are not Unicode code points. Note
  that this is only fixed when Beautiful Soup is used with the
  html.parser parser -- html5lib already worked and I couldn't fix it
  with lxml.

* Improved the warning given when no parser is specified.

* When markup contains duplicate elements, a select() call that
  includes multiple match clauses will match all relevant
  elements.

* Fixed code that was causing deprecation warnings in recent Python 3
  versions.

* Fixed a Windows crash in diagnose() when checking whether a long
  markup string is a filename.

* Stopped HTMLParser from raising an exception in very rare cases of
  bad markup.

* Fixed a bug where find_all() was not working when asked to find a
  tag with a namespaced name in an XML document that was parsed as
  HTML.

* You can get finer control over formatting by subclassing
  bs4.element.Formatter and passing a Formatter instance into (e.g.)
  encode().

* You can pass a dictionary of `attrs` into
  BeautifulSoup.new_tag. This makes it possible to create a tag with
  an attribute like 'name' that would otherwise be masked by another
  argument of new_tag.

* Clarified the deprecation warning when accessing tag.fooTag, to cover
  the possibility that you might really have been looking for a tag
  called 'fooTag'.

(adam)

py-braintree: updated to 3.47.0

3.47.0
Add processor respone code and processor response text to authorization adjustments subfield in transaction response.
Add support for Samsung Pay

(adam)

py-pylint: updated to 2.1.0

What's New in Pylint 2.1?

  * trailing-comma-tuple gets emitted for yield statements as well.

  * Get only the arguments of the scope function for redefined-argument-from-local

  * Add a check misplaced-format-function which is emitted if format function is used on
    non str object.

  * chain.from_iterable no longer emits dict-{}-not-iterating when dealing with dict values and keys

  * Demote the try-except-raise message from an error to a warning (E0705 -> W0706)

  * Correctly handle the new name of the Python implementation of the abc module.

    * Modules with __getattr__ are exempted by default from no-member

      There's no easy way to figure out if a module has a particular member when
      the said module uses __getattr__, which is a new addition to Python 3.7.
      Instead we assume the safe thing to do, in the same way we do for classes,
      and skip those modules from checking.

    * Fix a false positive invalid name message when method or attribute name is longer then 30 characters.

    * Include the type of the next branch in no-else-return

    * Fix inconsistent behaviour for bad-continuation on first line of file

    * Fix not being able to disable certain messages on the last line through
      the global disable option

    * Don't emit useless-return when we have a single statement that is the return itself

      We still want to be explicit when a function is supposed to return
      an optional value; even though pass could still work, it's not explicit
      enough and the function might look like it's missing an implementation.

  * Fix false-positive undefined-variable for self referential class name in lamdbas

    * Don't crash when pylint is unable to infer the value of an argument to next()

    * Don't emit not-an-iterable when dealing with async iterators.

      But do emit it when using the usual iteration protocol against
      async iterators.

  * Can specify a default docstring type for when the check cannot guess the type

(adam)

py-astroid: updated to 2.0.2

What's New in astroid 2.0.2?
  * Stop repeat inference attempt causing a RuntimeError in Python3.7

  *  infer_call_result can raise InferenceError so make sure to handle that for the call sites
      where it is used

    infer_call_result started recently to raise InferenceError for objects for which it
    could not find any returns. Previously it was silently raising a StopIteration,
    which was especially leaking when calling builtin methods.
    Since it is after all an inference method, it is expected that it
    could raise an InferenceError rather than returning nothing.

(adam)

Updated databases/redis, graphics/png, graphics/pngcrush

(adam)

pngcrush: updated to 1.8.13

Version 1.8.13 (built with libpng-1.6.32 and zlib-1.2.11)
  Add "exit(0)" after processing "-version" argument, to avoid
    displaying the Usage information
  Fix problem with MacOS prior to Sierra; it uses CLOCK_MONOTONIC
    for some other purpose

Version 1.8.12 (built with libpng-1.6.31 and zlib-1.2.11)
  Added POWERPC-VSX support.
  Report whether using optimizations.
  Added filter_method 6 (same as filter 5 with -speed).
  Added "methods" 149-176 (that use filter_method 6).
  Changed default verbosity from 1 (normal) to 0 (quiet). Use "-v" to get
    the previous default behavior and "-v -v" to get the previous "verbose"
    behavior. The "-s" (silent) and "-q" (quiet) options behave as before.

