Upstream changes:
Features
- add type CAA to libpyunbound (accessing libunbound from python).
- Fix#17: Add python module example from Jan Janak, that is a
plugin for the Unbound DNS resolver to resolve DNS records in
multicast DNS [RFC 6762] via Avahi. The plugin communicates
with Avahi via DBus. The comment section at the beginning of
the file contains detailed documentation.
- travis build file.
- PR #16: XoT support, AXFR over TLS, turn it on with
master: <ip>#<authname> in unbound.conf. This uses TLS to
download the AXFR (or IXFR).
Bug Fixes
- Fix for #4233: guard use of NDEBUG, so that it can be passed in
CFLAGS into configure.
- Add log message, at verbosity 4, that says the query is encrypted
with TLS, if that is enabled for the query.
- Fix#4239: set NOTIMPL when deny-any is enabled, for RFC8482.
- Fix#4240: Fix whitespace cleanup in example.conf.
- Fix that tls-session-ticket-keys: "" on its own in unbound.conf
disables the tls session ticker key calls into the OpenSSL API.
- Fix crash if tls-servic-pem not filled in when necessary.
- Fix auth-zone NSEC3 response for empty nonterminals with exact
match nsec3 records.
- Fix for out of bounds integers, thanks to OSTIF audit. It is in
allocation debug code.
- Fix for auth zone nsec3 ent fix for wildcard nodata.
- Move goto label in answer_from_cache to the end of the function
where it is more visible.
- Fix auth-zone NSEC3 response for wildcard nodata answers,
include the closest encloser in the answer.
- Fix spelling error in log output for event method.
- Fix to reinit event structure for accepted TCP (and TLS) sockets.
- Fix to use event_assign with libevent for thread-safety.
- verbose information about auth zone lookup process, also lookup
start, timeout and fail.
- Fix to wipe ssl ticket keys from memory with explicit_bzero,
if available.
- Fix that auth zone uses correct network type for sockets for
SOA serial probes. This fixes that probes fail because earlier
probe addresses are unreachable.
- Fix that auth zone fails over to next master for timeout in tcp.
- Squelch SSL read and write connection reset by peer and broken pipe
messages. Verbosity 2 and higher enables them.
- Update python documentation for init_standard().
- Typos.
- Fix tls write event for read state change to re-call SSL_write and
not resume the TLS handshake.
- Better braces in if statement in TCP fastopen code.
- iana portlist updated.
- Scrub RRs from answer section when reusing NXDOMAIN message for
subdomain answers.
- For harden-below-nxdomain: do not consider a name to be non-exitent
when message contains a CNAME record.
- Fix wrong query name in local zone redirect answers with a CNAME,
the copy of the local alias is in unpacked form.
- contrib/fastrpz.patch updated for code changes, and with git diff.
- Fix#29: Solaris 11.3 and missing symbols be64toh, htobe64.
- Fix#30: AddressSanitizer finding in lookup3.c. This sets the
hash function to use a slower but better auditable code that does
not read beyond array boundaries. This makes code better security
checkable, and is better for security. It is fixed to be slower,
but not read outside of the array.
- Fix edns-subnet locks, in error cases the lock was not unlocked.
- Fix doxygen output error on readme markdown vignettes.
- Squelch log messages from tcp send about connection reset by peer.
They can be enabled with verbosity at higher values for diagnosing
network connectivity issues.
- Attempt to fix malformed tcp response.
- Fix#31: swig 4.0 and python module.
- Note that so-reuseport at extreme load is better turned off,
otherwise queries are not distributed evenly, on Linux 4.4.x.
- Fix that spoolbuf is not used to store tcp pipelined response
between mesh send and callback end.
- Fix double file close in tcp pipelined response code.
- Fix to define _OPENBSD_SOURCE to get reallocarray on NetBSD.
- Fix to guard _OPENBSD_SOURCE from redefinition.
- Fix that fixes the Fix that spoolbuf is not used to store tcp
pipelined response between mesh send and callback end, this fixes
error cases that did not use the correct spoolbuf.
- Fix that fixes the Fix that spoolbuf is not used to store tcp
pipelined response between mesh send and callback end, this fixes
error cases that did not use the correct spoolbuf.
