This header-only library is accurate in decimal representation of
doubles; provides API that allows to create custom representation
- so it's possible to use custom decimal separators and grouping.
This allows to unify all corner cases: integers, numbers close to
DBL_MAX, up-rounding to the next decade.
Note that Dragonbox creates the shortest decimal representation
of the number, that is unambiguously convertible back to the same
number; thus it may hide trailing digits that are unneeded for
such conversion.
The functional changes are minimal, and beneficial:
1. Rounding numbers close to DBL_MAX now takes into account the
bEraseTrailingDecZeros argument, as it should, allowing to have
"1.8E+308" for rounding DBL_MAX to 2 decimals without trailing
zeroes, instead of previous "1.80E+308".
2. Incorrect rounding is fixed in some cases, e.g. 9.9999999999999929
rounded to 10 previously using rtl_math_DecimalPlaces_Max.
3. Representing the number in the shortest way may change display
of some printed numbers. E.g., 5th greatest double is represented
as "1.797693134862315E+308" instead of a bit longer, but giving
the same double on roundtrip, "1.7976931348623149E+308". This would
generally look better for some numbers similar to the famous 0.1,
where users would likely expect more "round" representation where
it's unambiguous (but we still truncate to 15 significant decimals
anyway - so there's no point in pretending to provide exact digits
for actual binary representation).
These are reflected in the unit tests affected by the change.
Change-Id: I05e20274a30eec499593ee3e9ec070e1269232a2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129948
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
This commit implements a WebP reader and writer for both lossless
and lossy WebP, export dialog options for selecting lossless/lossy
and quality for lossy, and various internal support for the format.
Since writing WebP to e.g. ODT documents would make those images
unreadable by previous versions with no WebP support, support
for that is explicitly disabled in GraphicFilter, to be enabled
somewhen later.
Change-Id: I9b10f6da6faa78a0bb74415a92e9f163c14685f7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/128920
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
Remove code in ucb/source/ucp/webdav-neon, and now unused external
neon.
The --with-webdav=no option is retained for now.
Change-Id: I4ce429587e3991fa82009da2f8e4a068abe36435
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/126839
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
we are loading a spreadsheet in parallel here, but the parallel threads
achievei very little actual concurrency because of heavy contention
in the SharedStringPool mutex.
So switch to a concurrent hash map. I looked at a couple of different
ones (including the Folly one), and this was the one with the simplest
resulting code.
This takes my load time from 12.5s to 8s
Change-Id: I04d6d8e11d613b510eb3bc981f3338819b7ac813
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/121717
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
...using Java 1.4 java.util.logging.Logger instead also for the last remaining
uses in reportbuilder.
(The mention in swext/mediawiki/src/THIRDPARTYLICENSEREADME.html was presumably
a leftover from 4b6ceed4a4 "swext: Wiki Publisher
does not use those apache-commons libraries".)
Change-Id: Ia0bc598fe5844ced11cae497548ec7d09453a99d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/113939
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
It's no longer used by Android Viewer and use in
the online-based Android app has already been removed
in online commit
commit 2a52d768dd61f2ef8fedccb32f015c9095915935
Date: Wed Feb 19 09:05:56 2020 +0100
android shell: Remove the 'storage framework', we have content providers.
Change-Id: I468c7121eb495eb8b1a8892f14f2c289b94b7a93
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/112766
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
We must link nss statically, including the three dylibs that normally
are loaded at run-time, because including bare dylibs in an iOS appp
on the App Store is not OK. See
https://developer.apple.com/forums/thread/125796 .
For linking the softokn3 library statically, NSS already had code,
behind NSS_STATIC_SOFTOKEN ifdefs. Introduce two more macros:
NSS_STATIC_FREEBL for the freebl library and NSS_STATIC_PKCS11 for the
nssckbi library.
Turn off parallelism for the sub-make building nss. There seems to be
race conditions or something when running simultaneous instances of
the nsinstall.py script or the nsinstall program in nss (used when
building nss for the build platform).
When cross-compiling from macOS, use python3 to run the nsinstall.py
script, as it is Python 3.
Change-Id: Idd427b5ebf21f802b3feb0d5a3d259317ba8fc67
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103106
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Tor Lillqvist <tml@collabora.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/103218
Tested-by: Jenkins
Bundled box2d with the build system of LO as a static library.
If --with-system-box2d was specified checks for instance box2d in
the system, defines SYSTEM_BOX2D and uses the library from the system.
Change-Id: Ifb05912f2acaff273b25abebafc5af1cac3afec4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/94103
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
... representation of given decimal.
Use dtoa.c from https://www.netlib.org/fp/dtoa.c to build a custom
static library that doesn't use current locale (unlike strtod from
stdlib.h). This is the implementation used by e.g. python and nss
(search for "dtoa.c" under UnpackedTarball).
