]>
granicus.if.org Git - graphviz/log
Matthew Fernandez [Sat, 12 Feb 2022 03:18:29 +0000 (14:18 +1100)]
API BREAK: remove no longer used 'MAXFLOAT'
Matthew Fernandez [Sat, 12 Feb 2022 03:15:55 +0000 (14:15 +1100)]
neatogen: use '-FLT_MAX' as a sentinel instead of '-MAXFLOAT'
This seems a more correct/appropriate way of achieving this in C99.
Matthew Fernandez [Sat, 12 Feb 2022 03:13:49 +0000 (14:13 +1100)]
neatogen: replace 'MAXFLOAT' in clamping logic with 'FLT_MAX'
This seems a more accurate definition of what overflow would mean under C99.
Though as far as I can tell, it is not possible to reach this code with negative
values in the array, so the `< 0` check is always false. The `>= FLT_MAX` is
true for both `FLT_MAX` itself as well as positive infinity.
Matthew Fernandez [Sat, 12 Feb 2022 03:10:45 +0000 (14:10 +1100)]
common: [nfc] remove commented out 'MAXFLOAT' usage
Matthew Fernandez [Sat, 12 Feb 2022 03:09:55 +0000 (14:09 +1100)]
neatogen: [nfc] remove 'USE_MAXFLOAT' branch
This is never defined in the build.
Matthew Fernandez [Tue, 15 Feb 2022 03:37:17 +0000 (03:37 +0000)]
Merge branch 'smattr/
EAA625F3 -EE98-439D-B175-
F947666A46F7 ' into 'main'
fix incorrect clearing of bits in bitarray
See merge request graphviz/graphviz!2440
Matthew Fernandez [Fri, 11 Feb 2022 10:40:22 +0000 (21:40 +1100)]
fix incorrect clearing of bits in bitarray
The wrong bitwise operations were used here, resulting in e.g. a set of bit 2 of
of `
0b011000 ` updating to `
0b011100 ` instead of being a no-op.
This bug was present since the first introduction of this interface in
3ced90bd756bb8e56920620f2686e0ba6b482bbb , but this has not yet made it into a
release.
Matthew Fernandez [Thu, 10 Feb 2022 08:45:34 +0000 (19:45 +1100)]
add unit tests for bitarray.h
This demonstrates a current bug in bit clearing.
Matthew Fernandez [Mon, 14 Feb 2022 21:18:53 +0000 (21:18 +0000)]
Merge branch 'smattr/
3680D475 -238E-4402-AFF9-
1B68D2860757 ' into 'main'
CI: fix missing 'OS_ID' env var in ctest jobs
See merge request graphviz/graphviz!2436
Matthew Fernandez [Wed, 9 Feb 2022 05:55:00 +0000 (16:55 +1100)]
CI: fix missing 'OS_ID' env var in ctest jobs
Jobs inheriting from this template were attempting to set `$OS_ID` from the file
OS_ID that did not exist before `install-package.sh` was run. The result was
output in the build log like:
export OS_ID=$( cat OS_ID )
cat: OS_ID: No such file or directory
Somewhat surprisingly, tests were still passing. The reason is that `$OS_ID` is
only accessed in ci/tests.py and on Ubuntu (the platform used for both ctest
jobs) the only `$OS_ID`-based skip is for `mingle`, that is also coincidentally
skipped due to these jobs using the CMake build system where `mingle` is not
build. That is, the mingle check is meant to be skipped on lines 104-108, but
actually ends up skipped on lines 116-122 due to `$OS_ID` being empty.
In future, all of this should be refactored to use a less brittle technique for
asking “what distro are we running on?” When moving to Python 3.10,
`platform.freedesktop_os_release()` would be the obvious candidate.
Matthew Fernandez [Wed, 9 Feb 2022 05:53:46 +0000 (16:53 +1100)]
CI: shell escape echoed output during Windows build
Makes it slightly easier to copy and paste these commands if necessary.