- Fix another spoolbuf storage code point, in prefetch.
1.9.74.123:
- Correct calculation in haDelta indicator
- Use initial datalabel for non-overlaid volume plot
1.9.73.123:
- Add utility NonZeroDifference indicator
- Redefine CrossUp, CrossDown and CrossOver indicators using
NonZeroDifference to cover the case in which the crossing entities
converge right before crossing up and down
1.9.72.122:
- Cover case in which result in high-level overridden operations have
multiple lines and wer not be taken into account for minimum period
calculations
- Add "Int" variants of percentage based sizers to import
- Trades observer to show net profit instead of brutto, with parameter
to control behavior
1.9.71.122:
- Improve on indicator legend plotting to overcome matplotlib legend
reordering
- Added PercenSizerInt and AllSizerInt which truncate the returned size
to an int, suited better for stocks/futures
1.9.70.122:
- Use opening price for submission check for Market orders when
cheat-on-open is active
- Update pnlcomm on all operations and not just profit/loss locking
- Correct comment for fillalpha and add baralpha for candlestick opacity
- Use internal dict for data feed presence test and update trade observer
1.9.69.122:
- Fix offline Yahoo feed by moving the new adjclose line up to the offline
feed
- Adapt the yahoodownload tool to the current status (ex: data not reversed)
- Redownload all yahoo data feeds
1.9.68.122
- Fix call to _nextday in TradingCalendar
- Clean up and rework of Yahoo Data. The data feeds seems to be reliable
again
- IBStore Support for IND prices
1.9.67.122
- Fix compression only scenarios when resampling and resampling after
changes in 1.9.66.122
- Final correction for rollover fix introduced in 1.9.66.122
- Cover use case for mininum period calculation when all
operations/indicators don't use the data feeds directly but lines of it
1.9.66.122
- Fix regression introduced with 8f537a1c2c271eb5cfc592b373697732597d26d6
which voids the count of lost trades
- Allow rollover to distinguish between no values temporarily (with None)
and no values permanently (with False)
- Avoid math domain error for negative returns in logarithmic calculations
- Fix local variable declaration for compound returns
- Fix typo in date2num tz conversion which shows up in direct usage
1.9.65.122
- Fix commission info assigment and orderref seeking in OandaStore
- Add strategy type to OptReturn
- Fix prepend_constant for OLS_Transformation
- Fix LogReturnsRolling compression when not specified
- Have ints instead of bools in some values with 1 Trade in TradeAnalyzer
1.9.64.122
- Avoid stage2 comparison using [0] in API methods
- Support plotname, if given, as name of indicator in csv output
1.9.63.122
- Add optimization callbacks when running with 1 Core
- Correct sell_bracket by removing old append code
- Correct typo in store.py
- Pass period from RateOfChange100 to underlying ROC
1.9.62.122
- Correct PSAR acceleration capping
- Enable PandasData line extension without the need to extend datafields
1.9.61.122
- Add `_skipnan` to plotlines to allow joining two points with a line
- buy_bracket/sell_bracket allow suppressing stop/limit orders
- Add stop-loss approaches sample
- Correct codes for minutes compression
1.9.60.122
- Remove unused files
- README update, Docstring corrections, documentation corrections
- Update travis settings
1.9.58.122
- Provide default fundmode methods for all brokers
- Correct order notification if positions exist when starting the broker
and will be simulated
- Correct csv values output if object has no length
1.9.57.122
- Fix set_fundmode in bbroker
- Synchronize fund history mode with master clock
- Allow relocation of legend in plotting charts
- Adapt broker observer to fund mode
1.9.56.122
- Handle volume as string null in YahooFinanceData
- Corrections/Improvements to order history support
- Add fund history support
- Increase plotting margin of trade observers
v0.3.6
Add option to compute forward returns non-cumulatively
v0.3.5
This is a minor release from 0.3.4 that includes bugfixes, speed enhancement and compatibility with more recent pandas versions. We recommend that all users upgrade to this version.
v0.3.4
This is a minor release from 0.3.3 that includes bugfixes, small enhancements and backward compatibility breakages. We recommend that all users upgrade to this version.
v0.3.3
TEST: added tests for perf.mean_return_by_quantile
Implementation of POSIX bc with GNU extensions
This bc is optimized for use by programmers. It includes several
extensions beyond the GNU extensions for that purpose.