To avoid name clash with the standard strtod, rename the function
to strtod_nolocale.
Size of buffer on stack in ImpSvNumberInputScan::StringToDouble is
256 characters. Logging function usage in make check, of ~124 600
invocations, the longest string was 14 characters, average being
2.1 characters. So heap allocation is unlikely in scenarios with
intensive function usage.
After std::from_chars is available in baseline compilers, external
library can be dropped, and call to strtod_nolocale replaced with
the standard function.
The artifact at https://dev-www.libreoffice.org/src/dtoa-20180411.tgz
is created with
mkdir dtoa && mkdir dtoa/src && wget https://www.netlib.org/fp/dtoa.c -O dtoa/src/dtoa.c && \
printf 'd8bab255476f39ea495c8c8ed164f9077da926e6ca7afb9ad3c56d337c4484fe dtoa/src/dtoa.c' | sha256sum -c && \
tar -c --owner=0 --group=0 --mode=go=r,u=rw --mtime='Wed, 11 Apr 2018 15:59:39 GMT' dtoa/src/dtoa.c | gzip -n > dtoa-20180411.tgz && \
printf '0082d0684f7db6f62361b76c4b7faba19e0c7ce5cb8e36c4b65fea8281e711b4 dtoa-20180411.tgz' | sha256sum -c
(where the date "Wed, 11 Apr 2018 15:59:39 GMT" is from
`wget -S https://www.netlib.org/fp/dtoa.c`
"Last-Modified: Wed, 11 Apr 2018 15:59:39 GMT" header).
Change-Id: Ia61b7678e257c4bc1ff193f3f856d611aa5c1a21
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/88854
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
CPython commit f40d4ddff3c800b3c956a5e8820aabe3aa87cddd "Closes #27979:
Remove bundled copy of libffi" causes a bit of a problem because it
turns out that libffi isn't all that stable; there's libffi.so.5 on
CentOS 6, libffi.so.6 on CentOS 7 and libffi.so.7 on
lo_daily_update_gandalf tinderbox.
So we have to bundle it in LO; it's only used on GNU/Linux currently.
CPython commit 32119e10b792ad7ee4e5f951a2d89ddbaf111cc5 "bpo-35947:
Update Windows to the current version of libffi (GH-11797)" also removes
the libffi for MSVC, so in a future python upgrade we will have to build
libffi for MSVC too.
The libffi fork for MacOSX is still in CPython git master.
(regression from b10be5d484)
Change-Id: Ibc20cf8cd3614cf9941b6970662bd930496776b2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/86493
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@cib.de>
Not quite complete yet, missing e.g. Vulkan sources, and the setup
is not completely correct either.
Change-Id: I2283bf12f0d226ff8a34554deae5a7bd69045971
This gets rid of the horrible hack in gbuild.mk to accomodate the
case-incorrect iOS platform makefiles that cannot be renamed without
upsetting git on file systems that sadly lack the case sensitivity
feature.
Keep the macro defined to IOS though.
Change-Id: I1022bfef4900da00e75fc1ccce786b20f8673234
Reviewed-on: https://gerrit.libreoffice.org/62705
Tested-by: Jenkins
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
Reviewed-by: Tor Lillqvist <tml@collabora.com>
Tested-by: Tor Lillqvist <tml@collabora.com>
Additionally I erased the remains of the old mysql C++ connector.
Also update the code where the clang plugins were unhappy after moving
the mysqlc module.
Remove mysql-cpp-connector external.
Change-Id: I771cc7afacf5012fc349afb6b36828fb0fee5a66
Reviewed-on: https://gerrit.libreoffice.org/59200
Tested-by: Jenkins
Reviewed-by: Tamás Bunth <btomi96@gmail.com>
in page number, chapter and outline numbering
in ~30 languages by integrating libnumbertext library.
- offapi: add linguistic2::NumberText
New NumberingType constants:
- ordinal indicators (1st, 2nd, 3rd...)
- cardinal number names (One, Two, Three...)
- ordinal number names (First, Second, Third...)
Note: these numberings are parts of OOXML, too.
Plain text files of Libnumbertext's language data
are installed in share/numbertext (similar to
share/fingerprint), allowing further customization.
Change-Id: I4034da0a40a8c926f14a3f591749a89a8d807d5a
Reviewed-on: https://gerrit.libreoffice.org/53313
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
We cannot silently convert user data or their used file formats
and then await them to accept it or reinstall an older version.
Let us make a soft change instead of an hard (heart) break
and avoid us to been attacked with fire and forks from our users.
This reverts commit 8d381ae8d6.
Change-Id: Ia153640935e355771acb85cf652f8fe4c21fafbb
Reviewed-on: https://gerrit.libreoffice.org/52731
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Tamás Bunth <btomi96@gmail.com>
except on windows, where gla11y will resort to python's internal xml parser,
which does not provide line numbers.