Matthew Fernandez [Mon, 14 Feb 2022 13:09:13 +0000 (13:09 +0000)]
Merge branch 'smattr/
705AB4C9 -1D6E-40FF-B0A9-
3668FE37D27D ' into 'main'
CI: redirect stderr→stdout on Windows during build
See merge request graphviz/graphviz!2439
Matthew Fernandez [Thu, 10 Feb 2022 11:37:45 +0000 (22:37 +1100)]
CI: redirect stderr→stdout on Windows during build
In CI, the Windows environments run ci/build_windows.py to orchestrate
compilation. This is run under PowerShell, which has the surprising behavior of
(1) hiding stderr and (2) considering any output to stderr a signal to abort the
running program. The combined effect of this is that benign CMake logging
statements like:
message(WARNING "hello world")
abruptly terminate compilation with the baffling output:
python : CMake Warning at cmake/FindGD.cmake:52 (message):
At line:1 char:1
+ python ci/build_windows.py --build-system $env:build_system --platfor ...
+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ CategoryInfo : NotSpecified: (CMake Warning a...e:52
(message)::String) [], RemoteException
+ FullyQualifiedErrorId : NativeCommandError
This behavior appears to be configurable, but it seems simpler to work around
this by avoiding any stderr output. The exit status of any build commands is
sufficient to signal failure.
Gitlab: #2097
Matthew Fernandez [Sun, 13 Feb 2022 20:16:34 +0000 (20:16 +0000)]
Merge branch 'smattr/
BA5AA4C3 -8A8A-42E9-A3C9-
F0357D8D7FE9 ' into 'main'
upgrade Pytest dependency
See merge request graphviz/graphviz!2444
Matthew Fernandez [Sat, 12 Feb 2022 23:31:17 +0000 (10:31 +1100)]
upgrade Pytest dependency
This upgrade brings with it a number of bug fixes for running with Python ≥ 3.8.
Matthew Fernandez [Sat, 12 Feb 2022 21:10:08 +0000 (21:10 +0000)]
Merge branch 'smattr/
E8A2F5A2 -2C7D-453C-8D7E-
A309515C594E ' into 'main'
add a test case for #1786
Closes #1786
See merge request graphviz/graphviz!2435
Matthew Fernandez [Tue, 8 Feb 2022 03:14:02 +0000 (14:14 +1100)]
add a test case for #1786
Matthew Fernandez [Tue, 8 Feb 2022 04:15:27 +0000 (15:15 +1100)]
fix 'dot' test helper failing to encode string input
When asking the `dot` helper function to process a graph from inline input to a
binary output format, the `universal_newlines=True` parameter would not be
passed to the underlying `subprocess.check_output` call, resulting in failure to
encode the input. That is, something like the following:
dot("png:gd", source="digraph { a -> b; }")
would result in:
self = <subprocess.Popen object at 0x7fbb862ae390>
def _communicate(self, input, endtime, orig_timeout):
…
> input_view = memoryview(self._input)
E TypeError: memoryview: a bytes-like object is required, not 'str'
/usr/lib64/python3.6/subprocess.py:1519: TypeError
Note that this problem was latent, as no usage fitting the above scenario
currently exists. Something like this will be introduced in a future commit.
Gitlab: related to #1786
Matthew Fernandez [Sat, 12 Feb 2022 19:58:25 +0000 (19:58 +0000)]
Merge branch 'smattr/
5DC1591E -4531-4D4F-AA8A-
BD596A1D2C0B ' into 'main'
sfdpgen: some clean up
See merge request graphviz/graphviz!2437
Matthew Fernandez [Thu, 10 Feb 2022 04:43:50 +0000 (15:43 +1100)]
sfdpgen: remove dead store of 'DistanceMatrix_restrict_matching' result
The returned value is immediately overwritten after being stored.
