Release 1.0.8:
Added named exceptions to XlsxWriter. See The Exceptions Class.
Removed the implicit close() in the destructor since it wasn’t guaranteed to work correctly and raised a confusing exception when any other exception was triggered. Note that this is a backward incompatible change. The with context manager is a better way to close automatically, see close().
Added border, fill, pattern and gradient formatting options to set_legend().
Added top_right position to set_legend().
omega:
* New OmegaScript $unique command. The existing $uniq only removes adjacent
entries (like the Unix uniq command) so to fully remove duplicates you need a
sorted input. Sometimes it is desirable to remove duplicates from an
unsorted list without changing the order of the entries which are left, so
add $unique to do that. If the list is sorted already, then $uniq is more
efficient.
* Fix $map to cleanly reject a single argument.
templates:
* templates/query: Merge multiple entries in the term frequency information,
which came from searching several prefixes by default. Reported by Alistair
Buxton on #xapian-discuss.
* When multiple words with the same stem are in the query string we now fully
eliminate duplicates when showing term frequency information.
API:
* Database::check(): Fix bogus error reports for documents with length zero
due to a new check added in 1.4.6 that the doclength was between the stored
upper and lower bounds, which failed to allow for the lower bound ignoring
documents with length zero (since documents indexed only by boolean terms
aren't involved in weighted searches). Reported by David Bremner.
* Query: Use of Query::MatchAll in multithreaded code causes problems because
the reference counting gets messed up by concurrent updates. Document that
Query(string()) should be used instead of MatchAll in multithreaded code, and
avoid using it in library code. Reported by Germán M. Bravo.
* Stem:
+ Stemming algorithms added for Irish, Lithuanian, Nepali and Tamil.
+ Merge Snowball compiler changes which improve code generation.
+ Merge optimisations to the Arabic and Turkish stemmers.
glass backend:
* A long-lived cursor on a table in a WritableDatabase could get into
an invalid state, which typically resulted in a DatabaseCorruptError
being thrown with the message:
Db block overwritten - are there multiple writers?
But in fact the on-disk database is not corrupted - it's just that
the cursor in memory has got into an inconsistent state. It looks
like we'll always detect the inconsistency before it can cause on-disk
corruption but it's hard to be completely certain.
The bug is in code to rebuild the cursor when the underlying table
changes in ways which require that, which is a fairly rare occurrence
to start with, and only triggers when a block in the cursor has been
released, reallocated, and we tried to load it in the cursor at the
same level - the cursor wrongly assumes it has the current version
of the block.
Reported with a reproducer by Sylvain Taverne. Confirmed by David
Bremner as also fixing a problem in notmuch for which he hadn't managed
to find a reduced reproducer.
GtkSpell provides word-processor-style highlighting and replacement
of misspelled words in a GtkTextView widget. Right-clicking a
misspelled word pops up a menu of suggested replacements.
GtkSpell is a library and is mostly of interest to GTK programmers.
This is major version 3 of the library, which is API incompatible
to version 2.
0.722 2018-08-19 PHRED
- Remove unnecessary tests [github/paultcochrane for all]
- Fix link to dist repo
- Added continuous integration configurations
- Minor fixes and cosmetic changes to make code adhere better to current
Perl best practices
- Documentation fixes
0.08 2018-08-21T03:21:47
- Update project metadata: The Perl Shop is now project
maintainer, distribution homepage is now on MetaCPAN, and the
project now includes an appropriate LICENSE file.
This is based on the decision The NetBSD Foundation made in 2008 to
do so, which was already applied to src.
This change has been applied to code which is likely not in other
repositories.
ok board@, reviewed by riastradh@
0.72 Sun Jul 8 17:03:08 CEST 2018
- Update to libyaml 0.2.1 (tinita) It's forbidden now to escape single
quotes inside double quotes
- When disabling $LoadBlessed, return scalars not refs (tinita)
- Save anchors also for blessed scalars (tinita)
Use TEST_DEPENDS.
3.101 2018-08-06T14:32:38Z
- Ensure local_module_url_prefix defaults to current perldoc_url_prefix.
3.100 2018-08-06T02:07:18Z
- Add "local_module_url_prefix" and "local_module_re" args
to allow pointing links for local modules to another site [rt-120452].