(adam)

png: updated to 1.6.35

Version 1.6.35:
* Replaced the remaining uses of png_size_t with size_t (Cosmin)
* Fixed the calculation of row_factor in png_check_chunk_length
* Added missing parentheses to a macro definition

(adam)

redis: updated to 4.0.10

Redis 4.0.10 fixes a number of important issues:

* Important security issues related to the Lua scripting engine.
  Please check https://github.com/antirez/redis/issues/5017
  for more information.

* A bug with SCAN, SSCAN, HSCAN and ZSCAN, that may not return all the elements.
  We also add a regression test that can trigger the issue often when present, and
  may in theory be able to find unrelated regressions.

* A PSYNC2 bug is fixed: Redis should not expire keys when saving RDB files
  because otherwise it is no longer possible to use such RDB file as a base
  for partial resynchronization. It no longer represents the right state.

* Compatibility of AOF with RDB preamble when the RDB checksum is disabled.

* Sentinel bug that in some cases prevented Sentinel to detect that the master
  was down immediately. A delay was added to the detection.

* Other minor issues.

(adam)

Updated www/py-django, devel/py-ipython

(adam)

py-ipython: updated to 6.5.0

IPython 6.5.0

Miscellaneous bug fixes and compatibility with Python 3.7.

* Autocompletion fix for modules with out __init__.py
* update the %pastebin magic to use dpaste.com instead of GitHub Gist
  which now requires authentication
* Fix crash with multiprocessing

(adam)

py-django: updated to 1.11.5

1.11.5:
Fix CVE-2018-14574: Open redirect possibility in CommonMiddleware

If the CommonMiddleware and the APPEND_SLASH setting are both enabled, and if the project has a URL pattern that accepts any path ending in a slash (many content management systems have such a pattern), then a request to a maliciously crafted URL of that site could lead to a redirect to another site, enabling phishing and other attacks.

CommonMiddleware now escapes leading slashes to prevent redirects to other domains.

(adam)

Updated www/py-WebOb, www/py-WebTest

(adam)

py-WebTest: updated to 2.0.30

2.0.30:
- Add Email class for input fields with type "email".
- Documentation bearer token and JWT authorization

(adam)

py-WebOb: updated to 1.8.2

1.8.2:
Bugfix
- SameSite may now be passed as str or bytes to Response.set_cookie and
  cookies.make_cookie. This was an oversight as all other arguments would be
  correctly coerced before being serialized.

(adam)

Updated games/wesnoth

(adam)