Matthew Fernandez [Thu, 10 Feb 2022 04:43:11 +0000 (15:43 +1100)]
sfdpgen: [nfc] remove DistanceMatrix_restrict_cluster no-op
Matthew Fernandez [Thu, 10 Feb 2022 04:27:43 +0000 (15:27 +1100)]
sfdpgen: [nfc] remove duplicated comment
Matthew Fernandez [Thu, 10 Feb 2022 04:27:13 +0000 (15:27 +1100)]
sfdpgen: [nfc] remove unnecessary casts
Matthew Fernandez [Wed, 9 Feb 2022 07:08:32 +0000 (07:08 +0000)]
Merge branch 'smattr/
14DF9C3A -34FA-4052-AAE4-
D6D57F7D1D11 ' into 'main'
enable 'dot_builtins' in RPM package
Closes #1839
See merge request graphviz/graphviz!2434
Matthew Fernandez [Mon, 7 Feb 2022 06:49:36 +0000 (17:49 +1100)]
enable 'dot_builtins' in RPM package
It is not clear why this was excluded. This partially reverts
e5f514b283fa5fe8690f7d82fe7555e8ce74e371 .
Gitlab: closes #1839
Matthew Fernandez [Mon, 7 Feb 2022 07:44:11 +0000 (18:44 +1100)]
CI: on RHEL distros, install all RPMs
Commit
c334e4e81aad15d3092d7c25038a4d2cf1cbff05 altered this step to only
install RPMs necessary for the test suite at the time. However, an upcoming
commit that enables `dot_builtins` causes graphviz-nox-* to depend on
libgvplugin_webp.so.6 that was not being installed. For simplicity, the present
commit reverts
c334e4e81aad15d3092d7c25038a4d2cf1cbff05 .
Note that this more closely matches the only reasonable way to use the
CI-produced RPMs: install all of them.
Gitlab: related to #1839
Matthew Fernandez [Mon, 7 Feb 2022 10:05:58 +0000 (21:05 +1100)]
CI: install R on Fedora
An upcoming commit will install the Graphviz R language bindings. These depend
on the `R` package that is not a dependency of any other currently installed
package.
Gitlab: related to #1839
Matthew Fernandez [Mon, 7 Feb 2022 10:03:26 +0000 (21:03 +1100)]
CI: install Perl on Fedora
An upcoming commit will install the Graphviz Perl language bindings. These
depend on the `perl` package that is not a dependency of any other currently
installed package.
Gitlab: related to #1839
Matthew Fernandez [Tue, 8 Feb 2022 03:32:15 +0000 (03:32 +0000)]
Merge branch 'smattr/
1301EC82 -E002-4551-8386-
BB449F2816D6 ' into 'main'
API BREAK: fix: typedef 'ssize_t' to 'SSIZE_T' on Windows
Closes #1804
See merge request graphviz/graphviz!2433
Matthew Fernandez [Mon, 7 Feb 2022 05:29:27 +0000 (16:29 +1100)]
API BREAK: fix: typedef 'ssize_t' to 'SSIZE_T' on Windows
In several places, code was using `int` as a drop-in replacement for `ssize_t`
on Windows where it does not exist. This is incorrect on some platforms. E.g. on
x86-64 this will result in `ssize_t` being a 32-bit type instead of a 64-bit
type. This change replaces it with the correct Windows equivalent, `SSIZE_T`.¹
¹ https://docs.microsoft.com/en-us/windows/win32/winprog/windows-data-types?redirectedfrom=MSDN#ssize_t
Gitlab: fixes #1804
Matthew Fernandez [Mon, 7 Feb 2022 08:52:43 +0000 (08:52 +0000)]
Merge branch 'smattr/
4F4E0C69 -AA2D-473F-A930-
83FC0A256AAF ' into 'main'
some non-functional changes for simplification/cleanup
See merge request graphviz/graphviz!2432
Matthew Fernandez [Sun, 6 Feb 2022 06:51:16 +0000 (17:51 +1100)]
netogen: [nfc] rewrite 'push' and 'pop' macros as functions
There is no need for these to be macros. Modern compilers will easily inline
these helpers. Making them functions provides additional type safety.
