v1.8.2
perf: *much* faster dictionary compression on small files
perf: improved decompression speed and binary size
perf: slightly faster HC compression and decompression speed
perf: very small compression ratio improvement
fix : compression compatible with low memory addresses (< 0xFFFF)
fix : decompression segfault when provided with NULL input
cli : new command --favor-decSpeed
cli : benchmark mode more accurate for small inputs
fullbench : can bench _destSize() variants
doc : clarified block format parsing restrictions
18.1.0:
- x=X(); x.cycle = x; repr(x) will no longer raise a RecursionError, and will instead show as X(x=...).
- attr.ib(factory=f) is now syntactic sugar for the common case of attr.ib(default=attr.Factory(f)).
- Added attr.field_dict() to return an ordered dictionary of attrs attributes for a class, whose keys are the attribute names.
- The order of attributes that are passed into attr.make_class() or the these argument of @attr.s() is now retained if the dictionary is ordered (i.e. dict on Python 3.6 and later, collections.OrderedDict otherwise).
Before, the order was always determined by the order in which the attributes have been defined which may not be desirable when creating classes programatically.
- In slotted classes, __getstate__ and __setstate__ now ignore the __weakref__ attribute.
- Setting the cell type is now completely best effort.
This fixes attrs on Jython.
We cannot make any guarantees regarding Jython though, because our test suite cannot run due to dependency incompatabilities.
- If attr.s is passed a *these* argument, it will not attempt to remove attributes with the same name from the class body anymore.
- The hash of attr.NOTHING is now vegan and faster on 32bit Python builds.
- The overhead of instantiating frozen dict classes is virtually eliminated.
- Generated __init__ methods now have an __annotations__ attribute derived from the types of the fields.
- We have restructured the documentation a bit to account for attrs' growth in scope.
4.5.0:
- Drop support for 3.3, avoid accidental dependence breakage via setup.py.
- Allow registering and unregistering instance methods as listeners.
- Synchronize and simplify zope/__init__.py.
Changes 2.7:
bug in span stats printing introduced by new scalable page heap change was fixed.
Christoph Müllner has contributed couple warnings fixes and initial support for aarch64_ilp32 architecture.
Ben Dang contributed documentation fix for heap checker.
Fabrice Fontaine contributed fixed for linking benchmarks with --disable-static.
Holy Wu has added sized deallocation unit tests.
Holy Wu has enabled support of sized deallocation (c++14) on recent MSVC.
Holy Wu has fixed MSVC build in WIN32_OVERRIDE_ALLOCATORS mode.
Holy Wu has contributed cleanup of config.h used on windows.
Mao Huang has contributed couple simple tcmalloc changes from chromium code base. Making our tcmalloc forks a tiny bit closer.
issue 946 that caused compilation failures on some Linux clang installations has been fixed. Much thanks to github user htuch for helping to diagnose issue and proposing a fix.
Tulio Magno Quites Machado Filho has contributed build-time fix for PPC (for problem introduced in one of commits since RC).
1.22.2:
* Include the Negator when reverse engineering SQL for operators
* Fix the psql plugin command on OSX to work more reliably
* Prevent a crash in some situtions where the database connection is lost and needs to be reset.
* Fix display of elapsed query time for queries running longer than 1 hour
1.22.1:
* Don't barf when connecting to Greenplum or Greenplum HAWQ
* Fix selection of conversion functions in the CREATE CONVERSION dialogue
* Properly quote type names in the Type dialogue, and ensure range information is displayed for existing types
Notable changes since 1.0.0:
- RSQLite has been rewritten (essentially from scratch) in C++ with
Rcpp.
- You can now use SQLite’s URL specification for databases.
- Queries, query parameters and table data are always converted to
UTF-8 before being sent to the database.
- New strategy for prepared queries. Create a prepared query with
dbSendQuery() or dbSendStatement() and bind values with dbBind().
- dbSendQuery(), dbGetQuery(), dbSendStatement() and dbExecute() also
support inline parameterised queries.
- Improve column type inference.
- dbFetch() uses the same row name strategy as dbReadTable().
- dbColumnInfo() will now return information even before you’ve
retrieved any data.
- New sqliteVersion() prints the header and library versions of
RSQLite.
- Deprecation warnings are given only once, with a clear reference to
the source.
- datasetsDb() now returns a read-only database, to avoid
modifications to the installed file.
- Values of class "integer64" are now supported for dbWriteTable() and
dbBind().
- New connections now automatically load default RSQLite extensions.
- Implement dbUnquoteIdentifier().
Fixes PR pkg/53263, tested by Mayuresh.
0.20.1:
Fixed
Avoid attempting to kill Firefox process that has stopped.
With the change to allow Firefox enough time to shut down in
0.20.0, geckodriver started unconditionally killing the process
to reap its exit status. This caused geckodriver to inaccurately
report a successful Firefox shutdown as a failure.
The regression should not have caused any functional problems, but
the termination cause and the exit status are now reported correctly.
0.20.0:
Added
New --jsdebugger flag to open the Browser Toolbox when Firefox
launches. This is useful for debugging Marionette internals.
Introduced the temporary, boolean capability
moz:useNonSpecCompliantPointerOrigin to disable the WebDriver
conforming behavior of calculating the Pointer Origin.
Changed
HTTP status code for the StaleElementReference error changed
from 400 (Bad Request) to 404 (Not Found).
Backtraces from geckodriver no longer substitute for missing
Marionette stacktraces.
webdriver crate upgraded to 0.35.0.
Fixed
The Firefox process is now given ample time to shut down, allowing
enough time for the Firefox shutdown hang monitor to kick in.
Firefox has an integrated background monitor that observes
long-running threads during shutdown. These threads will be
killed after 63 seconds in the event of a hang. To allow Firefox
to shut down these threads on its own, geckodriver has to wait
that time and some additional seconds.
Grapheme clusters are now accepted as input for keyboard input
to actions.
Input to the value field of the keyDown and keyUp action
primitives used to only accept single characters, which means
geckodriver would error when a valid grapheme cluster was sent in,
for example with the tamil nadu character U+0BA8 U+0BBF.
Thanks to Greg Fraley for fixing this bug.
Improved error messages for malformed capability values.
Changes:
- New dbAppendTable() that by default calls sqlAppendTableTemplate()
and then dbExecute() with a param argument, without support for
row.names argument.
- New dbCreateTable() that by default calls sqlCreateTable() and then
dbExecute(), without support for row.names argument.
- New dbCanConnect() generic with default implementation.
- New dbIsReadOnly() generic with default implementation.
- sqlAppendTable() now accepts lists for the values argument, to
support lists of SQL objects in R 3.1.
- Add default implementation for dbListFields(DBIConnection, Id), this
relies on dbQuoteIdentifier(DBIConnection, Id).
- Documentation updates and ug fixes.
The pick utility allows users to select from a set of choices using an interface
with fuzzy search functionality. The choices are read from stdin, and the
selected choice written to stdout.
Packaged in pkgsrc-wip by Jenz Guenther, thanks!
The package now depends on devel/ocaml-ppxlib instead of on its
constituent parts. There have also been some minor improvements from
version 0.9 to 0.10.
This is a package that subsumes several other packages (see DESCR for
details). All dependencies on these packages have already been removed
(or will be soon, as soon as I finish committing this batch of updates)
and the packages themselves will be removed from pkgsrc.