wesnoth: updated to 1.14.4

Version 1.14.4
* Security Fixes
  * Fixed Lua being able to escape sandboxing via load/loadstring (CVE-2018-1999023).
* Add-ons server
  * Made it so plain-text .po catalogues in add-ons are detected and added to
    the list of translations for them.
* Campaigns
  * Dead Water:
    * In 'Tirigaz', take the situation into account of orcs being killed either
      first or by undead.
  * Delfador's Memoirs:
    * Fix hero units costing upkeep
  * Eastern Invasion:
    * Fixed missing prisoners and loss of recallable units in 'Captured'.
  * Northern Rebirth:
    * Level 0 units are not available anymore after scenario 5
  * Secrets of the Ancients:
    * Adjust gender of enemies to better match story in S11 & 21
    * Simplify dialog to fix possible confusion in S16
    * Have nagas be able to recruit in S21
  * The South Guard:
    * S4: undead leader won't leave the castle anymore
    * S5: the untypical situation that one can defeat the lich before finding
      Urza Afalas is now handled
  * Under the Burning Suns:
    * S11: added custom graphics for the citadel.
    * S12: clarified the alien bodies' weaknesses.
    * Various visual improvements.
* Editor
  * Fixed saving a map as a scenario not enabling scenario editor tools.
* Graphics
  * New attack animation for the Peasant.
  * Tweaked the Ruffian's attack animation timing.
* Language and i18n
  * Updated translations: British English, Bulgarian, Chinese (Simplified),
    Chinese (Traditional), Czech, French, German, Hungarian, Japanese,
    Scottish Gaelic, Slovak, Spanish
* Lua API
  * Upgrade to Lua 5.3.5.
* Multiplayer server
  * Fixed lobby and whisper messages not having a maximum length.
* User interface
  * Improved the layout of the Statistics dialog.
  * Allow changing dropdown menu selections with the scrollwheel.
  * Fixed lobby chat box scrolling to top on a new message if it isn't at the
    bottom
  * Fixed the unit preview pane not showing the default race icon when detailing
    a single unit's stats.
  * Sort units secondarily by XP in the unit list dialog.
  * Whiteboard related bugfixes
* WML engine
  * Fixed errors about WESNOTH_VERSION not being defined when trying to load
    add-ons that have preprocessor errors.
* Miscellaneous and bug fixes
  * Added an advanced preference to enable experimental PRNG combat.
  * Campfires use illumination instead of a different ToD.
  * Linux builds now enable security hardening by default.
  * Fixed MP admins being unable to observe private games.
  * Fixed MP faction, leader, and leader gender changes persisting even if the
    selection dialog is dismissed.
  * Fixed an issue with positioned sound sources ignoring the volume set in
    Preferences after going off the audible radius and back.
  * Fixed wmllint choking on gzipped binary files (e.g. gzipped tarballs).
  * Fixed wmllint crashing on nonexistent paths provided in the command line.
  * Slight changes to the objectives dialogue
  * Greatly improved touch control support.
  * Fixed wmlindent crashing on nonexistent paths provided in the command line

(adam)

SDL2_mixer: removed unused PKGCONFIG_OVERRIDE lines

(adam)

Updated audio/pulseaudio, print/cups-filters

(adam)

cups-filters: updated to 1.20.4

CHANGES IN V1.20.4
- README: Added link to Issue Tracker on GitHub.
- gstoraster: Removed unneeded "if"s.
- cups-browsed: When checking whether there is already a local
  print queue with the same URI as the one of the discovered
  printer, consider also as equal URI if the URIs only differ
  by use of IPP or IPPS and/or use of HTTPS port 443 instead
  of IPP port 631.
- cups-browsed: Also upgrade from ipp: to ipps: when the ipps:
  URI is on HTTPS port 443 instead of IPP port 631. This is
  common on IPP network printers.
- pdftopdf: Removed support for hardware-implemented reversing
  of page order in PostScript printers. It was once not
  correctly implemented in cups-filters and second, such
  printers are extremely rare, and on Gutenprint PPDs with
  pseudo OutputOrder option hardware reversing was even
  wrongly assumed.
- pdftopdf: Accept option "output-order=normal/reverse" for
  reversing page order and also "page-delivery=
  same-order/reverse-order".
- libcupsfilters: Let the PPD generator add "*PageStackOrder
  ..."  lines to the choices of the "OutputBin" option, to
  mark which output bins need the pages printed in reverse
  order.
- libcupsfilters: Let the PPD generator correctly create a
  "*DefaultOutputOrder: ..."  entry, depending on whether the
  paper is put out face-up or face-down in the default output
  bin.
- libcupsfilters: Fixed human-readable name of the OutputBin
  option in the PPD generator.
- pdftoopvp: Silence compiler warning.
- cups-browsed: If the user modifies/overwrites a print queue
  created by cups-browsed, it will now not only be
  automatically released from the control of cups-browsed, but
  we also create a replacement for our generated local queue
  under a new name.
- cups-browsed: Make URIS for using the implicitclass backend
  correctly working also with queue names containing an '@'
  character.
- braille: Strengthen error checking.
- braille: Index: Replace bogus characters with space.
- braille: Add print and braille page number options.
- braille: Index: Use standard duplex cups option.
- cups-browsed: Moved auto-generation of PPD file for IPP
  network printers from create_remote_printer_entry()
  function to update_cups_queues(). This allows re-creating
  accidentally removed or overwritten local queues without
  losing the PPD file.
- braille: Add option to pick hyphenation rule according to
  current locale and make it the default for second
  translation table.
- braille: Remove generated defs on "make clean".
- braille: Turn non-breakable spaces to spaces.
- braille: Fix character encoding when extracting text. When
  extracing text from a zip file or a pdf, the resulting text
  is always utf-8 independently of the original locale, so we
  need to force that.
- braille: Warn when no text translation was selected in case
  the user didn't notice.
- braille: Fix spurious spacing after last Form-Feed