Matthew Fernandez [Sun, 6 Feb 2022 06:51:26 +0000 (17:51 +1100)]
neatogen: [nfc] remove unused 'read_top' macro
This macro also appears to be incorrect; it does not use its parameter `h` but
instead uses `s`.
Matthew Fernandez [Sun, 6 Feb 2022 06:19:35 +0000 (17:19 +1100)]
grammar.y startgraph: [nfc] remove a static local
A simpler way of achieving the warning squashing the comment here is implying
(uninitialized struct members) is to avoid the use of a static variable
altogether. Beyond being simpler, this explains to the compiler the contents of
this variable do not need to be preserved across calls and gets this function
closer to being thread safe.
Matthew Fernandez [Sun, 6 Feb 2022 06:03:48 +0000 (17:03 +1100)]
grammar.y pop: [nfc] fix incorrect pop from global stack
This function was incorrectly referencing the global stack, `S`, instead of the
stack passed in. This change is marked NFC because `pop` is only ever called
with `S` as a parameter, but it still seemed worth correcting.
Matthew Fernandez [Sun, 6 Feb 2022 21:39:02 +0000 (21:39 +0000)]
Merge branch 'smattr/
886DC5C9 -BF2B-47F2-9E00-
7083AAB2423F ' into 'main'
lib/pack cleanup
See merge request graphviz/graphviz!2430
Matthew Fernandez [Sat, 5 Feb 2022 06:19:57 +0000 (17:19 +1100)]
gml2gv: [nfc] remove unnecessary include
Matthew Fernandez [Sat, 5 Feb 2022 07:14:05 +0000 (18:14 +1100)]
lib/pack: [nfc] remove duplicate include
Matthew Fernandez [Sat, 5 Feb 2022 07:13:52 +0000 (18:13 +1100)]
lib/pack: [nfc] remove open coded NULLs
Matthew Fernandez [Sat, 5 Feb 2022 06:56:44 +0000 (17:56 +1100)]
lib/pack: [nfc] remove unnecessary parens
Matthew Fernandez [Sat, 5 Feb 2022 06:43:55 +0000 (17:43 +1100)]
insertFn: [nfc] remove unnecessary cast
Magnus Jacobsson [Sun, 6 Feb 2022 18:59:48 +0000 (18:59 +0000)]
Merge branch 'upgrade-to-macos-12-xcode-13' into 'main'
Upgrade to MacOS 12, Xcode 13
See merge request graphviz/graphviz!2370
Magnus Jacobsson [Mon, 27 Dec 2021 16:46:00 +0000 (17:46 +0100)]
CI: upgrade to macos-12-xcode-13
Matthew Fernandez [Sat, 5 Feb 2022 06:08:32 +0000 (06:08 +0000)]
Merge branch 'smattr/
F06FAC81 -75F7-4E90-9AA8-
368F7D31E8D8 ' into 'main'
cmd/tools: [nfc] some minor cleanup
See merge request graphviz/graphviz!2429
Matthew Fernandez [Sat, 5 Feb 2022 01:34:19 +0000 (12:34 +1100)]
gxl2gv: [nfc] remove the need to cast 'mapLookup' parameter
The equivalent of the prior commit on gxl2gv.
Matthew Fernandez [Sat, 5 Feb 2022 01:32:53 +0000 (12:32 +1100)]
graphml2gv: [nfc] remove the need to cast 'mapLookup' parameter
Matthew Fernandez [Sat, 5 Feb 2022 01:30:21 +0000 (12:30 +1100)]
graphml2gv: [nfc] remove the need to cast 'isAnonGraph' parameter
Equivalent of
3e09cc381e53ece08529dd35d27a7a436dd9ab0f on graphml2gv.