Release 1.7.7:
Bugs fixed
* document not in toctree warning when including files only for parallel
builds
* LaTeX: reduce "Token not allowed in a PDF string" hyperref warnings in latex
console output
* LaTeX: suppress "remreset Warning: The remreset package is obsolete" in latex
console output with recent LaTeX
* PDF output: usage of PAPER environment variable is broken since Sphinx
1.5
* LaTeX: fix the :confval:`latex_engine` documentation regarding Latin Modern
font with XeLaTeX/LuaLateX
* autodoc: Fix wrong type annotations for complex typing
* autodoc: Optional types are wrongly rendered
* autodoc crashed by ForwardRef types
* autodoc: No docs generated for functools.partial functions
* autodoc: ``getargspec()`` raises NameError for invalid typehints
* imgmath: math_number_all causes equations to have two numbers in html
* sphinx-quickstart blank prompts in PowerShell
Release 2.2.6 Sun August 12 2018
Bug fixes:
#170#206 Avoid doing arithmetic with NULL pointers in XML_GetBuffer
#204#205 Fix 2.2.5 regression with suspend-resume while parsing
a document like '<root/>'
Other changes:
#165#168 Autotools: Fix docbook-related configure syntax error
#166 Autotools: Avoid grep option `-q` for Solaris
#167 Autotools: Support
./configure DOCBOOK_TO_MAN="xmlto man --skip-validation"
#159#167 Autotools: Support DOCBOOK_TO_MAN command which produces
xmlwf.1 rather than XMLWF.1; also covers case insensitive
file systems
#181 Autotools: Drop -rpath option passed to libtool
#188 Autotools: Detect and deny SGML docbook2man as ours is XML
#188 Autotools/CMake: Support command db2x_docbook2man as well
#174 CMake: Introduce option WARNINGS_AS_ERRORS, defaults to OFF
#184#185 CMake: Introduce option MSVC_USE_STATIC_CRT, defaults to OFF
#207#208 CMake: Introduce option XML_UNICODE and XML_UNICODE_WCHAR_T,
both defaulting to OFF
#175 CMake: Prefer check_symbol_exists over check_function_exists
#176 CMake: Create the same pkg-config file as with GNU Autotools
#178#179 CMake: Use GNUInstallDirs module to set proper defaults for
install directories
#208 CMake: Utilize expat_config.h.cmake for XML_DEV_URANDOM
#180 Windows: Fix compilation of test suite for Visual Studio 2008
#131#173#202 Address compiler warnings
#187#190#200 Fix miscellaneous typos
Version info bumped from 7:7:6 to 7:8:6
Release 2.2.5 Tue October 31 2017
Bug fixes:
#8 If the parser runs out of memory, make sure its internal
state reflects the memory it actually has, not the memory
it wanted to have.
#11 The default handler wasn't being called when it should for
a SYSTEM or PUBLIC doctype if an entity declaration handler
was registered.
#137#138 Fix a case of mistakenly reported parsing success where
XML_StopParser was called from an element handler
#162 Function XML_ErrorString was returning NULL rather than
a message for code XML_ERROR_INVALID_ARGUMENT
introduced with release 2.2.1
Other changes:
#106 xmlwf: Add argument -N adding notation declarations
#75#106 Test suite: Resolve expected failure cases where xmlwf
output was incomplete
#127 Windows: Fix test suite compilation
#126#127 Windows: Fix compilation for Visual Studio 2012
Windows: Upgrade shipped project files to Visual Studio 2017
#33#132 tests: Mass-fix compilation for XML_UNICODE_WCHAR_T
#129 examples: Fix compilation for XML_UNICODE_WCHAR_T
#130 benchmark: Fix compilation for XML_UNICODE_WCHAR_T
#144 xmlwf: Fix compilation for XML_UNICODE_WCHAR_T; still needs
Windows or MinGW for 2-byte wchar_t
#9 Address two Clang Static Analyzer false positives
#59 Resolve troublesome macros hiding parser struct membership
and dereferencing that pointer
#6 Resolve superfluous internal malloc/realloc switch
#153#155 Improve docbook2x-man detection
#160 Undefine NDEBUG in the test suite (rather than rejecting it)
#161 Address compiler warnings
Version info bumped from 7:6:6 to 7:7:6
2018-08-18 -- 0.8.6
* Fixed: Bad/NULL .hostText.afterLast when parsing certain rather pathologic
but well-formed URIs with empty host (e.g. "//:%aa@") (GitHub #15)
Thanks to Kurt Schwehr for the report!
