diff options
author | coryan <coryan@ae88bc3d-4319-0410-8dbf-d08b4c9d3795> | 2001-03-27 18:09:38 +0000 |
---|---|---|
committer | coryan <coryan@ae88bc3d-4319-0410-8dbf-d08b4c9d3795> | 2001-03-27 18:09:38 +0000 |
commit | ec103072e1328ff1871e3da62d73c6dd916c75a0 (patch) | |
tree | d259eea7edaa5d5d3f849c976339b5a8f4a79d2e | |
parent | 8a776e69ce08b8587c9891268cebb9ade70f9839 (diff) | |
download | ATCD-ec103072e1328ff1871e3da62d73c6dd916c75a0.tar.gz |
ChangeLogTag:Tue Mar 27 10:08:44 2001 Carlos O'Ryan <coryan@uci.edu>
-rw-r--r-- | ChangeLog | 242 | ||||
-rw-r--r-- | ChangeLogs/ChangeLog-02a | 242 | ||||
-rw-r--r-- | ChangeLogs/ChangeLog-03a | 242 | ||||
-rw-r--r-- | ace/Select_Reactor_Base.h | 4 |
4 files changed, 376 insertions, 354 deletions
diff --git a/ChangeLog b/ChangeLog index 40fab2aeba8..f9d5eba9853 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,7 +1,13 @@ +Tue Mar 27 10:08:44 2001 Carlos O'Ryan <coryan@uci.edu> + + * ace/Select_Reactor_Base.h: + Add missing #include for builds that define + ACE_HAS_REACTOR_NOTIFICATION_QUEUE. + Tue Mar 27 05:35:38 2001 Christopher Kohlhoff <chris@kohlhoff.com> * include/makeinclude/make_flags.bor: - Prevent PASCAL build by default with BCB5 since it is not required. + Prevent PASCAL build by default with BCB5 since it is not required. Since both VCL and non-VCL applications share a common run-time library you can simply link your VCL code with a non-PASCAL build of the ACE+TAO libraries. This behaviour can be overridden by @@ -126,20 +132,20 @@ Mon Mar 26 13:00:37 2001 Carlos O'Ryan <coryan@uci.edu> Fri Mar 23 12:40:34 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> - * apps/Gateway/Gateway/Event_Forwarding_Discriminator.cpp: Make sure - to delete the Consumer_Dispatch_Set * that we get back from - unbind(). Thanks to the ever vigilant Mike Winter - <mwinter@sonic.net> for reporting this. + * apps/Gateway/Gateway/Event_Forwarding_Discriminator.cpp: Make sure + to delete the Consumer_Dispatch_Set * that we get back from + unbind(). Thanks to the ever vigilant Mike Winter + <mwinter@sonic.net> for reporting this. Thu Mar 22 10:18:24 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> - * ace/SString.h: Make the various data members protected - rather than private so subclasses can access them. Thanks to - Martin Krumpolec <krumpo@pobox.sk> for this suggestion. + * ace/SString.h: Make the various data members protected + rather than private so subclasses can access them. Thanks to + Martin Krumpolec <krumpo@pobox.sk> for this suggestion. - * tests/README: Clarified some things about the tests. - Thanks to Eric Desamore <Eric.Desamore@idtv.bull.net> for - motivating this. + * tests/README: Clarified some things about the tests. + Thanks to Eric Desamore <Eric.Desamore@idtv.bull.net> for + motivating this. Mon Mar 26 11:03:22 2001 Balachandran <bala@cs.wustl.edu> @@ -161,9 +167,9 @@ Fri Mar 23 18:01:05 2001 Balachandran Natarajan <bala@cs.wustl.edu> Fri Mar 23 16:25:18 2001 Balachandran Natarajan <bala@cs.wustl.edu> - * tests/New_Fail_Test.cpp: Reverted the change from "Fri Mar 23 - 06:34:33 2001 Balachandran Natarajan <bala@cs.wustl.edu>". The - change fails to comipile when exceptions are on. + * tests/New_Fail_Test.cpp: Reverted the change from "Fri Mar 23 + 06:34:33 2001 Balachandran Natarajan <bala@cs.wustl.edu>". The + change fails to comipile when exceptions are on. Fri Mar 23 08:47:29 2001 Carlos O'Ryan <coryan@uci.edu> @@ -189,9 +195,9 @@ Thu Mar 22 21:33:37 2001 Chris Cleeland <cleeland_c@ociweb.com> inet_ntoa_b to fill that in. * ace/INET_Addr.i (get_host_addr): For VxWorks only, changed this - to use the preferred inet_ntoa_b. This eliminate memory leaks - under VxWorks. Thanks to Erik Johannes <ejohannes@oresis.com> for - the suggested fix. + to use the preferred inet_ntoa_b. This eliminate memory leaks + under VxWorks. Thanks to Erik Johannes <ejohannes@oresis.com> for + the suggested fix. Wed Mar 21 19:20:02 2001 Balachandran Natarajan <bala@cs.wustl.edu> @@ -441,9 +447,9 @@ Thu Mar 15 11:26:35 2001 Ossama Othman <ossama@uci.edu> * ace/SSL/SSL_SOCK_Acceptor.i (ACE_SSL_SOCK_Acceptor): Make sure the handle is set for the ACE_SSL_SOCK_Acceptor, after - initializing the underlying ACE_SOCK_Acceptor. This ensures - that OpenSSL has the proper handle. Thanks to Greg McCain - <greg.mccain@veritas.com> for suggesting this fix. + initializing the underlying ACE_SOCK_Acceptor. This ensures + that OpenSSL has the proper handle. Thanks to Greg McCain + <greg.mccain@veritas.com> for suggesting this fix. * THANKS: Added Greg to the hall of the fame. @@ -452,8 +458,8 @@ Thu Mar 15 05:46:59 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> * ace/FILE_Addr.cpp (ACE_FILE_Addr): Make sure to divide the size of the filename_ by the sizeof (ACE_TCHAR) to handle unicode - correctly. Thanks to the ever vigilant Ivan Murphy for catching - this! + correctly. Thanks to the ever vigilant Ivan Murphy for catching + this! Wed Mar 14 01:47:10 2001 Nanbor Wang <nanbor@cs.wustl.edu> @@ -571,7 +577,7 @@ Sun Mar 11 18:41:08 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> * ace/CDR_Stream.h: Updated a comment to clarify that the buffer() method just returns a pointer to the first block in the chain. - Thanks to Johnny Willemsen for pointing this out. + Thanks to Johnny Willemsen for pointing this out. Sun Mar 11 17:46:45 2001 Balachandran Natarajan <bala@cs.wustl.edu> @@ -677,7 +683,7 @@ Sat Mar 10 12:39:16 2001 Craig Rodrigues <crodrigu@bbn.com> * ace/WFMO_Reactor.h: * ace/Select_Reactor_T.h: * ace/Reactor_Impl.h: - Cleaned up comment for schedule_timer() + Cleaned up comment for schedule_timer() Sat Mar 10 07:05:16 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> @@ -689,7 +695,7 @@ Sat Mar 10 07:05:16 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> * tests/run_test.lst: Added Log_Msg_Test and the new Logging_Strategy_Test. - * tests/Log_Msg_Test.cpp: Added ACE_TEXT macros to all the ACE_DEBUG and + * tests/Log_Msg_Test.cpp: Added ACE_TEXT macros to all the ACE_DEBUG and ACE_ERROR calls. * ace/Service_Manager.{h,cpp}: Refactored the request processing @@ -759,12 +765,12 @@ Wed Mar 7 08:57:42 2001 Darrell Brunsch <brunsch@uci.edu> * ace/Asynch_IO.h: Changed use of ACE_UNIMPLEMENTED_FUNC to just defining the - assignment operators outright. The point here is to just - provide any implementation, since the default one is flagged as - causing infinite recursion. + assignment operators outright. The point here is to just + provide any implementation, since the default one is flagged as + causing infinite recursion. Thanks to David Levine for pointing out this incorrect usage of - the macro. + the macro. * ace/config-win32-msvc-5.h: * ace/config-win32-msvc-6.h: @@ -791,7 +797,7 @@ Wed Mar 7 08:20:01 2001 Chad Elliott <elliott_c@ociweb.com> * bin/ace_ld: Added new functionality to filter out the harmless symbol size - change warnings for VxWorks 5.4. + change warnings for VxWorks 5.4. * examples/Shared_Malloc/test_multiple_mallocs.cpp: * examples/Shared_Malloc/test_persistence.cpp: @@ -804,13 +810,13 @@ Wed Mar 7 06:11:52 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> * ace/Asynch_IO.h: Removed the unnecessary/improper ';' at the end of the line. Thanks to Johnny Willemsen and Chris Kohlhoff - for reporting/fixing this. + for reporting/fixing this. * ace/*_Connector.h: Clarified in the comments that the ACE_Time_Value *timeout represents the *maximum* amount of time - to wait. If the server is not running, however, connect() can - return much sooner. Thanks to Dipti Jain <dijain@hss.hns.com> - for motivating this. + to wait. If the server is not running, however, connect() can + return much sooner. Thanks to Dipti Jain <dijain@hss.hns.com> + for motivating this. Tue Mar 6 14:14:06 2001 Darrell Brunsch <brunsch@uci.edu> @@ -887,8 +893,8 @@ Tue Mar 6 10:16:31 2001 Darrell Brunsch <brunsch@uci.edu> * ace/config-win32-common.h: Removed old comment about ACE_HAS_WINSOCK2 being related to - ACE_HAS_WINNT4. Thanks to Johnny Willemsen - <jwillemsen@remedy.nl> for pointing this out. + ACE_HAS_WINNT4. Thanks to Johnny Willemsen + <jwillemsen@remedy.nl> for pointing this out. Tue Mar 6 10:13:03 2001 Carlos O'Ryan <coryan@uci.edu> @@ -911,21 +917,21 @@ Mon Mar 5 19:20:45 2001 Fuzz Master D <brunsch@uci.edu> * bin/fuzz.pl: As Ossama pointed out, the ACE_CHECK test was seeing problems in - the tao_idl files since it has uses of ACE_TRY_ENV inside - quotes. Made sure not to detect problems in this case. + the tao_idl files since it has uses of ACE_TRY_ENV inside + quotes. Made sure not to detect problems in this case. Mon Mar 