Matthew Fernandez [Fri, 4 Feb 2022 09:53:50 +0000 (09:53 +0000)]
Merge branch 'smattr/
729B6BDD -7A42-455A-8B12-
AA2F3A929663 ' into 'main'
add a test case for #2092
See merge request graphviz/graphviz!2427
Matthew Fernandez [Thu, 3 Feb 2022 04:06:12 +0000 (15:06 +1100)]
dot2gxl: de-dupe 'gcalloc' implementation
Matthew Fernandez [Thu, 3 Feb 2022 03:55:57 +0000 (14:55 +1100)]
dot2gxl: [nfc] remove the need to cast 'isAnonGraph' parameter
Matthew Fernandez [Thu, 3 Feb 2022 01:02:06 +0000 (12:02 +1100)]
add a test case for #2092
Matthew Fernandez [Fri, 4 Feb 2022 08:00:00 +0000 (08:00 +0000)]
Merge branch 'smattr/
E428EEE4 -08B8-472D-B3B6-
87C93C5D321E ' into 'main'
minor test clean up
See merge request graphviz/graphviz!2425
Matthew Fernandez [Tue, 1 Feb 2022 10:26:56 +0000 (21:26 +1100)]
test 1594: [nfc] tell 'subprocess' to discard stdout
This test case was capturing stdout and then discarding it. It is more efficient
to just attach stdout directly to /dev/null.
Matthew Fernandez [Tue, 1 Feb 2022 10:23:20 +0000 (21:23 +1100)]
test 1594: remove work around for #1780
Issue #1780 was fixed in
a286e358006dd49d2b3348abd8a2aebcffd24dec .
Matthew Fernandez [Fri, 4 Feb 2022 06:56:16 +0000 (06:56 +0000)]
Merge branch 'smattr/
650B3501 -D9F6-4EDA-BE4C-
1C8EA4C2283F ' into 'main'
add a test case for #2179
See merge request graphviz/graphviz!2426
Matthew Fernandez [Wed, 2 Feb 2022 11:50:47 +0000 (22:50 +1100)]
textspan.c: [nfc] remove unnecessary casts
Matthew Fernandez [Tue, 1 Feb 2022 10:21:05 +0000 (21:21 +1100)]
add a test case for #2179
Matthew Fernandez [Tue, 1 Feb 2022 10:19:29 +0000 (21:19 +1100)]
test 1449: [nfc] do not capture stdout
Graphviz is being passed `-o /dev/null` in this test case, so there is no need
to capture stdout.
Matthew Fernandez [Thu, 3 Feb 2022 07:22:07 +0000 (07:22 +0000)]
Merge branch 'smattr/
629BA038 -E6B6-4140-8BA3-
CFB49323FE14 -2' into 'main'
CMake: fix CMAKE_LIBRARY_PATH is not honored
Closes #1973
See merge request graphviz/graphviz!2422
Matthew Fernandez [Thu, 25 Nov 2021 16:16:36 +0000 (08:16 -0800)]
CMake: use more standard, portable mechanism for setting install paths
Fixes #1973.
Suggested-by: Satadru Pramanik
Matthew Fernandez [Thu, 25 Nov 2021 17:15:24 +0000 (09:15 -0800)]
increase CMake log-level in CI
Allows us to more easily debug failures.
Matthew Fernandez [Sun, 30 Jan 2022 23:54:46 +0000 (10:54 +1100)]
CMake: add a variable substitution step for plugin configuration step
This CMake script hard codes some library paths in a way that thwarts
user/system preferences (#1973). Attempts to solve this by falling back on
standard CMake mechanisms have been unsuccessful because an install script like
this runs in an isolated environment where it (correctly) does not have access
to the configuration from the build environment. To resolve this we need a way
to pass the install script parameters/arguments from the build context.
Introducing a level of indirection here gives us such a mechanism.
This change by itself does nothing. However an upcoming change will take
advantage of the `@…@` substitution mechanism enabled by this change.
Note that the word “configure” in these files is used to refer to two distinct
mechanisms:
1. The CMake `configure_file` command does something analogous to Autotools’
variable substitution during `./configure`.