Prepared in pkgsrc-wip by Gavin Howard.
Changes since b158
Beta #159 - 15.06.2019
- The audio sampling tool in Smp. Ed. now supports stereo sampling
- Several tools in Smp. Ed. (Mix, Resample, Echo, Volume) now accepts
pressing Enter on the keyboard.
v0.9.0:
This is a major release with several substantial and long-desired new features. There are also updates/modifications to the themes and color palettes that give better consistency with matplotlib 2.0 and some notable API changes.
PyMC3 3.7 (May 29 2019)
New features
Add data container class (Data) that wraps the theano SharedVariable class and let the model be aware of its inputs and outputs.
Add function set_data to update variables defined as Data.
Mixture now supports mixtures of multidimensional probability distributions, not just lists of 1D distributions.
GLM.from_formula and LinearComponent.from_formula can extract variables from the calling scope. Customizable via the new eval_env argument.
Added the distributions.shape_utils module with functions used to help broadcast samples drawn from distributions using the size keyword argument.
Used numpy.vectorize in distributions.distribution._compile_theano_function. This enables sample_prior_predictive and sample_posterior_predictive to ask for tuples of samples instead of just integers.
Maintenance
All occurances of sd as a parameter name have been renamed to sigma. sd will continue to function for backwards compatibility.
HamiltonianMC was ignoring certain arguments like target_accept, and not using the custom step size jitter function with expectation 1.
Made BrokenPipeError for parallel sampling more verbose on Windows.
Added the broadcast_distribution_samples function that helps broadcasting arrays of drawn samples, taking into account the requested size and the inferred distribution shape. This sometimes is needed by distributions that call several rvs separately within their random method, such as the ZeroInflatedPoisson.
The Wald, Kumaraswamy, LogNormal, Pareto, Cauchy, HalfCauchy, Weibull and ExGaussian distributions random method used a hidden _random function that was written with scalars in mind. This could potentially lead to artificial correlations between random draws. Added shape guards and broadcasting of the distribution samples to prevent this.
Added a fix to allow the imputation of single missing values of observed data, which previously would fail.
The draw_values function was too permissive with what could be grabbed from inside point, which lead to an error when sampling posterior predictives of variables that depended on shared variables that had changed their shape after pm.sample() had been called.
draw_values now adds the theano graph descendants of TensorConstant or SharedVariables to the named relationship nodes stack, only if these descendants are ObservedRV or MultiObservedRV instances.
Fixed bug in broadcast_distrution_samples, which did not handle correctly cases in which some samples did not have the size tuple prepended.
Changed MvNormal.random's usage of tensordot for Cholesky encoded covariances. This lead to wrong axis broadcasting and seemed to be the cause for issue 3343.
Fixed defect in Mixture.random when multidimensional mixtures were involved. The mixture component was not preserved across all the elements of the dimensions of the mixture. This meant that the correlations across elements within a given draw of the mixture were partly broken.
Restructured Mixture.random to allow better use of vectorized calls to comp_dists.random.
Added tests for mixtures of multidimensional distributions to the test suite.
Fixed incorrect usage of broadcast_distribution_samples in DiscreteWeibull.
Mixture's default dtype is now determined by theano.config.floatX.
dist_math.random_choice now handles nd-arrays of category probabilities, and also handles sizes that are not None. Also removed unused k kwarg from dist_math.random_choice.
Changed Categorical.mode to preserve all the dimensions of p except the last one, which encodes each category's probability.
Changed initialization of Categorical.p. p is now normalized to sum to 1 inside logp and random, but not during initialization. This could hide negative values supplied to p as mentioned in 2082.
Categorical now accepts elements of p equal to 0. logp will return -inf if there are values that index to the zero probability categories.
Add sigma, tau, and sd to signature of NormalMixture.
Set default lower and upper values of -inf and inf for pm.distributions.continuous.TruncatedNormal. This avoids errors caused by their previous values of None.
Converted all calls to pm.distributions.bound._ContinuousBounded and pm.distributions.bound._DiscreteBounded to use only and all positional arguments.