5 18:58:32 2001 Fuzz Master D <brunsch@uci.edu> * bin/fuzz.pl: Fixed some false positives in the resolve_initial_reference test - and also added an ACE_CHECK test by Carlos's request. + and also added an ACE_CHECK test by Carlos's request. Mon Mar 5 18:54:42 2001 Darrell Brunsch <brunsch@uci.edu> * bin/PerlACE/Process_Unix.pm: Updated the signal detection to not pay attention to TERM and - KILL. Will print out better output when signals cause a process - to exit during most calls like Kill () or TerminateWaitKill (). + KILL. Will print out better output when signals cause a process + to exit during most calls like Kill () or TerminateWaitKill (). Mon Mar 5 13:56:33 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> @@ -1207,11 +1213,11 @@ Sun Mar 4 19:31:16 2001 Darrell Brunsch <brunsch@uci.edu> * tests/version_tests/version_tests.dsw: (removed) Moved the version_tests back into the tests/*.dsp files. The - reason they were originally moved is because it was easier to - select all the debug versions during a batch build. But now we - have pippen.pl and msvc_auto_compile.pl, so this is no longer an - issue, while this strange setup would compilicate the nightly - builds. Thus, it was standardized back. + reason they were originally moved is because it was easier to + select all the debug versions during a batch build. But now we + have pippen.pl and msvc_auto_compile.pl, so this is no longer an + issue, while this strange setup would compilicate the nightly + builds. Thus, it was standardized back. * bin/create_vt_dsps.pl: (removed) @@ -1272,8 +1278,8 @@ Sun Mar 04 03:12:17 2001 Nanbor Wang <nanbor@cs.wustl.edu> * ace/SSL/SSL_Export.h: * websvcs/lib/websvcs_export.h: These files didn't check for ACE_AS_STATIC_LIBS to determine if they should define their - export macros for static build. Thanks to Chris Kohlhoff - <chris@kohlhoff.com> for reporting this. [Bug 813] + export macros for static build. Thanks to Chris Kohlhoff + <chris@kohlhoff.com> for reporting this. [Bug 813] Sat Mar 3 19:07:31 2001 Pradeep Gore <pradeep@cs.wustl.edu> @@ -1390,24 +1396,24 @@ Fri Mar 02 19:36:22 2001 Steve Huston <shuston@riverace.com> * ace/OS.i (ACE_OS::send): (only non-Win32), if call fails and errno is EAGAIN, change it to EWOULDBLOCK. Improve portability - for platforms where EAGAIN and EWOULDBLOCK are different values - and both used, like HP-UX. Thanks to Michael Grove - <snowymike@aol.com> for reporting this problem. + for platforms where EAGAIN and EWOULDBLOCK are different values + and both used, like HP-UX. Thanks to Michael Grove + <snowymike@aol.com> for reporting this problem. * ace/ACE.i (send_i): Call ACE_OS::send, not ACE_OS::write, on HP-UX. This makes it pick up the EAGAIN->EWOULDBLOCK change. * tests/TSS_Test.cpp: Avoid doing ACE_Thread::keyfree on where ACE_HAS_PTHREADS_DRAFT4 or ACE_HAS_PTHREADS_DRAFT6 (unless - ACE_HAS_TSS_EMULATION is also defined) since ACE_Thread::keyfree - will get ENOTSUP anyway. + ACE_HAS_TSS_EMULATION is also defined) since ACE_Thread::keyfree + will get ENOTSUP anyway. Fri Mar 02 17:50:40 2001 Sharath R. Cholleti <sharath@cs.wustl.edu> * ace/QtReactor.cpp: Thanks to Stefan Scherer <stefan_scherer@sepp.de> for reporting - and fixing the problem - crashing of TAO/tests/QtTests/ client - example after moving the slider. + and fixing the problem - crashing of TAO/tests/QtTests/ client + example after moving the slider. Fri Mar 02 15:35:40 2001 Nanbor Wang <nanbor@cs.wustl.edu> @@ -1429,10 +1435,10 @@ Fri Mar 02 13:17:36 2001 Steve Huston <shuston@riverace.com> * examples/IOStream/server/iostream_server.cpp (main): Removed the call to register peer_acceptor with the reactor. The - acceptor::open call already registers for ACCEPT_MASK. Prevents - leaving a hanging READ_MASK registered after the acceptor is - destroyed when main() returns. This fixes Bugzilla # 801. Thanks - to Pim Philipse <pph@rogan-medical.com> for reporting this. + acceptor::open call already registers for ACCEPT_MASK. Prevents + leaving a hanging READ_MASK registered after the acceptor is + destroyed when main() returns. This fixes Bugzilla # 801. Thanks + to Pim Philipse <pph@rogan-medical.com> for reporting this. * THANKS: Added Pim Philipse to the Hall of Fame. @@ -1440,23 +1446,23 @@ Fri Mar 02 12:24:27 2001 Steve Huston <shuston@riverace.com> * ace/config-hpux-10.x-hpc++.h: If compiling without exceptions (+noeh) then don't set ACE_HAS_EXCEPTIONS (don't set - ACE_NEW_THROWS_EXCEPTIONS either). + ACE_NEW_THROWS_EXCEPTIONS either). * ace/Thread_Manager.cpp (join_thr): Moved the code for ACE_HAS_PTHREADS_DRAFT4 that does a pthread_detach from here to - ACE_OS::thr_join (which this function calls) so that - ACE_Thread::join also causes a pthread_detach. This is needed on - Pthreads Draft 4 (HP-UX 10.x) to properly clean up the thread - storage and avoid leaking memory. Thanks to Vladimir Kondratiev - <vladimir.kondratiev@intel.com> for reporting this and providing - excellent test cases to help resolve it quickly. + ACE_OS::thr_join (which this function calls) so that + ACE_Thread::join also causes a pthread_detach. This is needed on + Pthreads Draft 4 (HP-UX 10.x) to properly clean up the thread + storage and avoid leaking memory. Thanks to Vladimir Kondratiev + <vladimir.kondratiev@intel.com> for reporting this and providing + excellent test cases to help resolve it quickly. * ace/OS.i (ACE_OS::thr_join): Added the code that was removed from ACE_Thread_Manager::join_thr, above. * ace/Event_Handler.cpp (~ACE_Event_Handler): Add an ACE_Errno_Guard around purge_pending_notifications to prevent against an ENOTSUP - from purge_pending_notifications. + from purge_pending_notifications. Thu Mar 1 17:02:10 2001 Darrell Brunsch <brunsch@uci.edu> @@ -1464,22 +1470,22 @@ Thu Mar 1 17:02:10 2001 Darrell Brunsch <brunsch@uci.edu> * bin/PerlACE/Process_Win32.pm: Fixed up the Spawn code to use the Executable () instead of - {EXECUTABLE}, since that's how we get the adjusted name (with - .exe and stuff). And also skipped the file existance with - IgnoreExeSubDir, since IgnoreExeSubDir is often used when you - need to find the program in the path. + {EXECUTABLE}, since that's how we get the adjusted name (with + .exe and stuff). And also skipped the file existance with + IgnoreExeSubDir, since IgnoreExeSubDir is often used when you + need to find the program in the path. Thu Mar 1 15:48:10 2001 Darrell Brunsch <brunsch@uci.edu> * bin/fuzz.pl: Fuzz was being a bit to eager with the class matching in the - TRACE test. It was having problems with something like: + TRACE test. It was having problems with something like: char *foo::bar () since it thought the class was "char *foo" instead of just - "foo". + "foo". But it is feeling muuuuuuch better now. @@ -1532,8 +1538,8 @@ Thu Mar 1 10:54:23 2001 Ossama Othman <ossama@uci.edu> * include/makeinclude/wrapper_macros.GNU: Fixed all incorrect uses of CFLAGS and CCFLAGS where - preprocessor macros were defined. Preprocessor macros should be - defined in the CPPFLAGS variable. + preprocessor macros were defined. Preprocessor macros should be + defined in the CPPFLAGS variable. Define ACE_USE_RCSID in non-debug builds (i.e. debug=0). @@ -1541,7 +1547,7 @@ Thu Mar 01 07:29:58 2001 Carlos O'Ryan <coryan@uci.edu> * bin/auto_run_tests.lst: Fixed typos in the Service_Context_Manipulation test. It was - running in platforms where it shouldn't. + running in platforms where it shouldn't. Thu Mar 01 07:07:03 2001 Christopher Kohlhoff <chris@kohlhoff.com> @@ -1564,9 +1570,9 @@ Wed Feb 28 22:58:50 2001 Carlos O'Ryan <coryan@uci.edu> * html/index.html: Reference */index.html directly, instead of assuming that the - web browser will use index.html as the default file, this also - works better when browsing locally. Thanks to Alain Decamps - <Alain.Decamps@pidpa.be> for pointing this out. + web browser will use index.html as the default file, this also + works better when browsing locally. Thanks to Alain Decamps + <Alain.Decamps@pidpa.be> for pointing this out. Wed Feb 28 16:41:42 2001 Darrell Brunsch <brunsch@uci.edu> @@ -1584,40 +1590,40 @@ Wed Feb 28 15:35:59 2001 Joe Hoffert <joeh@cs.wustl.edu> * docs/bugzilla-build-setup.html: Added html file outlining steps for adding a bugzilla nightly - build. + build. Wed Feb 28 00:25:42 2001 Darrell Brunsch <brunsch@uci.edu> * bin/auto_run_tests.lst: The examples/POA/Adapter_Activator/run_test.pl test was being - run twice. In fairness to all the other tests that only get to - run once, I removed the second instance. + run twice. In fairness to all the other tests that only get to + run once, I removed the second instance. Tue