[Bug 252001] maintainer-update www/wt update to v4.5.0

[bugzilla-noreply at freebsd.org]

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=252001

            Bug ID: 252001
           Summary: maintainer-update www/wt update to v4.5.0
           Product: Ports & Packages
           Version: Latest
          Hardware: Any
                OS: Any
            Status: New
          Severity: Affects Only Me
          Priority: ---
         Component: Individual Port(s)
          Assignee: ports-bugs at FreeBSD.org
          Reporter: info at babaei.net

Created attachment 220759
  --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=220759&action=edit
www/wt v4.4.0 to v4.5.0 patch file

Release 4.5.0 (December 15, 2020)
Notable changes
Wt now targets C++14, and CMAKE_CXX_STANDARD defaults to "14". There appears to
be no good reason to maintain C++11 support, since every compiler that supports
enough C++11 for Wt also supports enough C++14. You should still be able to use
C++11 mode when using Wt as a library, but Wt itself no longer builds in C++11
mode.
Wt::Dbo: the Firebird backend has been marked as unmaintained. The Firebird
tests were not succeeding, likely because of an incompatibility with the
unmaintained IBPP that the Firebird backend is based on and recent versions of
Firebird.
Features and improvements
Issue #7586: The <behind-reverse-proxy> option has been deprecated in favor of
a new <trusted-proxy-config> option. It's now possible to specify the IP header
that Wt should look for using <remote-ip-header> (defaults to X-Forwarded-For).
Wt will trust all proxies listed in <trusted-proxies>. When
<behind-reverse-proxy> is set to true, the behavior stays the same as it was
before. See the library overview for more information about this option.
WFormModel now uses string comparison instead of pointer comparison for its
fields.
Issue #7666: In some cases Wt 4.4.0 would connect jsScrollVisibilityChanged
even if it was not actually being used, causing extra unnecessary requests.
Added Jenkins pipelines in the jenkins folder. We use Jenkins to test Wt at
Emweb but we would usually keep this configuration in Jenkins. Our pipelines
for Linux and FreeBSD have now been updated a bit and moved into the Wt
repository, since they may be of use to people looking to contribute to Wt.
Issue #7669: Wt now automatically uses addChild instead of addWidget when
addNew is used on a global widget, like WPopupWidget.
Bug fixes
Issue #6367: fixed issue where wthttp would no longer accept new connections
when an error occurred while accepting a connection, e.g. if the open file
limit was reached.
WDate::currentServerDate() and WTime::currentServerTime() now correctly return
the local date and time on the server instead of UTC.
Fixed visibility of OAuthProcess and OAuthAccessToken.
Fixed issues #6548 and #7749: URL parameters were being parsed more than once
for file uploads, causing a rapid accumulation of duplicate values during
uploads, and potentially causing segmentation faults. The wtwithqt example was
also updated so that it properly waits for condition variables in a loop, to
deal with spurious wakeups. Also fixed a potential use-after-free when updating
resource progress.
Fixed an issue where the HTTP Client would wait indefinitely for data when
Content-Length is zero.
WebSockets: If a bad value for server_max_window_bits or client_max_window_bits
was encountered, this would cause an uncaught exception and a crash. This is
now handled properly.
Fixed issue with DISTINCT ON in X3 based query parser: the fields listed there
would incorrectly be regarded as fields being returned.
Fixed issue #7677: when a font could not be loaded in WPdfImage, this would
throw an uncaught exception. This was changed to logging an error.
Fixed issue #7719: Wt::asString did not properly handle unsigned long.
PR #166: fixed mandelbrot example: missing call to done().
PR #167: we need to link with Boost unit_test_framework on Windows when
BOOST_DYNAMIC is ON.
Fixed an issue where WMediaPlayer would not get properly loaded when
reload-is-new-session is false, and the page is refreshed.

-- 
You are receiving this mail because:
You are the assignee for the bug.