Restructured distributions.distribution.generate_samples to use the shape_utils module. This solves issues 3421 and 3147 by using the size aware broadcating functions in shape_utils.
Fixed the Multinomial.random and Multinomial.random_ methods to make them compatible with the new generate_samples function. In the process, a bug of the Multinomial.random_ shape handling was discovered and fixed.
Fixed a defect found in Bound.random where the point dictionary was passed to generate_samples as an arg instead of in not_broadcast_kwargs.
Fixed a defect found in Bound.random_ where total_size could end up as a float64 instead of being an integer if given size=tuple().
Fixed an issue in model_graph that caused construction of the graph of the model for rendering to hang: replaced a search over the powerset of the nodes with a breadth-first search over the nodes.
Removed variable annotations from model_graph but left type hints. This means that we support python>=3.5.4.
Default target_acceptfor HamiltonianMC is now 0.65, as suggested in Beskos et. al. 2010 and Neal 2001.
Fixed bug in draw_values that lead to intermittent errors in python3.5. This happened with some deterministic nodes that were drawn but not added to givens.
Deprecations
nuts_kwargs and step_kwargs have been deprecated in favor of using the standard kwargs to pass optional step method arguments.
SGFS and CSG have been removed. They have been moved to pymc3-experimental.
References to live_plot and corresponding notebooks have been removed.
Function approx_hessian was removed, due to numdifftools becoming incompatible with current scipy. The function was already optional, only available to a user who installed numdifftools separately, and not hit on any common codepaths.
Deprecated vars parameter of sample_posterior_predictive in favor of varnames.
References to live_plot and corresponding notebooks have been removed.
Deprecated vars parameters of sample_posterior_predictive and sample_prior_predictive in favor of var_names. At least for the latter, this is more accurate, since the vars parameter actually took names.
Whats New in 0.24.2
Fixed Regressions
Bug Fixes
Whats New in 0.24.1
Changing the sort parameter for Index set operations
Fixed Regressions
Bug Fixes
What’s New in 0.24.0
This is a major release from 0.23.4 and includes a number of API changes, new features, enhancements, and performance improvements along with a large number of bug fixes.
Highlights include:
* Optional Integer NA Support
* New APIs for accessing the array backing a Series or Index
* A new top-level method for creating arrays
* Store Interval and Period data in a Series or DataFrame
* Support for joining on two MultiIndexes
pkgsrc changes:
- remove patch for totoridipjp (removed in upstream too)
- remove patch for upstream ticket #1323 (integrated in 3.9)
Upstream changes (from "new features of mikutter 3.9"):
* Mastodon plugin
* Support for UI scaling
* World account can be assigned to shortcut key
* Reorder World accounts
* Non-PNG image support for Skin
* Generic Model viewer
* Quick Step
* PulseAudio
* Reorgenize gem dependencies
Upstream pull request log:
https://github.com/twitter/twitter-text/pull/265
* New v3.json config file with emojiParsingEnabled config option. When
true, twitter-text will parse and discount emoji supported by the
twemoji library (see https://github.com/twitter/twemoji). The length
of these emoji will be the default weight (200 or two characters) even
if they contain multiple code points combined by zero-width
joiners. This means that emoji with skin tone and gender modifiers no
longer count as more characters than those without such modifiers.
All 4 implementations updated to reflect this change in counting.
* Updates known gTLDs to recognize recent additions by IANA (#261)
Upstream changes (from CHANGELOG.md):
# Addressable 2.6.0
- added `tld=` method to allow assignment to the public suffix
- most `heuristic_parse` patterns are now case-insensitive
- `heuristic_parse` handles more `file://` URI variations
- fixes bug in `heuristic_parse` when uri starts with digit
- fixes bug in `request_uri=` with query strings
- fixes template issues with `nil` and `?` operator
- `frozen_string_literal` pragmas added
- minor performance improvements in regexps
- fixes to eliminate warnings
meson `pkgconfig' module generate pkg-config files during the configure
phase and PKGCONFIG_OVERRIDE can be used in post-configure stage to
avoid directly patching meson.build files.
Discussed with <ryoon>, thanks!
What's new in h5py 2.9
New features
* A convenient high-level API for creating virtual datasets, HDF5 objects which
act as a view over one or more real datasets.