Feb 27 17:31:48 2001 Steve Huston <shuston@riverace.com> * include/makeinclude/platform_hpux_aCC.GNU: Turn on -D_REENTRANT for the threads=0 build - this enables the reentrant functions which - are present, threads or not. It does not turn on threads, which - are instead enabled by -D_POSIX_C_SOURCE=199506L. + are present, threads or not. It does not turn on threads, which + are instead enabled by -D_POSIX_C_SOURCE=199506L. * ace/config-hpux-11.00.h: Added ACE_HAS_2_PARAM_ASCTIME_R_AND_CTIME_R * ace/OS.i (rand_r): Added 'defined (HPUX_11)' to the ways to get to the POSIX.1c-conformant version of rand_r. Allows clean compile - on HP-UX 11 without thread support. Also lined up some - misaligned #if/else/endif lines to make it easier to find one's - way through. + on HP-UX 11 without thread support. Also lined up some + misaligned #if/else/endif lines to make it easier to find one's + way through. * ace/OS_Dirent.inl (readdir_r): Added 'defined (HPUX_11)' to the ways to get the POSIX.1c-conformant version of readdir_r. Allows - clean compile on HP-UX 11 without thread support. + clean compile on HP-UX 11 without thread support. Tue Feb 27 11:38:38 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> * performance-tests/TCP/tcp_test.cpp (main): Zapped the non-existent '-f' option. Thanks to Mike Curtis for reporting - this. + this. * performance-tests/TCP/tcp_test.cpp (main): Check for improper bufsz. Thanks to Mike Curtis for reporting this. @@ -1637,8 +1643,8 @@ Tue Feb 27 10:23:20 2001 Balachandran Natarajan <bala@cs.wustl.edu> * Makefile: Changed the ChangeLog listing in the CONTROLLED_FILES to ChangeLog*. This should help us to avoid missing ChangeLog - entries in the beta distribution, just about every time we split - and add a ChangeLog. + entries in the beta distribution, just about every time we split + and add a ChangeLog. Mon Feb 27 07:39:16 2001 Chad Elliott <elliott_c@ociweb.com> @@ -1650,9 +1656,9 @@ Mon Feb 27 07:39:16 2001 Chad Elliott <elliott_c@ociweb.com> * ace/OS.h: * ace/config-hpux-11.00.h: Modification to remove system header macro definition conflicts. - This change does not suppress the conflicts, but does an undef - of the two conflicting macros before includeing tcp.h. Thanks - to Sam Chong <sam@i2i-inc.com> for reporting this problem. + This change does not suppress the conflicts, but does an undef + of the two conflicting macros before includeing tcp.h. Thanks + to Sam Chong <sam@i2i-inc.com> for reporting this problem. Mon Feb 26 19:01:16 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> @@ -1669,9 +1675,9 @@ Mon Feb 26 18:18:38 2001 Steve Huston <shuston@riverace.com> (ACE_SString::find (char c, int pos) const): (ACE_WString::find (const ACE_WSTRING_TYPE *s, int pos) const): If match is found, return index from start of string, not from - 'pos'. Matches the C++ string standard, which was the model for - these classes. Thanks to Glen Osterhout <gosterho@elronsw.com> - for this fix. + 'pos'. Matches the C++ string standard, which was the model for + these classes. Thanks to Glen Osterhout <gosterho@elronsw.com> + for this fix. * ace/SString.h: Clarified comments on return value for above functions. @@ -1692,12 +1698,12 @@ Mon Feb 26 12:29:45 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> * ace/ARGV.h: Clarify that the int add (const ACE_TCHAR *next_arg) method assumes ownership of <next_arg>, rather than copying it. - Thanks to Steve Hespelt <shespelt@bear.com> for reporting this. - Fixes [Bug 807]. + Thanks to Steve Hespelt <shespelt@bear.com> for reporting this. + Fixes [Bug 807]. The generated version_tests dsp didn't work (since DLL_Test_Impl - creates a library). Manually created the configurations in the - tests/DLL_Test_Impl.dsp version instead. + creates a library). Manually created the configurations in the + tests/DLL_Test_Impl.dsp version instead. Mon Feb 26 15:54:33 2001 Phil Mesnier <mesnier_p@ociweb.com> @@ -1722,19 +1728,19 @@ Mon Feb 26 12:29:45 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> * ace/ARGV.h: Clarify that the int add (const ACE_TCHAR *next_arg) method assumes ownership of <next_arg>, rather than copying it. - Thanks to Steve Hespelt <shespelt@bear.com> for reporting this. - Fixes [Bug 807]. + Thanks to Steve Hespelt <shespelt@bear.com> for reporting this. + Fixes [Bug 807]. Mon Feb 26 12:27:46 2001 Jeff Parsons <parsons@cs.wustl.edu> * ace/CDR_Stream.h: * ace/CDR_Stream.i: Added method to reset the byte order in class OuputCDR. This - will be used by a gateway to set the byte order of a reply to a - reqest it has handled to one different than the gateway's own - byte order, if necessary. It should be called while the stream - is still empty, that is, before any part of the reply, including - the header, has been marshaled. + will be used by a gateway to set the byte order of a reply to a + reqest it has handled to one different than the gateway's own + byte order, if necessary. It should be called while the stream + is still empty, that is, before any part of the reply, including + the header, has been marshaled. Mon Feb 26 12:07:00 2001 Nanbor Wang <nanbor@cs.wustl.edu> @@ -1855,15 +1861,15 @@ Mon Feb 26 00:19:28 2001 Darrell Brunsch <brunsch@uci.edu> * tests/Upgradable_RW_Test.dsp: Removed the Alpha configurations. I don't think we've actually - looked at these for over a year, and they are missing from a lot - of projects. Thankfully, they are easy to recreate if they are - needed. + looked at these for over a year, and they are missing from a lot + of projects. Thankfully, they are easy to recreate if they are + needed. * ACE-INSTALL.html: Changed the Windows NT Alpha section to mention that we no - longer have configurations out of the box for Win32/Alpha, they - must be recreated if needed. + longer have configurations out of the box for Win32/Alpha, they + must be recreated if needed. * WindozeCE/ACEApp.cpp: (removed) * WindozeCE/ACEApp.h: (removed) @@ -2032,11 +2038,11 @@ Sat Feb 24 14:07:11 2001 Darrell Brunsch <brunsch@uci.edu> * tests/Thread_Pool_Reactor_Test.cpp: Did a careful study of what this test was doing in regards to - sending wide-character strings, and made it consistent. The - most troublesome aspect was the fact that the length is encoded - in the shutdown string ("\011shutdown") which meant that I had - to send the length in characters, not in bytes. This should - really fix [Bug 537]. + sending wide-character strings, and made it consistent. The + most troublesome aspect was the fact that the length is encoded + in the shutdown string ("\011shutdown") which meant that I had + to send the length in characters, not in bytes. This should + really fix [Bug 537]. Sat Feb 24 14:37:00 2001 Craig Rodrigues <crodrigu@bbn.com> diff --git a/ChangeLogs/ChangeLog-02a b/ChangeLogs/ChangeLog-02a index 40fab2aeba8..f9d5eba9853 100644 --- a/ChangeLogs/ChangeLog-02a +++ b/ChangeLogs/ChangeLog-02a @@ -1,7 +1,13 @@ +Tue Mar 27 10:08:44 2001 Carlos O'Ryan <coryan@uci.edu> + + * ace/Select_Reactor_Base.h: + Add missing #include for builds that define + ACE_HAS_REACTOR_NOTIFICATION_QUEUE. + Tue Mar 27 05:35:38 2001 Christopher Kohlhoff <chris@kohlhoff.com> * include/makeinclude/make_flags.bor: - Prevent PASCAL build by default with BCB5 since it is not required. + Prevent PASCAL build by default with BCB5 since it is not required. Since both VCL and non-VCL applications share a common run-time library you can simply link your VCL code with a non-PASCAL build of the ACE+TAO libraries. This behaviour can be overridden by @@ -126,20 +132,20 @@ Mon Mar 26 13:00:37 2001 Carlos O'Ryan <coryan@uci.edu> Fri Mar 23 12:40:34 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> - * apps/Gateway/Gateway/Event_Forwarding_Discriminator.cpp: Make sure - to delete the Consumer_Dispatch_Set * that we get back from - unbind(). Thanks to the ever vigilant Mike Winter - <mwinter@sonic.net> for reporting this. + * apps/Gateway/Gateway/Event_Forwarding_Discriminator.cpp: Make sure + to delete the Consumer_Dispatch_Set * that we get back from + unbind(). Thanks to the ever vigilant Mike Winter + <mwinter@sonic.net> for reporting this. Thu Mar 22 10:18:24 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> - * ace/SString.h: Make the various data members protected - rather than private so subclasses can access them. Thanks to - Martin Krumpolec <krumpo@pobox.sk> for this suggestion. + * ace/SString.h: Make the various data members protected + rather than private so subclasses can access them. Thanks to + Martin Krumpolec <krumpo@pobox.sk> for this suggestion. - * tests/README: Clarified some things about the