summaryrefslogtreecommitdiff
path: root/sw/CppunitTest_sw_odfexport2.mk
AgeCommit message (Collapse)AuthorFilesLines
2021-07-21Avoid external processes picking up instdir/program/libnspr4.soStephan Bergmann1-0/+4
...which is a problem in ASan builds, as seen with a failing CppunitTest_sd_import_tests (among lots of other similarly failing tests): > Exception in thread "main" java.lang.UnsatisfiedLinkError: /usr/lib/jvm/java-11-openjdk-11.0.11.0.9-5.fc34.x86_64/lib/libsystemconf.so: ~/lo/core/instdir/program/libnspr4.so: undefined symbol: __asan_option_detect_stack_use_after_return > at java.base/java.lang.ClassLoader$NativeLibrary.load0(Native Method) > at java.base/java.lang.ClassLoader$NativeLibrary.load(ClassLoader.java:2442) > at java.base/java.lang.ClassLoader$NativeLibrary.loadLibrary(ClassLoader.java:2498) > at java.base/java.lang.ClassLoader.loadLibrary0(ClassLoader.java:2694) > at java.base/java.lang.ClassLoader.loadLibrary(ClassLoader.java:2648) > at java.base/java.lang.Runtime.loadLibrary0(Runtime.java:830) > at java.base/java.lang.System.loadLibrary(System.java:1873) > at java.base/java.security.SystemConfigurator$1.run(SystemConfigurator.java:67) > at java.base/java.security.SystemConfigurator$1.run(SystemConfigurator.java:65) > at java.base/java.security.AccessController.doPrivileged(Native Method) > at java.base/java.security.SystemConfigurator.<clinit>(SystemConfigurator.java:65) > at java.base/java.security.Security.initialize(Security.java:208) > at java.base/java.security.Security$2.run(Security.java:93) > at java.base/java.security.Security$2.run(Security.java:91) > at java.base/java.security.AccessController.doPrivileged(Native Method) > at java.base/java.security.Security.<clinit>(Security.java:91) > at java.base/sun.security.jca.ProviderList.<init>(ProviderList.java:176) > at java.base/sun.security.jca.ProviderList$2.run(ProviderList.java:94) > at java.base/sun.security.jca.ProviderList$2.run(ProviderList.java:92) > at java.base/java.security.AccessController.doPrivileged(Native Method) > at java.base/sun.security.jca.ProviderList.fromSecurityProperties(ProviderList.java:91) > at java.base/sun.security.jca.Providers.<clinit>(Providers.java:54) > at java.base/java.security.SecureRandom.getDefaultPRNG(SecureRandom.java:264) > at java.base/java.security.SecureRandom.<init>(SecureRandom.java:219) > at java.base/java.util.UUID$Holder.<clinit>(UUID.java:101) > at java.base/java.util.UUID.randomUUID(UUID.java:147) > at org.probatron.officeotron.sessionstorage.Store.putZippedResource(Unknown Source) > at org.probatron.officeotron.CommandLineSubmission.<init>(Unknown Source) > at org.probatron.officeotron.Driver.main(Unknown Source) > warn:svl.items:1258842:1258842:svl/source/items/itempool.cxx:358: old secondary pool: EditEngineItemPool of pool: XOutdevItemPool must be empty. > ~/lo/core/test/source/bootstrapfixture.cxx:232:SdImportTest::testDocumentLayout > equality assertion failed > - Expected: 0 > - Actual : 256 > - failed to execute: sh ~/lo/core/bin/officeotron.sh ~/lo/tmp/SdImportTest__testDocumentLayout_16w8dw.tmp > ~/lo/tmp/SdImportTest__testDocumentLayout_16w8dy.tmp The solution reuses the arg-env bootstrap variable already used by test::OfficeConnection::setUp (unotest/source/cpp/officeconnection.cxx) and for that makes test::getArgumnet available outside Library_unotest. (For simplicity, it adds the relevant gb_CppunitTest_add_arguments calls unconditionally to the various *.mk files, even though the data is only used conditionally for non-_WIN32 in test::BootstrapFixture::validate in test/source/bootstrapfixture.cxx, at least for now.) Not sure what exactly started to cause this issue now for my ASan builds on Fedora 34. Change-Id: I7a4d6c0e36b94935442c6c91d5ee967fcad61763 Reviewed-on: https://gerrit.libreoffice.org/c/core/+/119317 Tested-by: Jenkins Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
2021-03-17tdf#52065 sw: revert tdf#45908: Cleaning up the tabs too earlyJustin Luth1-0/+72
... can cause loops from LO 3.5 commit 36c905d8c2874f6f984d5fbbc07784ec20c43524 The PostFormat for Center and Decimal tabstops is critical since the space they take up is unknown until the following text is evaluated. The search position in these two cases is invalid until the text position is calculated by PostFormat. PostFormat normally returns FALSE, so in almost all cases the rest of NewTabPortion completes. The function only returned early if the start of the tabstop was beyond the right margin. I assume the original intent was to avoid potentially unnecessary processing. However, returning early can cause looping, which the LO 3.5 commit "fixed" by removing everything. Well, just remove the early return and both problems are solved. Change-Id: I4afc4cce8249f90097bc1f263df1a39bc219098d Reviewed-on: https://gerrit.libreoffice.org/c/core/+/112559 Tested-by: Jenkins Reviewed-by: Justin Luth <justin_luth@sil.org> Reviewed-by: Miklos Vajna <vmiklos@collabora.com>