2. The Graphviz `dot -c …` configure step registers installed Graphviz
plugins.
Gitlab: #1973
Matthew Fernandez [Wed, 2 Feb 2022 12:55:51 +0000 (12:55 +0000)]
Merge branch 'smattr/
A8790232 -0AB5-4072-B45F-
50CCECF99A65 ' into 'main'
remove support for '-m' command-line option
See merge request graphviz/graphviz!2424
Matthew Fernandez [Tue, 1 Feb 2022 07:24:56 +0000 (18:24 +1100)]
remove support for '-m' command-line option
As described in the changelog entry, LSan and friends are a more effective way
of performing this testing now.
Matthew Fernandez [Wed, 2 Feb 2022 07:48:28 +0000 (07:48 +0000)]
Merge branch 'smattr/
224C0C90 -0ABD-44D2-9D3F-
DCAD4702E26D ' into 'main'
fix Gvedit `-?` and enable building in CI on macOS
See merge request graphviz/graphviz!2419
Matthew Fernandez [Wed, 26 Jan 2022 09:07:04 +0000 (20:07 +1100)]
CI: add installation of Qt5 on macOS
This is a dependency for building Gvedit on macOS. Note that installation is not
added to the CMake test job because Gvedit is not currently built in the CMake
build system on macOS.
Gitlab: #1842, #1854
Matthew Fernandez [Thu, 27 Jan 2022 22:08:03 +0000 (09:08 +1100)]
fix: teach Gvedit `-?` on macOS
This is the equivalent of
aa695fabf4ff342488229aabd43c0970470d711f and friends,
but for Gvedit. It was not previously detected that this problem also affects
Gvedit because it is not currently built on macOS in CI. That will be improved
in an upcoming commit.
Gitlab: #1842, #1852
Matthew Fernandez [Wed, 2 Feb 2022 06:45:28 +0000 (06:45 +0000)]
Merge branch 'smattr/
3D701FC9 -FA35-4BB3-BD91-
3C6492C688BF ' into 'main'
CI: add a job for enforcing CMake file formatting
See merge request graphviz/graphviz!2423
Matthew Fernandez [Mon, 31 Jan 2022 02:34:12 +0000 (13:34 +1100)]
CI: add a job for enforcing CMake file formatting
This should hopefully reduce the mental load on MR reviewers going forwards.
Matthew Fernandez [Tue, 1 Feb 2022 04:10:18 +0000 (15:10 +1100)]
CMake: [nfc] add a leading comment to a macro
Squashes a “[C0111] Missing docstring on function or macro declaration”
`cmake-lint` warning.
Matthew Fernandez [Tue, 1 Feb 2022 03:55:26 +0000 (14:55 +1100)]
CMake: [nfc] remove excess blank line
Squashes a “[C0305] too many newlines between statements” `cmake-lint` warning.
Matthew Fernandez [Tue, 1 Feb 2022 03:55:01 +0000 (14:55 +1100)]
CMake: [nfc] uppercase 'dotlink' variable
Upcoming changes that introduce `cmake-format` checks complain that this
identifier does not comply with `[A-Z][0-9A-Z_]+|_[0-9a-z_]+`
Matthew Fernandez [Tue, 1 Feb 2022 03:54:13 +0000 (14:54 +1100)]
CMake: [nfc] uppercase 'dotcopy' variable
Upcoming changes that introduce `cmake-format` checks complain that this
identifier does not comply with `[A-Z][0-9A-Z_]+|_[0-9a-z_]+`
Matthew Fernandez [Tue, 1 Feb 2022 03:51:59 +0000 (14:51 +1100)]
CMake: [nfc] uppercase 'sources' variable
Upcoming changes that introduce `cmake-format` checks complain that this
identifier does not comply with `[A-Z][0-9A-Z_]+|_[0-9a-z_]+`
Matthew Fernandez [Tue, 1 Feb 2022 03:50:45 +0000 (14:50 +1100)]
CMake: add comments to 'add_custom_command', 'add_custom_target' steps
Squashes “[C0113] Missing COMMENT in statement which allows it” `cmake-lint`
warnings.