* Fixed: Fix uriRemoveBaseUri for case where scheme, host name,
IPvFuture address or path segments of the source address were
string prefixes of the related counterpart in the base URI.
Thanks to Yang Yu for the patch! (GitHub #19, #20)
* Fixed: Make UriStringToUnixFilename and UriStringToWindowsFilename
support minimal representation a la RFC 8089, e.g. file:/bin/bash
(compare to file:///bin/bash with three slashes) (GitHub #12, #14)
Thanks to Zane van Iperen for the report!
* Fixed: Documentation typos (GitHub #10, #11)
Thanks to Graham Percival!
* Improved: Made API docs of uriRemoveBaseUri more clear
(related to GitHub #19)
* Soname: 1:22:0
0.12.2:
Fixed IPv4 formatted IP address returning True on ipv6
Fixed IPv6 address parsing
Fixed domain validator for international domains and certain edge cases
0.12.1:
Fixed IDNA encoded TLDs in domain validator
Fixed URL validator for URLs with invalid characters in userinfo part
0.12.0:
Added hash validators for md5, sha1, sha224, sha256 and sha512
Made ipv6 validator support IPv4-mapped IPv6 addresses
0.11.3:
Fixed URL validator for URLs containing localhost
0.11.2:
Fixed URL validator for urls with query parameters but without path
0.11.1:
Fixed pyp2rpm build problem
0.11.0:
Fixed public url validation
Made URL validator case insensitive
Drop Python 2.6 support
0.10.3:
Added public parameter to url validator
0.10.2:
Fixed various URL validation issues
0.10.1:
Fixed domain name validation for numeric domain names
Fixed IBAN validation for Norwegian and Belgian IBANs
0.10.0:
Added support for internationalized domain names in domain validator
0.9.0:
Added new validator: domain
Added flake8 and isort checks in travis config
0.8.0:
Added new validator: iban
Changes in version 1.14.4, released on August 8, 2018
--- MAJOR NEW FEATURES ---
* In ASCII output, render mathematical symbols and greek letters
as transliterations conveying the characters' meanings rather
than trying to imitate their shape. Consequently, such characters
can now be used in portable manual pages. All the same, please
limit their use to contexts where they really matter, for example
when showing complicated mathematical formulae.
* First steps towards better support for small screens in HTML
output (responsive design): avoid most style= attributes, in
particular all hard-coded indentations and column widths, and
provide a better mandoc.css style sheet with a @media query,
using em units throughout, and avoiding redundancy in selectors.
* Better HTML output with some more fitting HTML elements, eliminating
needless class= attributes, and avoiding various HTML syntax errors
(element nesting, URL-fragment syntax, duplicate id= attributes).
--- MINOR NEW FEATURES ---
* When a man(1) argument contains a slash, imply -l like in man-db.
* Use TIOCGWINSZ to reduce the default -Owidth and -Oindent during
interactive use on terminals narrower than 79 columns.
* Generated PostScript files are now more than 50% smaller.
* Terminal rendering of eqn(7) is improved in several respects.
* Simplified and nicer output from the mdoc(7) .Lk macro, formatting
all links in-line, even long ones.
* roff(7) \n+ and \n- numerical register auto-increment and -decrement
* roff(7) .nr optional third argument (auto-increment step size)
* Autodetect in ./configure whether the compiler can use -W and -static,
allowing to build on Solaris 10 and 11 without any configure.local.
--- RELIABILITY BUGFIXES ---
* Only activate UTF-8 output when the user really selected UTF-8,
not some other multibyte character encoding.
* Prevent excessive .ll arguments from generating infinite output.
* Fix out of bounds accesses to parse buffers that could happen when
using renamed or user defined macros after roff(7) conditionals.
* Avoid an assertion failure in certain .Bl -column lists.
* Avoid a NULL pointer access on deroff() failure after '.SS ""'.
* Fix a segfault that could be triggered by two invalid .Dt macros.
* Fix two syntax errors in generated PDF files.
* Properly state the page size in generated PostScript files.
* Close a memory leak caused by missing gzclose(3).
* Fix misformatting of man(7) documents lacking .SH macros
in PostScript and PDF output.
* And many minor bugfixes.
--- THANKS TO ---
* Marc Espie (OpenBSD) for implementing the size reduction of
PostScript files, one additional patch for code simplification,
and two bug reports.
* Theo Buehler (OpenBSD) for a bugfix patch,
and Theo de Raadt (OpenBSD) for checking it.