tests. - Thanks to Eric Desamore <Eric.Desamore@idtv.bull.net> for - motivating this. + * tests/README: Clarified some things about the tests. + Thanks to Eric Desamore <Eric.Desamore@idtv.bull.net> for + motivating this. Mon Mar 26 11:03:22 2001 Balachandran <bala@cs.wustl.edu> @@ -161,9 +167,9 @@ Fri Mar 23 18:01:05 2001 Balachandran Natarajan <bala@cs.wustl.edu> Fri Mar 23 16:25:18 2001 Balachandran Natarajan <bala@cs.wustl.edu> - * tests/New_Fail_Test.cpp: Reverted the change from "Fri Mar 23 - 06:34:33 2001 Balachandran Natarajan <bala@cs.wustl.edu>". The - change fails to comipile when exceptions are on. + * tests/New_Fail_Test.cpp: Reverted the change from "Fri Mar 23 + 06:34:33 2001 Balachandran Natarajan <bala@cs.wustl.edu>". The + change fails to comipile when exceptions are on. Fri Mar 23 08:47:29 2001 Carlos O'Ryan <coryan@uci.edu> @@ -189,9 +195,9 @@ Thu Mar 22 21:33:37 2001 Chris Cleeland <cleeland_c@ociweb.com> inet_ntoa_b to fill that in. * ace/INET_Addr.i (get_host_addr): For VxWorks only, changed this - to use the preferred inet_ntoa_b. This eliminate memory leaks - under VxWorks. Thanks to Erik Johannes <ejohannes@oresis.com> for - the suggested fix. + to use the preferred inet_ntoa_b. This eliminate memory leaks + under VxWorks. Thanks to Erik Johannes <ejohannes@oresis.com> for + the suggested fix. Wed Mar 21 19:20:02 2001 Balachandran Natarajan <bala@cs.wustl.edu> @@ -441,9 +447,9 @@ Thu Mar 15 11:26:35 2001 Ossama Othman <ossama@uci.edu> * ace/SSL/SSL_SOCK_Acceptor.i (ACE_SSL_SOCK_Acceptor): Make sure the handle is set for the ACE_SSL_SOCK_Acceptor, after - initializing the underlying ACE_SOCK_Acceptor. This ensures - that OpenSSL has the proper handle. Thanks to Greg McCain - <greg.mccain@veritas.com> for suggesting this fix. + initializing the underlying ACE_SOCK_Acceptor. This ensures + that OpenSSL has the proper handle. Thanks to Greg McCain + <greg.mccain@veritas.com> for suggesting this fix. * THANKS: Added Greg to the hall of the fame. @@ -452,8 +458,8 @@ Thu Mar 15 05:46:59 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> * ace/FILE_Addr.cpp (ACE_FILE_Addr): Make sure to divide the size of the filename_ by the sizeof (ACE_TCHAR) to handle unicode - correctly. Thanks to the ever vigilant Ivan Murphy for catching - this! + correctly. Thanks to the ever vigilant Ivan Murphy for catching + this! Wed Mar 14 01:47:10 2001 Nanbor Wang <nanbor@cs.wustl.edu> @@ -571,7 +577,7 @@ Sun Mar 11 18:41:08 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> * ace/CDR_Stream.h: Updated a comment to clarify that the buffer() method just returns a pointer to the first block in the chain. - Thanks to Johnny Willemsen for pointing this out. + Thanks to Johnny Willemsen for pointing this out. Sun Mar 11 17:46:45 2001 Balachandran Natarajan <bala@cs.wustl.edu> @@ -677,7 +683,7 @@ Sat Mar 10 12:39:16 2001 Craig Rodrigues <crodrigu@bbn.com> * ace/WFMO_Reactor.h: * ace/Select_Reactor_T.h: * ace/Reactor_Impl.h: - Cleaned up comment for schedule_timer() + Cleaned up comment for schedule_timer() Sat Mar 10 07:05:16 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> @@ -689,7 +695,7 @@ Sat Mar 10 07:05:16 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> * tests/run_test.lst: Added Log_Msg_Test and the new Logging_Strategy_Test. - * tests/Log_Msg_Test.cpp: Added ACE_TEXT macros to all the ACE_DEBUG and + * tests/Log_Msg_Test.cpp: Added ACE_TEXT macros to all the ACE_DEBUG and ACE_ERROR calls. * ace/Service_Manager.{h,cpp}: Refactored the request processing @@ -759,12 +765,12 @@ Wed Mar 7 08:57:42 2001 Darrell Brunsch <brunsch@uci.edu> * ace/Asynch_IO.h: Changed use of ACE_UNIMPLEMENTED_FUNC to just defining the - assignment operators outright. The point here is to just - provide any implementation, since the default one is flagged as - causing infinite recursion. + assignment operators outright. The point here is to just + provide any implementation, since the default one is flagged as + causing infinite recursion. Thanks to David Levine for pointing out this incorrect usage of - the macro. + the macro. * ace/config-win32-msvc-5.h: * ace/config-win32-msvc-6.h: @@ -791,7 +797,7 @@ Wed Mar 7 08:20:01 2001 Chad Elliott <elliott_c@ociweb.com> * bin/ace_ld: Added new functionality to filter out the harmless symbol size - change warnings for VxWorks 5.4. + change warnings for VxWorks 5.4. * examples/Shared_Malloc/test_multiple_mallocs.cpp: * examples/Shared_Malloc/test_persistence.cpp: @@ -804,13 +810,13 @@ Wed Mar 7 06:11:52 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> * ace/Asynch_IO.h: Removed the unnecessary/improper ';' at the end of the line. Thanks to Johnny Willemsen and Chris Kohlhoff - for reporting/fixing this. + for reporting/fixing this. * ace/*_Connector.h: Clarified in the comments that the ACE_Time_Value *timeout represents the *maximum* amount of time - to wait. If the server is not running, however, connect() can - return much sooner. Thanks to Dipti Jain <dijain@hss.hns.com> - for motivating this. + to wait. If the server is not running, however, connect() can + return much sooner. Thanks to Dipti Jain <dijain@hss.hns.com> + for motivating this. Tue Mar 6 14:14:06 2001 Darrell Brunsch <brunsch@uci.edu> @@ -887,8 +893,8 @@ Tue Mar 6 10:16:31 2001 Darrell Brunsch <brunsch@uci.edu> * ace/config-win32-common.h: Removed old comment about ACE_HAS_WINSOCK2 being related to - ACE_HAS_WINNT4. Thanks to Johnny Willemsen - <jwillemsen@remedy.nl> for pointing this out. + ACE_HAS_WINNT4. Thanks to Johnny Willemsen + <jwillemsen@remedy.nl> for pointing this out. Tue Mar 6 10:13:03 2001 Carlos O'Ryan <coryan@uci.edu> @@ -911,21 +917,21 @@ Mon Mar 5 19:20:45 2001 Fuzz Master D <brunsch@uci.edu> * bin/fuzz.pl: As Ossama pointed out, the ACE_CHECK test was seeing problems in - the tao_idl files since it has uses of ACE_TRY_ENV inside - quotes. Made sure not to detect problems in this case. + the tao_idl files since it has uses of ACE_TRY_ENV inside + quotes. Made sure not to detect problems in this case. Mon Mar 5 18:58:32 2001 Fuzz Master D <brunsch@uci.edu> * bin/fuzz.pl: Fixed some false positives in the resolve_initial_reference test - and also added an ACE_CHECK test by Carlos's request. + and also added an ACE_CHECK test by Carlos's request. Mon Mar 5 18:54:42 2001 Darrell Brunsch <brunsch@uci.edu> * bin/PerlACE/Process_Unix.pm: Updated the signal detection to not pay attention to TERM and - KILL. Will print out better output when signals cause a process - to exit during most calls like Kill () or TerminateWaitKill (). + KILL. Will print out better output when signals cause a process + to exit during most calls like Kill () or TerminateWaitKill (). Mon Mar 5 13:56:33 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> @@ -1207,11 +1213,11 @@ Sun Mar 4 19:31:16 2001 Darrell Brunsch <brunsch@uci.edu> * tests/version_tests/version_tests.dsw: (removed) Moved the version_tests back into the tests/*.dsp files. The - reason they were originally moved is because it was easier to - select all the debug versions during a batch build. But now we - have pippen.pl and msvc_auto_compile.pl, so this is no longer an - issue, while this strange setup would compilicate the nightly - builds. Thus, it was standardized back. + reason they were originally moved is because it was easier to + select all the debug versions during a batch build. But now we + have pippen.pl and msvc_auto_compile.pl, so this is no longer an + issue, while this strange setup would compilicate the nightly + builds. Thus, it was standardized back. * bin/create_vt_dsps.pl: (removed) @@ -1272,8 +1278,8 @@ Sun Mar 04 03:12:17 2001 Nanbor Wang <nanbor@cs.wustl.edu> * ace/SSL/SSL_Export.h: * websvcs/lib/websvcs_export.h: These files didn't check for ACE_AS_STATIC_LIBS to determine if they should define their - export macros for static build. Thanks to Chris Kohlhoff - <chris@kohlhoff.com> for reporting this. [Bug 813] + export macros for static build. Thanks to Chris Kohlhoff + <chris@kohlhoff.com> for reporting this. [Bug 813] Sat Mar 3 19:07:31 2001 Pradeep Gore <pradeep@cs.wustl.edu> @@ -1390,24 +1396,24 @@ Fri Mar 02 19:36:22 2001 Steve Huston <shuston@riverace.com> * ace/OS.i (ACE_OS::send): (only non-Win32), if call fails and errno is EAGAIN, change it to EWOULDBLOCK. Improve portability - for platforms where EAGAIN and EWOULDBLOCK are different values - and both used, like HP-UX. Thanks to Michael Grove - <snowymike@aol.com> for reporting this problem. + for platforms where EAGAIN and EWOULDBLOCK are different values + and both used, like HP-UX. Thanks to Michael Grove + <snowymike@aol.com> for