Matthew Fernandez [Tue, 1 Feb 2022 03:23:54 +0000 (14:23 +1100)]
CMake: [nfc] uppercase 'create_test' macro
Upcoming changes that introduce `cmake-format` checks complain that this
identifier does not comply with `[0-9A-Z_]+`.
Matthew Fernandez [Tue, 1 Feb 2022 02:41:36 +0000 (13:41 +1100)]
CMake: [nfc] wrap to 80 character column width
Upcoming changes to introduce `cmake-format` checks default to this, which suits
Graphviz well as it matches the (claimed) `clang-format` C/C++ style.
Matthew Fernandez [Tue, 1 Feb 2022 02:39:57 +0000 (13:39 +1100)]
CMake: [nfc] fix a comment typo
Matthew Fernandez [Tue, 1 Feb 2022 01:27:04 +0000 (12:27 +1100)]
CMake: [nfc] standardize on 2-space indentation
Upcoming changes to introduce `cmake-format` checks default to this, which suits
Graphviz well as it matches the (claimed) `clang-format` C/C++ style.
Matthew Fernandez [Tue, 1 Feb 2022 01:25:33 +0000 (12:25 +1100)]
CMake: [nfc] uppercase 'PangoCairo_*' variables
Upcoming changes that introduce `cmake-format` checks complain that these
identifiers do not comply with `[A-Z][0-9A-Z_]+|_[0-9a-z_]+`.
Matthew Fernandez [Tue, 1 Feb 2022 01:05:53 +0000 (12:05 +1100)]
CMake: [nfc] uppercase 'Svgpp_*' variables
Upcoming changes that introduce `cmake-format` checks complain that these
identifiers do not comply with `[A-Z][0-9A-Z_]+|_[0-9a-z_]+`.
Matthew Fernandez [Mon, 31 Jan 2022 12:00:10 +0000 (23:00 +1100)]
CMake: [nfc] uppercase 'RapidXml_*' variables
Upcoming changes that introduce `cmake-format` checks complain that these
identifiers do not comply with `[A-Z][0-9A-Z_]+|_[0-9a-z_]+`.
Matthew Fernandez [Mon, 31 Jan 2022 11:15:38 +0000 (22:15 +1100)]
CMake: [nfc] uppercase 'Getopt_*' variables
Upcoming changes that introduce `cmake-format` checks complain that these
identifiers do not comply with `[A-Z][0-9A-Z_]+|_[0-9a-z_]+`.
Matthew Fernandez [Mon, 31 Jan 2022 06:57:50 +0000 (17:57 +1100)]
CMake: [nfc] uppercase 'Cairo_*' variables
Upcoming changes that introduce `cmake-format` checks complain that these
identifiers do not comply with `[A-Z][0-9A-Z_]+|_[0-9a-z_]+`.
Magnus Jacobsson [Mon, 31 Jan 2022 18:11:55 +0000 (18:11 +0000)]
Merge branch 'fix-stdout-and-stderr-flushing-at-exit-on-mingw' into 'main'
Fix stdout and stderr flushing at exit on MinGW
Closes #2178
See merge request graphviz/graphviz!2416
Matthew Fernandez [Mon, 31 Jan 2022 11:01:03 +0000 (11:01 +0000)]
Merge branch 'smattr/
B165A187 -BED3-4E13-8876-
799331BFE43F ' into 'main'
remove unused ltmain.sh.patch
See merge request graphviz/graphviz!2420
Magnus Jacobsson [Thu, 20 Jan 2022 20:09:54 +0000 (21:09 +0100)]
add a CHANGELOG entry for the stdout/stderr not flushed fix
Magnus Jacobsson [Wed, 19 Jan 2022 21:30:40 +0000 (22:30 +0100)]
replace all instances of return from main with graphviz_exit()
Fixes https://gitlab.com/graphviz/graphviz/-/issues/2178.