This allows gla11y to be runnable on all systems.
Change-Id: Ica4eb90f59bddfcefd783fc2ed9c8c27357e7572
Reviewed-on: https://gerrit.libreoffice.org/50115
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
This patch allows us to use our own bundled TWAINDSM.dll (from
twain-dsm project, http://www.twain.org, https://github.com/twain/twain-dsm)
on 64-bit Windows systems that don't have their own TWAIN DLL,
and without requiring users to install it from some other source.
The DLL is put into INSTALLOCATION/program along soffice.bin.
Of course, TWAIN sources still need to be 64-bit themselves (i.e.,
true 64-bit TWAIN drivers required) to be usable by 64-bit program.
Also this drops external/twain/inc/twain/twain.h, which is an old
version copied from twain-dsm, and instead uses the tarball to get
the header (required by Library_scn on both 32- and 64-bit Windows).
Change-Id: I191027f3221ce46db17f50db91c9cb9315900810
Reviewed-on: https://gerrit.libreoffice.org/46992
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>
This moves the external to gpgmepp, since that's what we _actually_
link against; plus tons of enablement for Windows build, mostly
related to linker probs integration.
There's still no good way to build a DLL with autotools, so we fall
back to gbuild manual make, see also tdf#91480
Change-Id: Ifd8217ef58536612d2389d48e343db133a13fb9c
Reviewed-on: https://gerrit.libreoffice.org/44970
Reviewed-by: Siegmund Gorr <siegmund.gorr@cib.de>
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
Tested-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
The used version is actually named mariadb_client but since 2.1.0
upstream calls it MariaDB Connector/C.
Change-Id: I51b12a9be55151919585dec91a5fc802ff4fc86b
Reviewed-on: https://gerrit.libreoffice.org/42295
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
And backport a post-release fix that fixes the Windows build.
EPUBExportFilter in writerperfect will want to use this.
Change-Id: I020cd2abfd0daa731f4079320b18a6d857874278
Reviewed-on: https://gerrit.libreoffice.org/40679
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
The only remaining difference is that in the system-xmlsec case we work
with the default key manager, not with the one that's only added by our
xmlsec patches.
This works for me for the uses I know of (see
<https://lists.freedesktop.org/archives/libreoffice/2017-February/076947.html>
for the motivation): signing and verifying of different signatures (bad
signature, good with non-trusted CA, good with trusted CA) with
software-based certificates all behave as expected.
Change-Id: If3f3e2b8373ab7397db3f98070a5a2ce51fa7c06
Reviewed-on: https://gerrit.libreoffice.org/39075
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>
prefer having nasm/yasm but if its not there warn and fallback
to disabling assembler optional goodness
Change-Id: Ib31ad81717842f743c2910d575a9ebbc279a9c79
Reviewed-on: https://gerrit.libreoffice.org/35189
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
In OOo times, there'd originally been efforts to allow building on Windows with
MinGW. Later, in LO times, this has been shifted to an attempt of cross-
compiling for Windows on Linux. That attempt can be considered abandoned, and
the relevant code rotting.
Due to this heritage, there are now three kinds of MinGW-specific code in LO:
* Code from the original OOo native Windows effort that is no longer relevant
for the LO cross-compilation effort, but has never been removed properly.
* Code from the original OOo native Windows effort that is re-purposed for the
LO cross-compilation effort.
* Code that has been added specifially for the LO cross-compilation effort.
All three kinds of code are removed.
(An unrelated, remaining use of MinGW is for --enable-build-unowinreg, utilizing
--with-mingw-cross-compiler, MINGWCXX, and MINGWSTRIP.)
Change-Id: I49daad8669b4cbe49fa923050c4a4a6ff7dda568
Reviewed-on: https://gerrit.libreoffice.org/34127
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Initial use case is to avoid creating a whole Draw document + a poppler
process for each and every PDF image we load in a document.
The MSVC patch is only to support MSVC 2013, as upstream already moved
to MSVC 2015.
Change-Id: I3c9dbac3e3de9f2e874ca4cfec0a9dd8a388b87c
Reviewed-on: https://gerrit.libreoffice.org/34022
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Embedded firebird is now version 3.0.
Supports MSVC 14.0.
Instead of fbembed, there are now two libraries: fbclient and
Engine12. fbclient is linked as fbembed before, Engine12 is loaded
at runtime from fbclient.
fb now needs system libtommath, which is supplied
as a new ExternalProject of LO.
Change-Id: I132939bdee745795b22f675e4265e9590079c45f
Reviewed-on: https://gerrit.libreoffice.org/27642
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Lionel Elie Mamane <lionel@mamane.lu>
Rene Engelhard