The ! operator would have priority over < and would transform currentTime before the
comparison. Moreso, the logic was reversed. It is supposed to return false if it's
not time yet to run the task, not the opposite.
CID 1273447.
The call to _CopyForWrite() invalidated the pointer previously assigned.
There was also an unrelated null-pointer derefence situation in the copy constructor.
Fixes#3074 and CID 610886 and CID 610887.
1) The default font color is now B_DOCUMENT_TEXT_COLOR
2) The font color menu now shows a palette
3) The font color menu now includes a "default" item, set
to B_DOCUMENT_TEXT_COLOR
4) Added a Todo mentionning it would be ideal to not save the default color
itself, but instead saving the fact the default color was used. Maybe allow the
StyleBuffer to use a Null color or something similar.
* Fix wrong nesting of loops which are supposed to skip the first
(*cookie) number of elements and those with the wrong address
family
* With a start value of first = 0 (the usual case), the old code
would always just return the first element of the addrinfo list, no
matter which address family was actually requested
The Open POSIX Testsuite can now be found at
https://github.com/haiku/open_posix_testsuite - most of the patches
can be found in "master", and the full Git history of this directory
can be found in the "from_haiku_trunk" branch (the patches there
should eventually be merged to "master", whenever someone has time
to look at it.)
Commits merged from the semi-official Git mirror of NetBSD
trunk (https://github.com/IIJ-NetBSD/netbsd-src/ - NetResolv
is in the tree at lib/libc/netresolv/).
Commit authors/messages in chronological order follow:
---------------------------------------
From: wiz <[email protected]>
Date: Mon, 9 Dec 2013 09:35:17 +0000
Subject: Fix typo ("then" instead of "than")
From: christos <[email protected]>
Date: Tue, 16 Sep 2014 01:30:00 +0000
Subject: close on exec hostaliases
From: christos <[email protected]>
Date: Tue, 11 Nov 2014 03:29:24 +0000
Subject: PR/48475: YAMAGUCHI Takanori: EDNS option broken with
unbound. struct querybuf in net/gethnamaddr.c defines MAXPACKET to 64K. This
in turn gets passed down until it reached res_nopt(..., answer, anslen), where
the size of the buffer must fit in 16 bits. Unfortunately we end up
being one more than the max so we end up sending a 0 as the size and
unbound does not like that. Instead we clip now to 64K - 1, and everyone
is happy.
XXX: Pullup to 7.
From: christos <[email protected]>
Date: Tue, 24 Feb 2015 17:57:08 +0000
Subject: belated removal of the advertising clause.
Commits merged from the semi-official Git mirror of NetBSD
trunk (https://github.com/IIJ-NetBSD/netbsd-src/ - inet
is in the tree at lib/libc/inet/).
Commit authors/messages in chronological order follow:
---------------------------------------
From: christos <[email protected]>
Date: Mon, 10 Feb 2014 16:29:30 +0000
Subject: PR/48585: Henning Petersen: Always set errno when returning NULL.
From: christos <[email protected]>
Date: Mon, 10 Feb 2014 16:30:54 +0000
Subject: remove unneeded code, and kill parens from return
Under some circumstances the pthread_rwlock protecting the name service
configuration was not initialized, and the calls to rdlock/wrlock would
silently fail.
* It overwrote the request buffer, and tried to access its previous
contents afterwards.
* It incorrectly copied sockaddrs.
* It did not initialize all members.
* It did not check for error codes from ioctl().
* It did not properly set errno in many cases, but just returned an
error right away.
* This hopefully fixes bug #12156. Adrien obviously had a bad day :-)
ImageFunctionsView:
- For function nodes, also display the start address of the function in
question. Makes it somewhat more convenient to determine which function a
function pointer actually refers to.
VariableTableModel:
- When attempting to retrieve the type for a given node, ensure
that it actually has a value node first. This might not necessarily
be the case if no appropriate type handler was found.
AddressValueNode:
- If the address node is pointing to a function, don't bother creating
a child, as there isn't really any useful information that can be displayed
for such a node besides its target address, which is already shown by the
address node anyways.
Philippe Saint-Pierre