(adam)

pulseaudio: updated to 12.2

PulseAudio 12.2
The previous release tarball contained a broken configure script, this release
fixes the tarball. There are no changes in the source, except for this NEWS
file update.

PulseAudio 12.1
A bug fix release.
* Fixed crash when switching to A2DP bluetooth profile
* Fixed plugin search path in module-ladspa-sink
* Fixed file permissions for the pipes created by module-pipe-sink and
  module-pipe-source

(adam)

Added devel/py-hamcrest; Updated net/py-twisted

(adam)

py-twisted: updated to 18.7.0

Twisted 18.7.0:

Features
--------
- Cancelling a Deferred returned by twisted.internet.defer.inlineCallbacks now cancels the Deferred it is waiting on.
- twisted.application.internet.ClientService now accepts a function to initialize or validate a connection before it is returned by the whenConnected method as the prepareConnection argument.
- Traceback generated for twisted.internet.defer.inlineCallbacks now includes the full stack of inlineCallbacks generators between catcher and raiser (before it only contained raiser's stack).
- Add optional cwd argument to twisted.runner.procmon.ProcMon.addProcess
- twisted.python.failure.Failure tracebacks generated by coroutines scheduled with twisted.internet.defer.ensureDeferred - i.e. any Deferred-awaiting coroutine - now contain fewer extraneous frames from the trampoline implementation, and correctly indicate the source of exceptions raised in other call stacks - i.e. the function that raised the exception.  In other words: if you 'await' a function that raises an exception, you'll be able to see where the error came from.

Bugfixes
--------
- On UNIX-like platforms, Twisted attempts to recover from EMFILE when accepting connections on TCP and UNIX ports by shedding incoming clients.
- The documentation of IReactorTime.getDelayedCalls() has been corrected to indicate that the method returns a list, not a tuple.
- "python -m twisted web --help" now refers to "--listen" instead of the non-existing "--http"
- twisted.python.htmlizer.TokenPrinter now explicitly works on bytestrings.
- twisted.enterprise.adbapi.ConnectionPool.runWithConnection and runInteraction now use the reactor that is passed to ConnectionPool's constructor.

Improved Documentation
----------------------
- The Twisted Coding Standard now contains examples of how to mark up a feature as added in the next Twisted release.

Deprecations and Removals
-------------------------
- Deprecate direct introspection of ProcMon's processes: processes should not be directly accessed or pickled.
- twisted.internet.address.IPv4Address._bwHack and twisted.internet.address.UNIXAddress._bwHack, as well as the parameters to the constructors, deprecated since Twisted 11.0, have been removed.

(adam)

py-hamcrest: added version 1.9.0

PyHamcrest is a framework for writing matcher objects, allowing you to
declaratively define "match" rules. There are a number of situations where
matchers are invaluable, such as UI validation, or data filtering, but it is in
the area of writing flexible tests that matchers are most commonly used.

When writing tests it is sometimes difficult to get the balance right between
overspecifying the test (and making it brittle to changes), and not specifying
enough (making the test less valuable since it continues to pass even when the
thing being tested is broken). Having a tool that allows you to pick out
precisely the aspect under test and describe the values it should have, to a
controlled level of precision, helps greatly in writing tests that are "just
right." Such tests fail when the behavior of the aspect under test deviates
from the expected behavior, yet continue to pass when minor, unrelated changes
to the behaviour are made.

