[qt/qt/qtbase]: Summary of bulk changes made

KDE Git Services - Bulk Change <[email protected]>
Newsgroups gmane.comp.kde.cvs
Message-ID <[email protected]>
Git repository change summary for qt/qt/qtbase
Pushed by mirror-service into branch '6.11'.
Changed from 4d4ab82cde1897910bb72fe9a2971f290a05ed9a to 6db4f05b6d11e84fe112b28b4dfb65b2a00ed4e7
Acknowledgement was received that this change introduces only existing code that has been pushed to another public open source repository.

This change contains the following new commits:

Git commit de5bddd1de17dfc01fbc42a9a340a2f7447b62d6 by Qt Cherry-pick Bot (on behalf of Marc Mutz) on 17/07/2026 at 08:43..
QRandomNumber: don't make fallback_fill() cold just to -Os it (cont'd)

Amends 256ff836956f9989eae8004e6a62f935964bb8e1 (picked to 6.8), which
merged incomplete (forgot to git add -p): The Clang part was missing.

Original commit message:

Use optimize("Os") (available since GCC 4.4¹) or minsize (Clang²),
using __has_attribute() to switch between them. Both don't mark the
function as [[unlikely]] and don't co-locate it with other cold
functions (like error handling or exception tables).

Amends 629f3c0277632b9e799b30f4dccbbe41ec211eef (5.10).

¹ https://gcc.gnu.org/gcc-4.4/changes.html → "A new optimize"
² https://clang.llvm.org/docs/AttributeReference.html#minsize

Pick-to: 6.8
Change-Id: I4e731d639b203c1eda40821d8609eeb616c23937
Reviewed-by: Ivan Solovev <[email protected]>
(cherry picked from commit b79323efe257801bea84ba8e041f03f80f1e23fe)
Reviewed-by: Qt Cherry-pick Bot <[email protected]>
(cherry picked from commit 49f5578b1a3f8a714bc9ed40221e45056940ca07)
https://invent.kde.org/qt/qt/qtbase/-/commit/de5bddd1de17dfc01fbc42a9a340a2f7447b62d6

Git commit bccc97c517f5b9423d559d08ffa27680cf3fcd8e by Qt Cherry-pick Bot (on behalf of Marc Mutz) on 17/07/2026 at 08:43..
QXmlStreamAttribute: fix duplicate m_namespaceUri store

Found by Claude reviewing my patch that inits m_isDefault.

Amends the start of the public history.

Pick-to: 6.8
Change-Id: Ic96f422392f6ccd36ae29d8af9fb1bb0c20026a0
Reviewed-by: Ivan Solovev <[email protected]>
(cherry picked from commit f03a6f0a11516e2ac6c4827fa219115506e2a476)
Reviewed-by: Qt Cherry-pick Bot <[email protected]>
(cherry picked from commit 78ce83871712674f50359c4dd474a0f0a930b116)
https://invent.kde.org/qt/qt/qtbase/-/commit/bccc97c517f5b9423d559d08ffa27680cf3fcd8e

Git commit 37b325e7b4a85b7bb78013aa897903792b7ea7cd by Qt Cherry-pick Bot (on behalf of Marc Mutz) on 17/07/2026 at 08:43..
QDirEntryInfo: assert the variant is not valueless_by_exception

Coverity keeps complaining about the std::visit()s in this class
potentially throwing std::bad_variant_access, even after we closed the
one real source of valueless_by_exception() (never realized in
practice) by giving QFileInfo a nothrow move constructor
(cf. eb4b67478a1a36c476ad20a90e7d5a3fd080516c).

Next try is to add asserts, which this patch provides. If this also
fails, we need to port away from std::visit and switch over index()
instead. This is causing way too much fallout to not do anything about
it.

Amends 336ff06b648e2e7419f9234e4d1018a63ac9fb5b (picked to 6.11).

Since even QFileInfo copy wouldn't throw in practice, we can pick this
to 6.11, it doesn't hinge on a nothrow-move-constructible QFileInfo.