See :ref:vds for details.
* :class:File can now be constructed with a Python file-like object,
making it easy to create an HDF5 file in memory using :class:io.BytesIO
See :ref:file_fileobj for details.
* :class:File now accepts parameters to control the chunk cache.
See :ref:file_cache for details.
* New options to record the order of insertion for attributes and group entries.
Iterating over these collections now follows insertion order if it was
recorded, or name order if not.
* A new method :meth:Group.create_dataset_like to create a new dataset with
similar properties to an existing one.
* Datasets can now be created with storage backed by external non-HDF5 files
* Lists or tuples of unicode strings can now be stored as HDF5 attributes
* Inspecting the view returned by .keys() now shows the key names, for
convenient interactive use.
Exposing HDF5 functions
* H5LTopen_file_image as :func:h5py.h5f.open_file_image.
* External dataset storage functions H5Pset_external, H5Pget_external
and H5Pget_external_count as methods on :class:h5py.h5p.PropDCID
Bugfixes
* Fix reading/writing of float128 data.
* Converting data to float16 when creating a dataset.
Support for old Python
Support for Python 3.3 has been dropped.
Support for Python 2.6 has been dropped.
0.14.2
- [CHANGE] Travis CI builds now use tox to lint and run tests.
- [FIX] Fixed UnicodeDecodeError on certain locales.
0.14.1
- [FIX] Fixed "ImportError: No module named 'dateutil'".
0.14.0
- [NEW] Added provisional support for Python 3.8.
- [CHANGE] Removed support for EOL Python 3.4.
- [FIX] Updated setup.py with modern Python standards.
- [FIX] Upgraded dependencies to latest versions.
- [FIX] Enabled flake8 and black on travis builds.
- [FIX] Formatted code using black and isort.
REL: v3.1.0
The first release of the v3.1 series
This version of Matplotlib support Python 3.6+
Highlights of this release include:
- A helper method for scatter legends
- Secondary axis support (ex, degrees and radians)
- A concise date formatter
- No longer requires a framework build of Python to use OSX backend
- Major and minor ticks will no longer collide by default
- A progress callback for animations
- mplot3D is always registered
- deprecated mlab functions have been removed
- Many function calls have become stricter about invalid or ignored input
REL: v3.0.3
The third and last planned bug-fix release of the 3.0 series
* fix several possible memory leaks
* documentation fixes
jquery-ui is no longer bundled in git repo and is downloaded during
the install or sdist process. If you are installing from pypi the
files will be included, but if you are installing from git you will
need internet access the first time you run `pip install .`. The
install process tries to cache the files and will not need internet
access on future installations.
REL: v3.0.2
This is the second bug-fix release for the v3.0 series.
- Un-breaks basemap which was broken by partially restoring
private APIs for cartopy.
- Fixes bug in warning code when used in an embedded context.
- Fixes crash when using Tk and closing the first open window before showing it
- Many documentation improvements.
- Restore a corner case on ColorBar tick usage.
- Change the default behavior of `matplotlib.use` to silently allow
more 'safe' switching after auto-discovery, but before starting an event loop.
- Improvements to bounding box calculations.
- Provide the correct length for RcParams instances.
REL: v3.0.1
This is the first bug fix release for the 3.0 series which fixes
several
- Fix failure to import bug when used with backend-fallback on Python
3.6.7 and 3.7.1
- Fixed a number of failure to import bugs around finding fonts
- Fix Qt4 backend
- Fix bug on OSX that recursively searched current directory for fonts
- Fix bouncing-rocket on OSX when doing backend fallback and not
selecting OSX
- Temporarily restore several private APIs to unbreak cartopy
- Make pyplot more tolerant of varying signatures in 3rd-party
sub-classe
- Improve datetime64 unit handling
- Fixed several poor interactions with tight_layout
REL: v3.0.0
The first release of the Matplotlib 3.0 series
This is the first version of Matplotlib to only support Python 3.
Highlights of this release include:
- GUI backend is selected at run-time based on what toolkits are
installed. A GUI toolkit will not be selected on a headless
server.
- New cyclic color map *twilight*
- Improvements to automatic layout of titles, ticks, and GridSpec
- Many bug fixes!