(adam)

Updated devel/py-archinfo, devel/py-requests-mock

(adam)

py-requests-mock: updated to 1.5.2

1.5.2:
Prelude
Fix py.test plugin with py.test < 3.0

Bug Fixes
Fixed a bug relating to how the pytest version was being discovered that meant new versions of pytest were being treated as old versions and would receive bad configuration.
The py.test plugin was broken when using py.test < 3.0. The version of py.test that ships in EPEL is only 2.7 so we need to make sure we support at least that version.

1.5.1:
New Features
The stream parameter is recorded when the request is sent and available in request history in the same was as parameters like verify or timeout.

(adam)

py-archinfo: updated to 7.8.7.1

7.8.7.1:
Unknown changes.

(adam)

Updated devel/py-test-testmon, devel/py-test-timeout, devel/py-test-xdist

(adam)

py-test-xdist: updated to 1.22.3

pytest-xdist 1.22.3:

Bug Fixes
Fix issue of virtualized or containerized environments not reporting the number of CPUs correctly.

Trivial Changes
Make all classes subclass from object and fix super() call in LoadFileScheduling.

(adam)

py-test-timeout: updated to 1.3.1

1.3.1:
Fix deprecation warning on Python 3.6.
Create a valid tag for the release. Somehow this didn't happen for 1.3.0, that tag points to a non-existing commit.

(adam)

py-test-testmon: updated to 0.9.13

0.9.13:
Encourage bug reporting in the README.

(adam)

Updated databases/py-sqlalchemy, databases/py-alembic

(adam)

py-alembic: updated to 1.0.0

1.0.0:
[feature] [general] For Alembic 1.0, Python 2.6 / 3.3 support is being dropped, allowing a fixed setup.py to be built as well as universal wheels. Pull request courtesy Hugo.

[feature] [general] With the 1.0 release, Alembic窶冱 minimum SQLAlchemy support version moves to 0.9.0, previously 0.7.9.

[bug] [batch] Fixed issue in batch where dropping a primary key column, then adding it back under the same name but without the primary_key flag, would not remove it from the existing PrimaryKeyConstraint. If a new PrimaryKeyConstraint is added, it is used as-is, as was the case before.

(adam)

py-sqlalchemy: updated to 1.2.10

1.2.10:
[orm] [bug] Fixed bug in Bundle construct where placing two columns of the same name would be de-duplicated, when the Bundle were used as part of the rendered SQL, such as in the ORDER BY or GROUP BY of the statement.

[orm] [bug] Fixed regression in 1.2.9 due to 4287 where using a Load option in conjunction with a string wildcard would result in a TypeError.

[sql] [bug] Fixed bug where a Sequence would be dropped explicitly before any Table that refers to it, which breaks in the case when the sequence is also involved in a server-side default for that table, when using MetaData.drop_all(). The step which processes sequences to be dropped via non server-side column default functions is now invoked after the table itself is dropped.

(adam)

www/py-django-cors-headers, www/py-django-extensions

(adam)

py-django-extensions: updated to 2.1.0

2.1.0:
Fix: travis

2.0.9:
Improvement: use README as project description on PyPI

2.0.8:
Please stop using ForeignKeyAutocompleteAdmin edition :-)
Fix: special markers in runserver_plus.rst
Fix: shell_plus, refactor reading pythonrc file outside of exec(compile(...))
Fix: reset_db, fix default utf8 support
Fix: autoslugfield, Fix autoslug generation when existing model is copied
Improvement: Cleanup management commands options after argparse migration 916
Improvement: sqldiff, add more tests
Improvement: sqldiff, add DurationField and SearchVectorField
Improvement: shell_plus, add more tests
Improvement: shell_plus, backport macos fix for tab completion
Improvement: clear_cache, add --all option
Improvement: pipchecker, treat dev versions as unstable
Deprecation: ForeignKeyAutocompleteAdmin, Django 2.0 has similar capabilities, which are much better supported.

(adam)

py-django-cors-headers: updated to 2.4.0

2.4.0:
Always add 'Origin' to the 'Vary' header for responses to enabled URL's, to prevent caching of responses intended for one origin being served for another.

(adam)