Magnus Jacobsson [Sun, 16 Jan 2022 15:46:41 +0000 (16:46 +0100)]
replace all instances of exit() with graphviz_exit()
Partial fix for https://gitlab.com/graphviz/graphviz/-/issues/2178.
Magnus Jacobsson [Mon, 17 Jan 2022 22:02:03 +0000 (23:02 +0100)]
lneato: autotools: add 'lib' to include directories
An upcoming commit need this in order to be able to add an include of
cgraph/exit.h.
Towards https://gitlab.com/graphviz/graphviz/-/issues/2178.
Magnus Jacobsson [Mon, 17 Jan 2022 07:07:40 +0000 (08:07 +0100)]
lefty: autotools: add 'lib' to include directories
An upcoming commit need this in order to be able to add an include of
cgraph/exit.h.
Towards https://gitlab.com/graphviz/graphviz/-/issues/2178.
Magnus Jacobsson [Mon, 17 Jan 2022 20:18:44 +0000 (21:18 +0100)]
dotty: autotools: add 'lib' to include directories
An upcoming commit need this in order to be able to add an include of
cgraph/exit.h.
Towards https://gitlab.com/graphviz/graphviz/-/issues/2178.
Magnus Jacobsson [Mon, 17 Jan 2022 07:07:40 +0000 (08:07 +0100)]
diffimg: autotools: add 'lib' to include directories
An upcoming commit need this in order to be able to add an include of
cgraph/exit.h.
Towards https://gitlab.com/graphviz/graphviz/-/issues/2178.
Magnus Jacobsson [Sun, 16 Jan 2022 23:05:45 +0000 (00:05 +0100)]
lneato: MSBuild: add 'lib' to additional include directories
An upcoming commit need this in order to be able to add an include of
cgraph/exit.h.
Towards https://gitlab.com/graphviz/graphviz/-/issues/2178.
Magnus Jacobsson [Sun, 16 Jan 2022 23:05:21 +0000 (00:05 +0100)]
lefty: MSBuild: add 'lib' to additional include directories
An upcoming commit need this in order to be able to add an include of
cgraph/exit.h.
Towards https://gitlab.com/graphviz/graphviz/-/issues/2178.
Magnus Jacobsson [Sun, 16 Jan 2022 23:04:40 +0000 (00:04 +0100)]
dotty: MSBuild: add 'lib' to additional include directories
An upcoming commit need this in order to be able to add an include of
cgraph/exit.h.
Towards https://gitlab.com/graphviz/graphviz/-/issues/2178.
Magnus Jacobsson [Sun, 16 Jan 2022 23:06:18 +0000 (00:06 +0100)]
diffimg: MSBuild: add 'lib' to additional include directories
An upcoming commit need this in order to be able to add an include of
cgraph/exit.h.
Towards https://gitlab.com/graphviz/graphviz/-/issues/2178.
Magnus Jacobsson [Sun, 16 Jan 2022 16:10:59 +0000 (17:10 +0100)]
sparse: add a graphviz_exit function to general.h if STANDALONE
An previous commit added the same function to cgraph/exit.h, but that
library is not supposed to be used when STANDALONE is defined.
Towards https://gitlab.com/graphviz/graphviz/-/issues/2178.
Magnus Jacobsson [Sun, 16 Jan 2022 17:39:21 +0000 (18:39 +0100)]
sparse: remove useless include of cgraph/sprint.h
The last use of anything from sprint.h was removed in
https://gitlab.com/graphviz/graphviz/-/commit/
c2d3526427a7259aa656ab72c83b43232ebfd806 .
Magnus Jacobsson [Sun, 16 Jan 2022 15:40:29 +0000 (16:40 +0100)]
cgraph: add a graphviz_exit function in a new header file exit.h
Towards https://gitlab.com/graphviz/graphviz/-/issues/2178.