reporting this problem. * ace/ACE.i (send_i): Call ACE_OS::send, not ACE_OS::write, on HP-UX. This makes it pick up the EAGAIN->EWOULDBLOCK change. * tests/TSS_Test.cpp: Avoid doing ACE_Thread::keyfree on where ACE_HAS_PTHREADS_DRAFT4 or ACE_HAS_PTHREADS_DRAFT6 (unless - ACE_HAS_TSS_EMULATION is also defined) since ACE_Thread::keyfree - will get ENOTSUP anyway. + ACE_HAS_TSS_EMULATION is also defined) since ACE_Thread::keyfree + will get ENOTSUP anyway. Fri Mar 02 17:50:40 2001 Sharath R. Cholleti <sharath@cs.wustl.edu> * ace/QtReactor.cpp: Thanks to Stefan Scherer <stefan_scherer@sepp.de> for reporting - and fixing the problem - crashing of TAO/tests/QtTests/ client - example after moving the slider. + and fixing the problem - crashing of TAO/tests/QtTests/ client + example after moving the slider. Fri Mar 02 15:35:40 2001 Nanbor Wang <nanbor@cs.wustl.edu> @@ -1429,10 +1435,10 @@ Fri Mar 02 13:17:36 2001 Steve Huston <shuston@riverace.com> * examples/IOStream/server/iostream_server.cpp (main): Removed the call to register peer_acceptor with the reactor. The - acceptor::open call already registers for ACCEPT_MASK. Prevents - leaving a hanging READ_MASK registered after the acceptor is - destroyed when main() returns. This fixes Bugzilla # 801. Thanks - to Pim Philipse <pph@rogan-medical.com> for reporting this. + acceptor::open call already registers for ACCEPT_MASK. Prevents + leaving a hanging READ_MASK registered after the acceptor is + destroyed when main() returns. This fixes Bugzilla # 801. Thanks + to Pim Philipse <pph@rogan-medical.com> for reporting this. * THANKS: Added Pim Philipse to the Hall of Fame. @@ -1440,23 +1446,23 @@ Fri Mar 02 12:24:27 2001 Steve Huston <shuston@riverace.com> * ace/config-hpux-10.x-hpc++.h: If compiling without exceptions (+noeh) then don't set ACE_HAS_EXCEPTIONS (don't set - ACE_NEW_THROWS_EXCEPTIONS either). + ACE_NEW_THROWS_EXCEPTIONS either). * ace/Thread_Manager.cpp (join_thr): Moved the code for ACE_HAS_PTHREADS_DRAFT4 that does a pthread_detach from here to - ACE_OS::thr_join (which this function calls) so that - ACE_Thread::join also causes a pthread_detach. This is needed on - Pthreads Draft 4 (HP-UX 10.x) to properly clean up the thread - storage and avoid leaking memory. Thanks to Vladimir Kondratiev - <vladimir.kondratiev@intel.com> for reporting this and providing - excellent test cases to help resolve it quickly. + ACE_OS::thr_join (which this function calls) so that + ACE_Thread::join also causes a pthread_detach. This is needed on + Pthreads Draft 4 (HP-UX 10.x) to properly clean up the thread + storage and avoid leaking memory. Thanks to Vladimir Kondratiev + <vladimir.kondratiev@intel.com> for reporting this and providing + excellent test cases to help resolve it quickly. * ace/OS.i (ACE_OS::thr_join): Added the code that was removed from ACE_Thread_Manager::join_thr, above. * ace/Event_Handler.cpp (~ACE_Event_Handler): Add an ACE_Errno_Guard around purge_pending_notifications to prevent against an ENOTSUP - from purge_pending_notifications. + from purge_pending_notifications. Thu Mar 1 17:02:10 2001 Darrell Brunsch <brunsch@uci.edu> @@ -1464,22 +1470,22 @@ Thu Mar 1 17:02:10 2001 Darrell Brunsch <brunsch@uci.edu> * bin/PerlACE/Process_Win32.pm: Fixed up the Spawn code to use the Executable () instead of - {EXECUTABLE}, since that's how we get the adjusted name (with - .exe and stuff). And also skipped the file existance with - IgnoreExeSubDir, since IgnoreExeSubDir is often used when you - need to find the program in the path. + {EXECUTABLE}, since that's how we get the adjusted name (with + .exe and stuff). And also skipped the file existance with + IgnoreExeSubDir, since IgnoreExeSubDir is often used when you + need to find the program in the path. Thu Mar 1 15:48:10 2001 Darrell Brunsch <brunsch@uci.edu> * bin/fuzz.pl: Fuzz was being a bit to eager with the class matching in the - TRACE test. It was having problems with something like: + TRACE test. It was having problems with something like: char *foo::bar () since it thought the class was "char *foo" instead of just - "foo". + "foo". But it is feeling muuuuuuch better now. @@ -1532,8 +1538,8 @@ Thu Mar 1 10:54:23 2001 Ossama Othman <ossama@uci.edu> * include/makeinclude/wrapper_macros.GNU: Fixed all incorrect uses of CFLAGS and CCFLAGS where - preprocessor macros were defined. Preprocessor macros should be - defined in the CPPFLAGS variable. + preprocessor macros were defined. Preprocessor macros should be + defined in the CPPFLAGS variable. Define ACE_USE_RCSID in non-debug builds (i.e. debug=0). @@ -1541,7 +1547,7 @@ Thu Mar 01 07:29:58 2001 Carlos O'Ryan <coryan@uci.edu> * bin/auto_run_tests.lst: Fixed typos in the Service_Context_Manipulation test. It was - running in platforms where it shouldn't. + running in platforms where it shouldn't. Thu Mar 01 07:07:03 2001 Christopher Kohlhoff <chris@kohlhoff.com> @@ -1564,9 +1570,9 @@ Wed Feb 28 22:58:50 2001 Carlos O'Ryan <coryan@uci.edu> * html/index.html: Reference */index.html directly, instead of assuming that the - web browser will use index.html as the default file, this also - works better when browsing locally. Thanks to Alain Decamps - <Alain.Decamps@pidpa.be> for pointing this out. + web browser will use index.html as the default file, this also + works better when browsing locally. Thanks to Alain Decamps + <Alain.Decamps@pidpa.be> for pointing this out. Wed Feb 28 16:41:42 2001 Darrell Brunsch <brunsch@uci.edu> @@ -1584,40 +1590,40 @@ Wed Feb 28 15:35:59 2001 Joe Hoffert <joeh@cs.wustl.edu> * docs/bugzilla-build-setup.html: Added html file outlining steps for adding a bugzilla nightly - build. + build. Wed Feb 28 00:25:42 2001 Darrell Brunsch <brunsch@uci.edu> * bin/auto_run_tests.lst: The examples/POA/Adapter_Activator/run_test.pl test was being - run twice. In fairness to all the other tests that only get to - run once, I removed the second instance. + run twice. In fairness to all the other tests that only get to + run once, I removed the second instance. Tue Feb 27 17:31:48 2001 Steve Huston <shuston@riverace.com> * include/makeinclude/platform_hpux_aCC.GNU: Turn on -D_REENTRANT for the threads=0 build - this enables the reentrant functions which - are present, threads or not. It does not turn on threads, which - are instead enabled by -D_POSIX_C_SOURCE=199506L. + are present, threads or not. It does not turn on threads, which + are instead enabled by -D_POSIX_C_SOURCE=199506L. * ace/config-hpux-11.00.h: Added ACE_HAS_2_PARAM_ASCTIME_R_AND_CTIME_R * ace/OS.i (rand_r): Added 'defined (HPUX_11)' to the ways to get to the POSIX.1c-conformant version of rand_r. Allows clean compile - on HP-UX 11 without thread support. Also lined up some - misaligned #if/else/endif lines to make it easier to find one's - way through. + on HP-UX 11 without thread support. Also lined up some + misaligned #if/else/endif lines to make it easier to find one's + way through. * ace/OS_Dirent.inl (readdir_r): Added 'defined (HPUX_11)' to the ways to get the POSIX.1c-conformant version of readdir_r. Allows - clean compile on HP-UX 11 without thread support. + clean compile on HP-UX 11 without thread support. Tue Feb 27 11:38:38 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> * performance-tests/TCP/tcp_test.cpp (main): Zapped the non-existent '-f' option. Thanks to Mike Curtis for reporting - this. + this. * performance-tests/TCP/tcp_test.cpp (main): Check for improper bufsz. Thanks to Mike Curtis for reporting this. @@ -1637,8 +1643,8 @@ Tue Feb 27 10:23:20 2001 Balachandran Natarajan <bala@cs.wustl.edu> * Makefile: Changed the ChangeLog listing in the CONTROLLED_FILES to ChangeLog*. This should help us to avoid missing ChangeLog - entries in the beta distribution, just about every time we split - and add a ChangeLog. + entries in the beta distribution, just about every time we split + and add a ChangeLog. Mon Feb 27 07:39:16 2001 Chad Elliott <elliott_c@ociweb.com> @@ -1650,9 +1656,9 @@ Mon Feb 27 07:39:16 2001 Chad Elliott <elliott_c@ociweb.com> * ace/OS.h: * ace/config-hpux-11.00.h: Modification to remove system header macro definition conflicts. - This change does not suppress the conflicts, but does an undef - of the two conflicting macros before includeing tcp.h. Thanks - to Sam Chong <sam@i2i-inc.com> for reporting this problem. + This change does not suppress the conflicts, but does an undef + of the two conflicting macros before includeing tcp.h. Thanks + to Sam Chong <sam@i2i-inc.com> for reporting this problem. Mon Feb 26 19:01:16 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> @@ -1669,9 +1675,9 @@ Mon Feb 26 18:18:38 2001 Steve Huston <shuston@riverace.com> (ACE_SString::find (char c, int pos) const): (ACE_WString::find (const ACE_WSTRING_TYPE *s, int pos) const): If match is found, return index from start of string, not from - 'pos'. Matches the C++ string standard, which was the model for - these classes. Thanks to Glen Osterhout <gosterho@elronsw.com> - for this fix. + 'pos'. Matches the C++ string standard, which was the model for + these classes. Thanks to Glen Osterhout <gosterho@elronsw.com> + for this fix. * ace/SString.h: Clarified comments on return value for above functions. @@ -1692,12 +1698,12 @@ Mon Feb 26 12:29:45 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> * ace/ARGV.h: Clarify that the int add (const ACE_TCHAR *next_arg) method assumes ownership of <next_arg>, rather than copying it. - Thanks to Steve Hespelt <shespelt@bear.com> for reporting this. - Fixes [Bug 807]. + Thanks to Steve Hespelt <shespelt@bear.com> for reporting this. + Fixes [Bug 807]. The generated version_tests dsp didn't work (since DLL_Test_Impl - creates a library). Manually created the configurations in the - tests/DLL_Test_Impl.dsp version instead. + creates a library). Manually created the configurations in the + tests/DLL_Test_Impl.dsp version instead. Mon Feb 26 15:54:33 2001 Phil Mesnier <mesnier_p@ociweb.com> @@ -1722,19 +1728,19 @@ Mon Feb 26 12:29:45 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> * ace/ARGV.h: Clarify that the int add (const ACE_TCHAR *next_arg) method assumes ownership of <next_arg>, rather than copying it. - Thanks to Steve Hespelt <shespelt@bear.com> for reporting this. - Fixes [Bug 807]. + Thanks to Steve Hespelt <shespelt@bear.com> for reporting this. + Fixes [Bug 807]. Mon Feb 26 12:27:46 2001 Jeff Parsons <parsons@cs.wustl.edu> * ace/CDR_Stream.h: * ace/CDR_Stream.i: Added method to reset the byte order in class OuputCDR. This - will be used by a gateway to set the byte order of a reply to a - reqest it has handled to one different than the gateway's own - byte order, if necessary. It should be called while the stream - is still empty, that is, before any part of the reply, including - the header, has been marshaled. + will be used by a gateway to set the byte order of a reply to a + reqest it has handled to one different than the gateway's own + byte order, if necessary. It should be called while the stream + is still empty, that is, before any part of the reply, including + the header, has been marshaled. Mon Feb 26 12:07:00 2001 Nanbor Wang <nanbor@cs.wustl.edu> @@ -1855,15 +1861,15 @@ Mon Feb 26 00:19:28 2001 Darrell Brunsch <brunsch@uci.edu> * tests/Upgradable_RW_Test.dsp: Removed the Alpha configurations. I don't think we've actually - looked at these for over a year, and they are missing from a lot - of projects. Thankfully, they are easy to recreate if they are - needed. + looked at these for over a year, and they are missing from a lot + of projects. Thankfully, they are easy to recreate if they are + needed. * ACE-INSTALL.html: Changed the Windows NT Alpha section to mention that we no - longer have configurations out of the box for Win32/Alpha, they - must be recreated if needed. + longer have configurations out of the box for Win32/Alpha, they + must be recreated if needed. * WindozeCE/ACEApp.cpp: (removed) * WindozeCE/ACEApp.h: (removed) @@ -2032,11 +2038,11 @@ Sat Feb 24 14:07:11 2001 Darrell Brunsch <brunsch@uci.edu> * tests/Thread_Pool_Reactor_Test.cpp: Did a careful study of what this test was doing in regards to - sending wide-character strings, and made it consistent. The - most troublesome aspect was the fact that the length is encoded - in the shutdown string ("\011shutdown") which meant that I had - to send the length in characters, not in bytes. This should - really fix [Bug 537]. + sending wide-character strings, and made it consistent. The + most troublesome aspect was the fact that the length is encoded + in the shutdown string ("\011shutdown") which meant that I had + to send the length in characters, not in bytes. This should + really fix [Bug 537]. Sat Feb 24 14:37:00 2001 Craig Rodrigues <crodrigu@bbn.com> diff --git a/ChangeLogs/ChangeLog-03a b/ChangeLogs/ChangeLog-03a index 40fab2aeba8..f9d5eba9853 100644 --- a/ChangeLogs/ChangeLog-03a +++ b/ChangeLogs/ChangeLog-03a @@ -1,7 +1,13 @@ +Tue Mar 27 10:08:44 2001 Carlos O'Ryan <coryan@uci.edu> + + * ace/Select_Reactor_Base.h: + Add missing #include for builds that define + ACE_HAS_REACTOR_NOTIFICATION_QUEUE. + Tue Mar 27 05:35:38 2001 Christopher Kohlhoff <chris@kohlhoff.com> * include/makeinclude/make_flags.bor: - Prevent PASCAL build by default with BCB5 since it is not required. + Prevent PASCAL build by default with BCB5 since it is not required. Since both VCL and non-VCL applications share a common run-time library you can simply link your VCL code with a non-PASCAL build of the ACE+TAO libraries. This behaviour can be overridden by @@ -126,20 +132,20 @@ Mon Mar 26 13:00:37 2001 Carlos O'Ryan <coryan@uci.edu> Fri Mar 23 12:40:34 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> - * apps/Gateway/Gateway/Event_Forwarding_Discriminator.cpp: Make sure - to delete the Consumer_Dispatch_Set * that we get back from - unbind(). Thanks to the ever vigilant Mike Winter - <mwinter@sonic.net> for reporting this. + * apps/Gateway/Gateway/Event_Forwarding_Discriminator.cpp: Make sure + to delete the Consumer_Dispatch_Set * that we get back from + unbind(). Thanks to the ever vigilant Mike Winter + <mwinter@sonic.net> for reporting this. Thu Mar 22 10:18:24 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> - * ace/SString.h: Make the various data members protected - rather than private so subclasses can access them. Thanks to - Martin Krumpolec <krumpo@pobox.sk> for this suggestion. + * ace/SString.h: Make the various data members protected + rather than private so subclasses can access them. Thanks to + Martin Krumpolec <krumpo@pobox.sk> for this suggestion. - * tests/README: Clarified some things about the tests. - Thanks to Eric Desamore <Eric.Desamore@idtv.bull.net> for - motivating this. + * tests/README: Clarified some things about the tests. + Thanks to Eric Desamore <Eric.Desamore@idtv.bull.net> for + motivating this. Mon Mar 26 11:03:22 2001 Balachandran <bala@cs.wustl.edu> @@ -161,9 +167,9 @@ Fri Mar 23 18:01:05 2001 Balachandran Natarajan <bala@cs.wustl.edu> Fri Mar 23 16:25:18 2001 Balachandran Natarajan <bala@cs.wustl.edu> - * tests/New_Fail_Test.cpp: Reverted the change from "Fri Mar 23 - 06:34:33 2001 Balachandran Natarajan <bala@cs.wustl.edu>". The - change fails to comipile when exceptions are on. + * tests/New_Fail_Test.cpp: Reverted the change from "Fri Mar 23 + 06:34:33 2001 Balachandran Natarajan <bala@cs.wustl.edu>". The + change fails to comipile when exceptions are on. Fri Mar 23 08:47:29 2001 Carlos O'Ryan <coryan@uci.edu> @@ -189,9 +195,9 @@ Thu Mar 22 21:33:37 2001 Chris Cleeland <cleeland_c@ociweb.com> inet_ntoa_b to fill that in. * ace/INET_Addr.i (get_host_addr): For VxWorks only, changed this - to use the preferred inet_ntoa_b. This eliminate memory leaks - under VxWorks. Thanks to Erik Johannes <ejohannes@oresis.com> for - the suggested fix. + to use the preferred inet_ntoa_b. This eliminate memory leaks + under VxWorks. Thanks to Erik Johannes <ejohannes@oresis.com> for + the suggested fix. Wed Mar 21 19:20:02 2001 Balachandran Natarajan <bala@cs.wustl.edu> @@ -441,9 +447,9 @@ Thu Mar 15 11:26:35 2001 Ossama Othman <ossama@uci.edu> * ace/SSL/SSL_SOCK_Acceptor.i (ACE_SSL_SOCK_Acceptor): Make sure the handle is set for the ACE_SSL_SOCK_Acceptor, after - initializing the underlying ACE_SOCK_Acceptor. This ensures - that OpenSSL has the proper handle. Thanks to Greg McCain - <greg.mccain@veritas.com> for suggesting this fix. + initializing the underlying ACE_SOCK_Acceptor. This ensures + that OpenSSL has the proper handle. Thanks to Greg McCain + <greg.mccain@veritas.com> for suggesting this fix. * THANKS: Added Greg to the hall of the fame. @@ -452,8 +458,8 @@ Thu Mar 15 05:46:59 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> * ace/FILE_Addr.cpp (ACE_FILE_Addr): Make sure to divide the size of the filename_ by the sizeof (ACE_TCHAR) to handle unicode - correctly. Thanks to the ever vigilant Ivan Murphy for catching - this! + correctly. Thanks to the ever vigilant Ivan Murphy for catching + this! Wed Mar 14 01:47:10 2001 Nanbor Wang <nanbor@cs.wustl.edu> @@ -571,7 +577,7 @@ Sun Mar 11 18:41:08 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> * ace/CDR_Stream.h: Updated a comment to clarify that the buffer() method just returns a pointer to the first block in the chain. - Thanks to Johnny Willemsen for pointing this out. + Thanks to Johnny Willemsen for pointing this out. Sun Mar 11 17:46:45 2001 Balachandran Natarajan <bala@cs.wustl.edu> @@ -677,7 +683,7 @@ Sat Mar 10 12:39:16 2001 Craig Rodrigues <crodrigu@bbn.com> * ace/WFMO_Reactor.h: * ace/Select_Reactor_T.h: * ace/Reactor_Impl.h: - Cleaned up comment for schedule_timer() + Cleaned up comment for schedule_timer() Sat Mar 10 07:05:16 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> @@ -689,7 +695,7 @@ Sat Mar 10 07:05:16 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> * tests/run_test.lst: Added Log_Msg_Test and the new Logging_Strategy_Test. - * tests/Log_Msg_Test.cpp: Added ACE_TEXT macros to all the ACE_DEBUG and + * tests/Log_Msg_Test.cpp: Added ACE_TEXT macros to all the ACE_DEBUG and ACE_ERROR calls. * ace/Service_Manager.