* John Gardner for more than a dozen suggestions regarding HTML output.
* Mike Williams for teaching me how to use %%DocumentMedia and
setpagedevice in PostScript files.
* Werner Lemberg (groff) for feedback on mdoc(7) language changes.
* Colin Watson (man-db) for feedback on man-db semantics.
* Jason McIntyre (OpenBSD) for lots of feedback and suggestions
on diagnostic messages and on the documentation.
* Thomas Klausner (NetBSD) for suggesting two new style messages
and one new feature, for two bug reports, and for release testing.
* Leah Neukirchen (Void Linux) for suggesting a new style message,
five bug reports, and release testing.
* Anthony Bentley (OpenBSD) for reporting multiple bugs and missing
features.
* Paul Irofti (OpenBSD) and Nate Bargmann for suggesting new features.
* Michael Stapelberg (Debian) for bug reports and release testing.
* Christian Weisgerber, Jonathan Gray, Stuart Henderson,
Ted Unangst (OpenBSD), Takeshi Nakayama (NetBSD),
Anton Lazarov, Jakub Klinkovsky, Jan Stary, Jesper Wallin,
Will Backmam, and Wolfgang Mueller for bug reports.
* Sevan Janiyan (NetBSD) for additions to lib.in.
* George Brown for suggesting code simplifications.
* David Coppa, Igor Sobrado (OpenBSD), and Alexander Kuleshov
for documentation improvements.
* Laura Morales and Raf Czlonka for questions resulting in better
documentation.
* Yuri Pankov (illumos) for release testing.
No upstream changelog for UCD itself, but updated for new Unicode 11.0:
http://blog.unicode.org/2018/06/announcing-unicode-standard-version-110.html.
Announcing The Unicode(R) Standard, Version 11.0
Version 11.0 of the Unicode Standard is now available, both the core
specification and data files.
Version 11.0 adds 684 characters, for a total of 137,374 characters.
These additions include seven new scripts, for a total of 146 scripts,
as well as 145 new emoji.
No quotable upstream change summary about unicode emoji, but
updated for Unicode 11.0:
http://www.unicode.org/reports/tr51/index.html
"Starting with Version 11.0 of this specification, the repertoire of
emoji characters is synchronized with the Unicode Standard, and has
the same version numbering system."
Upstream announcemnet:
http://cldr.unicode.org/index/downloads/cldr-33-1
Unicode CLDR 33.1 is an update to CLDR 33 that focuses on
Unicode 11.0 support. Improvements in this release include:
* Data
* Updates to Unicode 11.0
* Adds annotations (names and keywords) for Unicode 11.0 emoji, and makes
improvements to previously-existing annotations.
* Updates Chinese collation stroke order from Unicode 7.0 to Unicode 11.0,
after tooling bug fixes
* Structure*
* No changes. The DTD deltas and DTD Diffs links above point to v33.
* Specification*
* There is no LDML 33.1 document. Instead, only amendments to v33 are
provided, as described below in Specification Amendments.
* Charts*
* There are no charts specifically for v33.1. The link above provides
the Charts34 (currently under development). The differences from Charts33
are mostly captured in the annotations charts and in the annotations
fields of other charts such as delta for German or By Type: smileys.
For more details, see the list of bug fixes.
http://unicode.org/cldr/trac/query?resolution=fixed&milestone=33.1&group=component&max=999
Quotes argument lists to be used in Win32 in several different
situations.
Windows passes its arguments as a single string instead of an array as
other platforms do. In almost all cases, the standard Win32
CommandLineToArgvW function is used to parse this string. cmd.exe has
different rules for handling quoting, so extra work has to be done if it
is involved. It isn't possible to consistantly create a single string
that will be handled the same by cmd.exe and the stardard parsing rules.
Perl will try to detect if you need the shell by detecting shell
metacharacters. The routine that checks that uses different quoting
rules from both cmd.exe and the native Win32 parsing. Extra work must
therefore be done to protect against this autodetection.
Changes:
2.2.0
-----
- Add support for Idris, APL and J
- Add .r to R file extensions
- Add file extensions for TLA+, Isabelle, Coq
- Add support for D
- Add support for PDB
- Add support for thrift
- Add filter for devicetree source files
- Document -W/--width option in the man page
- Add support for pug
- Add support for terraform files
- Misc bug fixes and improvements
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.
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().