Coverity-Id: 909813
Coverity-Id: 909676
Coverity-Id: 909674
Coverity-Id: 909672
Coverity-Id: 909666
Coverity-Id: 909663
Coverity-Id: 909661
Coverity-Id: 909660
Coverity-Id: 909656
Coverity-Id: 909650
Coverity-Id: 909648
Coverity-Id: 909645
Coverity-Id: 909640
Coverity-Id: 909639
Coverity-Id: 909637
Coverity-Id: 909636
Coverity-Id: 909633
Coverity-Id: 909632
Coverity-Id: 909629
Coverity-Id: 909628
Coverity-Id: 909627
Coverity-Id: 909626
Coverity-Id: 909624
Coverity-Id: 909622
Coverity-Id: 909621
Coverity-Id: 909618
Coverity-Id: 909616
Coverity-Id: 909615
Coverity-Id: 909614
Coverity-Id: 909613
Coverity-Id: 909612
Coverity-Id: 909611
Coverity-Id: 909598
Coverity-Id: 909597
Coverity-Id: 909596
Coverity-Id: 909593
Coverity-Id: 909591
Coverity-Id: 909590
Coverity-Id: 909589
Coverity-Id: 909588
Coverity-Id: 909585
Coverity-Id: 909582
Coverity-Id: 909580
Coverity-Id: 909579
Coverity-Id: 909576
Coverity-Id: 909572
Coverity-Id: 909568
Coverity-Id: 909566
Coverity-Id: 909561
Coverity-Id: 909559
Coverity-Id: 909557
Coverity-Id: 909555
Coverity-Id: 909553
Coverity-Id: 909550
Coverity-Id: 909546
Coverity-Id: 909544
Coverity-Id: 909542
Coverity-Id: 909539
Coverity-Id: 909534
Coverity-Id: 909533
Coverity-Id: 909532
Coverity-Id: 909526
Coverity-Id: 909522
Coverity-Id: 909517
Coverity-Id: 909516
Coverity-Id: 909515
Coverity-Id: 909514
Coverity-Id: 909513
Coverity-Id: 909510
Coverity-Id: 906992
Coverity-Id: 895839
Coverity-Id: 895838
Coverity-Id: 895835
Change-Id: I8b14c193054318c8710f3dcb6bad3093e8b56a78
Reviewed-by: Fabian Kosmale <[email protected]>
Reviewed-by: Ulf Hermann <[email protected]>
(cherry picked from commit 8bd5d7ee8f73dc70871f39e15ccaa785a077dcfc)
Reviewed-by: Qt Cherry-pick Bot <[email protected]>
(cherry picked from commit 8a383d62a212b4061b1dcc8beab420857549de5d)
https://invent.kde.org/qt/qt/qtbase/-/commit/37b325e7b4a85b7bb78013aa897903792b7ea7cd

Git commit 357a76db0d9f05d71303f4d8ea782084a326ef02 by Qt Cherry-pick Bot (on behalf of Marc Mutz) on 17/07/2026 at 08:43..
QMatrix4x4: deduplicate code between two ctors

The QGenericMatrix<N,M,float> ctor was performing exactly the same
operations as the (const float *, int cols, int rows) one, except the
former is inline and the latter is out-of-line.

While an inline function is more likely to be optimized properly,
incl. strength-reducing the multiplication in the inner loop, we can
do these optimizations to the out-of-line function manually, and
reduce code duplication (both source and executable) with very little
performance overhead. There's only one in-tree user of this contructor
(from QMatrix3x3), so I think the advantages outweight the
disadvantages here.

Amends the start of the public history, which, of course, didn't have
constructor delegation available, as this was a C++11 addition.

Change-Id: Ieb237961d1f664f633fd576985b9376cd9225ed0
Reviewed-by: Volker Hilsheimer <[email protected]>
(cherry picked from commit 44d63c50836a4752b1ff3b9015d5f31528b74b79)
Reviewed-by: Qt Cherry-pick Bot <[email protected]>
(cherry picked from commit 0af5003d17aacb29e2ed35e55aa696295864c128)
https://invent.kde.org/qt/qt/qtbase/-/commit/357a76db0d9f05d71303f4d8ea782084a326ef02

Git commit 6db4f05b6d11e84fe112b28b4dfb65b2a00ed4e7 by Qt Cherry-pick Bot (on behalf of Michael Weghorn) on 17/07/2026 at 08:43..
CUPS print dialog: Support entering custom values for PPD options

CUPS not only supports PPD options that have a fixed
set of values to choose from, but also PPD options
that allow entering a custom value.

CUPS documentation on custom options: [1]

Implement support for such custom PPD options by
adding a QLineEdit to enter custom text when a non-null
return value of ppdFindCustomOption [2] indicates that
a custom option exists.

Show that one only if the "Custom" value has been
selected for the PPD option. In that case, set
manually entered value in the print options
in QPrintPropertiesDialog::setPrinterAdvancedCupsOptions
as "Custom.<entered_value>" which matches the behavior
passed of other CUPS client application implementations
like GTK 3 applications or LibreOffice.

This commit implements passing a value manually entered
in the Qt print dialog to the CUPS print job.

So far, responsibility that the user-entered text
meets the type and minimum/maximum length supported
for the custom option is fully on the user, i.e.
the value is simply passed to CUPS as is.

[1] https://www.cups.org/doc/spec-ppd.html#OPTIONS
[2] https://www.cups.org/doc/api-ppd.html#ppdFindCustomOption

Task-number: QTBUG-137426
Pick-to: 6.8
Change-Id: Ia0d9260170a691f1e42a4e611eed86759e97dcdf
Reviewed-by: Axel Spoerl <[email protected]>
(cherry picked from commit 7d54cbd406890d9a54c58b3ec9f9cf7eb71ec9c0)
Reviewed-by: Qt Cherry-pick Bot <[email protected]>
(cherry picked from commit 50487bd929d6ee82c0d368b09213573e40617ad1)
https://invent.kde.org/qt/qt/qtbase/-/commit/6db4f05b6d11e84fe112b28b4dfb65b2a00ed4e7
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.