{h,cpp}: Refactored the request processing @@ -759,12 +765,12 @@ Wed Mar 7 08:57:42 2001 Darrell Brunsch <brunsch@uci.edu> * ace/Asynch_IO.h: Changed use of ACE_UNIMPLEMENTED_FUNC to just defining the - assignment operators outright. The point here is to just - provide any implementation, since the default one is flagged as - causing infinite recursion. + assignment operators outright. The point here is to just + provide any implementation, since the default one is flagged as + causing infinite recursion. Thanks to David Levine for pointing out this incorrect usage of - the macro. + the macro. * ace/config-win32-msvc-5.h: * ace/config-win32-msvc-6.h: @@ -791,7 +797,7 @@ Wed Mar 7 08:20:01 2001 Chad Elliott <elliott_c@ociweb.com> * bin/ace_ld: Added new functionality to filter out the harmless symbol size - change warnings for VxWorks 5.4. + change warnings for VxWorks 5.4. * examples/Shared_Malloc/test_multiple_mallocs.cpp: * examples/Shared_Malloc/test_persistence.cpp: @@ -804,13 +810,13 @@ Wed Mar 7 06:11:52 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> * ace/Asynch_IO.h: Removed the unnecessary/improper ';' at the end of the line. Thanks to Johnny Willemsen and Chris Kohlhoff - for reporting/fixing this. + for reporting/fixing this. * ace/*_Connector.h: Clarified in the comments that the ACE_Time_Value *timeout represents the *maximum* amount of time - to wait. If the server is not running, however, connect() can - return much sooner. Thanks to Dipti Jain <dijain@hss.hns.com> - for motivating this. + to wait. If the server is not running, however, connect() can + return much sooner. Thanks to Dipti Jain <dijain@hss.hns.com> + for motivating this. Tue Mar 6 14:14:06 2001 Darrell Brunsch <brunsch@uci.edu> @@ -887,8 +893,8 @@ Tue Mar 6 10:16:31 2001 Darrell Brunsch <brunsch@uci.edu> * ace/config-win32-common.h: Removed old comment about ACE_HAS_WINSOCK2 being related to - ACE_HAS_WINNT4. Thanks to Johnny Willemsen - <jwillemsen@remedy.nl> for pointing this out. + ACE_HAS_WINNT4. Thanks to Johnny Willemsen + <jwillemsen@remedy.nl> for pointing this out. Tue Mar 6 10:13:03 2001 Carlos O'Ryan <coryan@uci.edu> @@ -911,21 +917,21 @@ Mon Mar 5 19:20:45 2001 Fuzz Master D <brunsch@uci.edu> * bin/fuzz.pl: As Ossama pointed out, the ACE_CHECK test was seeing problems in - the tao_idl files since it has uses of ACE_TRY_ENV inside - quotes. Made sure not to detect problems in this case. + the tao_idl files since it has uses of ACE_TRY_ENV inside + quotes. Made sure not to detect problems in this case. Mon Mar 5 18:58:32 2001 Fuzz Master D <brunsch@uci.edu> * bin/fuzz.pl: Fixed some false positives in the resolve_initial_reference test - and also added an ACE_CHECK test by Carlos's request. + and also added an ACE_CHECK test by Carlos's request. Mon Mar 5 18:54:42 2001 Darrell Brunsch <brunsch@uci.edu> * bin/PerlACE/Process_Unix.pm: Updated the signal detection to not pay attention to TERM and - KILL. Will print out better output when signals cause a process - to exit during most calls like Kill () or TerminateWaitKill (). + KILL. Will print out better output when signals cause a process + to exit during most calls like Kill () or TerminateWaitKill (). Mon Mar 5 13:56:33 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> @@ -1207,11 +1213,11 @@ Sun Mar 4 19:31:16 2001 Darrell Brunsch <brunsch@uci.edu> * tests/version_tests/version_tests.dsw: (removed) Moved the version_tests back into the tests/*.dsp files. The - reason they were originally moved is because it was easier to - select all the debug versions during a batch build. But now we - have pippen.pl and msvc_auto_compile.pl, so this is no longer an - issue, while this strange setup would compilicate the nightly - builds. Thus, it was standardized back. + reason they were originally moved is because it was easier to + select all the debug versions during a batch build. But now we + have pippen.pl and msvc_auto_compile.pl, so this is no longer an + issue, while this strange setup would compilicate the nightly + builds. Thus, it was standardized back. * bin/create_vt_dsps.pl: (removed) @@ -1272,8 +1278,8 @@ Sun Mar 04 03:12:17 2001 Nanbor Wang <nanbor@cs.wustl.edu> * ace/SSL/SSL_Export.h: * websvcs/lib/websvcs_export.h: These files didn't check for ACE_AS_STATIC_LIBS to determine if they should define their - export macros for static build. Thanks to Chris Kohlhoff - <chris@kohlhoff.com> for reporting this. [Bug 813] + export macros for static build. Thanks to Chris Kohlhoff + <chris@kohlhoff.com> for reporting this. [Bug 813] Sat Mar 3 19:07:31 2001 Pradeep Gore <pradeep@cs.wustl.edu> @@ -1390,24 +1396,24 @@ Fri Mar 02 19:36:22 2001 Steve Huston <shuston@riverace.com> * ace/OS.i (ACE_OS::send): (only non-Win32), if call fails and errno is EAGAIN, change it to EWOULDBLOCK. Improve portability - for platforms where EAGAIN and EWOULDBLOCK are different values - and both used, like HP-UX. Thanks to Michael Grove - <snowymike@aol.com> for reporting this problem. + for platforms where EAGAIN and EWOULDBLOCK are different values + and both used, like HP-UX. Thanks to Michael Grove + <snowymike@aol.com> for reporting this problem. * ace/ACE.i (send_i): Call ACE_OS::send, not ACE_OS::write, on HP-UX. This makes it pick up the EAGAIN->EWOULDBLOCK change. * tests/TSS_Test.cpp: Avoid doing ACE_Thread::keyfree on where ACE_HAS_PTHREADS_DRAFT4 or ACE_HAS_PTHREADS_DRAFT6 (unless - ACE_HAS_TSS_EMULATION is also defined) since ACE_Thread::keyfree - will get ENOTSUP anyway. + ACE_HAS_TSS_EMULATION is also defined) since ACE_Thread::keyfree + will get ENOTSUP anyway. Fri Mar 02 17:50:40 2001 Sharath R. Cholleti <sharath@cs.wustl.edu> * ace/QtReactor.cpp: Thanks to Stefan Scherer <stefan_scherer@sepp.de> for reporting - and fixing the problem - crashing of TAO/tests/QtTests/ client - example after moving the slider. + and fixing the problem - crashing of TAO/tests/QtTests/ client + example after moving the slider. Fri Mar 02 15:35:40 2001 Nanbor Wang <nanbor@cs.wustl.edu> @@ -1429,10 +1435,10 @@ Fri Mar 02 13:17:36 2001 Steve Huston <shuston@riverace.com> * examples/IOStream/server/iostream_server.cpp (main): Removed the call to register peer_acceptor with the reactor. The - acceptor::open call already registers for ACCEPT_MASK. Prevents - leaving a hanging READ_MASK registered after the acceptor is - destroyed when main() returns. This fixes Bugzilla # 801. Thanks - to Pim Philipse <pph@rogan-medical.com> for reporting this. + acceptor::open call already registers for ACCEPT_MASK. Prevents + leaving a hanging READ_MASK registered after the acceptor is + destroyed when main() returns. This fixes Bugzilla # 801. Thanks + to Pim Philipse <pph@rogan-medical.com> for reporting this. * THANKS: Added Pim Philipse to the Hall of Fame. @@ -1440,23 +1446,23 @@ Fri Mar 02 12:24:27 2001 Steve Huston <shuston@riverace.com> * ace/config-hpux-10.x-hpc++.h: If compiling without exceptions (+noeh) then don't set ACE_HAS_EXCEPTIONS (don't set - ACE_NEW_THROWS_EXCEPTIONS either). + ACE_NEW_THROWS_EXCEPTIONS either). * ace/Thread_Manager.cpp (join_thr): Moved the code for ACE_HAS_PTHREADS_DRAFT4 that does a pthread_detach from here to - ACE_OS::thr_join (which this function calls) so that - ACE_Thread::join also causes a pthread_detach. This is needed on - Pthreads Draft 4 (HP-UX 10.x) to properly clean up the thread - storage and avoid leaking memory. Thanks to Vladimir Kondratiev - <vladimir.kondratiev@intel.com> for reporting this and providing - excellent test cases to help resolve it quickly. + ACE_OS::thr_join (which this function calls) so that + ACE_Thread::join also causes a pthread_detach. This is needed on + Pthreads Draft 4 (HP-UX 10.x) to properly clean up the thread + storage and avoid leaking memory. Thanks to Vladimir Kondratiev + <vladimir.kondratiev@intel.com> for reporting this and providing + excellent test cases to help resolve it quickly. * ace/OS.i (ACE_OS::thr_join): Added the code that was removed from ACE_Thread_Manager::join_thr, above. * ace/Event_Handler.cpp (~ACE_Event_Handler): Add an ACE_Errno_Guard around purge_pending_notifications to prevent against an ENOTSUP - from purge_pending_notifications. + from purge_pending_notifications. Thu Mar 1 17:02:10 2001 Darrell Brunsch <brunsch@uci.edu> @@ -1464,22 +1470,22 @@ Thu Mar 1 17:02:10 2001 Darrell Brunsch <brunsch@uci.edu> * bin/PerlACE/Process_Win32.pm: Fixed up the Spawn code to use the Executable () instead of - {EXECUTABLE}, since that's how we get the adjusted name (with - .exe and stuff). And also skipped the file existance with - IgnoreExeSubDir, since IgnoreExeSubDir is often used when you - need to find the program in the path. + {EXECUTABLE}, since that's how we get the adjusted name (with + .exe and stuff). And also skipped the file existance with + IgnoreExeSubDir, since IgnoreExeSubDir is often used when you + need to find the program in the path. Thu Mar 1 15:48:10 2001 Darrell Brunsch <brunsch@uci.edu> * bin/fuzz.pl: Fuzz was being a bit to eager with the class matching in the - TRACE test. It was having problems with something like: + TRACE test. It was having problems with something like: char *foo::bar () since it thought the class was "char *foo" instead of just - "foo". + "foo". But it is feeling muuuuuuch better now. @@ -1532,8 +1538,8 @@ Thu Mar 1 10:54:23 2001 Ossama Othman <ossama@uci.edu> * include/makeinclude/wrapper_macros.GNU: Fixed all incorrect uses of CFLAGS and CCFLAGS where - preprocessor macros were defined. Preprocessor macros should be - defined in the CPPFLAGS variable. + preprocessor macros were defined. Preprocessor macros should be + defined in the CPPFLAGS variable. Define ACE_USE_RCSID in non-debug builds (i.e. debug=0). @@ -1541,7 +1547,7 @@ Thu Mar 01 07:29:58 2001 Carlos O'Ryan <coryan@uci.edu> * bin/auto_run_tests.lst: Fixed typos in the Service_Context_Manipulation test. It was - running in platforms where it shouldn't. + running in platforms where it shouldn't. Thu Mar 01 07:07:03 2001 Christopher Kohlhoff <chris@kohlhoff.com> @@ -1564,9 +1570,9 @@ Wed Feb 28 22:58:50 2001 Carlos O'Ryan <coryan@uci.edu> * html/index.html: Reference */index.html directly, instead of assuming that the - web browser will use index.html as the default file, this also - works better when browsing locally. Thanks to Alain Decamps - <Alain.Decamps@pidpa.be> for pointing this out. + web browser will use index.html as the default file, this also + works better when browsing locally. Thanks to Alain Decamps + <Alain.Decamps@pidpa.be> for pointing this out. Wed Feb 28 16:41:42 2001 Darrell Brunsch <brunsch@uci.edu> @@ -1584,40 +1590,40 @@ Wed Feb 28 15:35:59 2001 Joe Hoffert <joeh@cs.wustl.edu> * docs/bugzilla-build-setup.html: Added html file outlining steps for adding a bugzilla nightly - build. + build. Wed Feb 28 00:25:42 2001 Darrell Brunsch <brunsch@uci.edu> * bin/auto_run_tests.lst: The examples/POA/Adapter_Activator/run_test.pl test was being - run twice. In fairness to all the other tests that only get to - run once, I removed the second instance. + run twice. In fairness to all the other tests that only get to + run once, I removed the second instance. Tue Feb 27 17:31:48 2001 Steve Huston <shuston@riverace.com> * include/makeinclude/platform_hpux_aCC.GNU: Turn on -D_REENTRANT for the threads=0 build - this enables the reentrant functions which - are present, threads or not. It does not turn on threads, which - are instead enabled by -D_POSIX_C_SOURCE=199506L. + are present, threads or not. It does not turn on threads, which + are instead enabled by -D_POSIX_C_SOURCE=199506L. * ace/config-hpux-11.00.h: Added ACE_HAS_2_PARAM_ASCTIME_R_AND_CTIME_R * ace/OS.i (rand_r): Added 'defined (HPUX_11)' to the ways to get to the POSIX.1c-conformant version of rand_r. Allows clean compile - on HP-UX 11 without thread support. Also lined up some - misaligned #if/else/endif lines to make it easier to find one's - way through. + on HP-UX 11 without thread support. Also lined up some + misaligned #if/else/endif lines to make it easier to find one's + way through. * ace/OS_Dirent.inl (readdir_r): Added 'defined (HPUX_11)' to the ways to get the POSIX.1c-conformant version of readdir_r. Allows - clean compile on HP-UX 11 without thread support. + clean compile on HP-UX 11 without thread support. Tue Feb 27 11:38:38 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> * performance-tests/TCP/tcp_test.cpp (main): Zapped the non-existent '-f' option. Thanks to Mike Curtis for reporting - this. + this. * performance-tests/TCP/tcp_test.cpp (main): Check for improper bufsz. Thanks to Mike Curtis for reporting this. @@ -1637,8 +1643,8 @@ Tue Feb 27 10:23:20 2001 Balachandran Natarajan <bala@cs.wustl.edu> * Makefile: Changed the ChangeLog listing in the CONTROLLED_FILES to ChangeLog*. This should help us to avoid missing ChangeLog - entries in the beta distribution, just about every time we split - and add a ChangeLog. + entries in the beta distribution, just about every time we split + and add a ChangeLog. Mon Feb 27 07:39:16 2001 Chad Elliott <elliott_c@ociweb.com> @@ -1650,9 +1656,9 @@ Mon Feb 27 07:39:16 2001 Chad Elliott <elliott_c@ociweb.com> * ace/OS.h: * ace/config-hpux-11.00.h: Modification to remove system header macro definition conflicts. - This change does not suppress the conflicts, but does an undef - of the two conflicting macros before includeing tcp.h. Thanks - to Sam Chong <sam@i2i-inc.com> for reporting this problem. + This change does not suppress the conflicts, but does an undef + of the two conflicting macros before includeing tcp.h. Thanks + to Sam Chong <sam@i2i-inc.com> for reporting this problem. Mon Feb 26 19:01:16 2001 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu> @@ -1669,9 +1675,9 @@ Mon Feb 26 18:18:38 2001 Steve Huston <shuston@riverace.com> (ACE_SString::find (char c, int pos) const): (ACE_WString::find (const ACE_WSTRING_TYPE *s, int pos) const): If match is found, return index from start of string, not from - 'pos'. Matches the C++ string standard, which was the model for - these classes. Thanks to Glen Osterhout <gosterho@elronsw.com> - for this fix. + 'pos'. Matches the C++ string standard, which was the model for + these classes. Thanks to Glen Osterhout <gosterho@elronsw.com> + for this fix. * ace/SString.h: Clarified comments on return value for above functions. @@ -1692,12 +1698,12 @@ Mon Feb 26 12:29:45 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> * ace/ARGV.h: Clarify that the int add (const ACE_TCHAR *next_arg) method assumes ownership of <next_arg>, rather than copying it. - Thanks to Steve Hespelt <shespelt@bear.com> for reporting this. - Fixes [Bug 807]. + Thanks to Steve Hespelt <shespelt@bear.com> for reporting this. + Fixes [Bug 807]. The generated version_tests dsp didn't work (since DLL_Test_Impl - creates a library). Manually created the configurations in the - tests/DLL_Test_Impl.dsp version instead. + creates a library). Manually created the configurations in the + tests/DLL_Test_Impl.dsp version instead. Mon Feb 26 15:54:33 2001 Phil Mesnier <mesnier_p@ociweb.com> @@ -1722,19 +1728,19 @@ Mon Feb 26 12:29:45 2001 Douglas C. Schmidt <schmidt@ace.cs.wustl.edu> * ace/ARGV.h: Clarify that the int add (const ACE_TCHAR *next_arg) method assumes ownership of <next_arg>, rather than copying it. - Thanks to Steve Hespelt <shespelt@bear.com> for reporting this. - Fixes [Bug 807]. + Thanks to Steve Hespelt <shespelt@bear.com> for reporting this. + Fixes [Bug 807]. Mon Feb 26 12:27:46 2001 Jeff Parsons <parsons@cs.wustl.edu> * ace/CDR_Stream.h: * ace/CDR_Stream.i: Added method to reset the byte order in class OuputCDR. This - will be used by a gateway to set the byte order of a reply to a - reqest it has handled to one different than the gateway's own - byte order, if necessary. It should be called while the stream - is still empty, that is, before any part of the reply, including - the header, has been marshaled. + will be used by a gateway to set the byte order of a reply to a + reqest it has handled to one different than the gateway's own + byte order, if necessary. It should be called while the stream + is still empty, that is, before any part of the reply, including + the header, has been marshaled. Mon Feb 26 12:07:00 2001 Nanbor Wang <nanbor@cs.wustl.edu> @@ -1855,15 +1861,15 @@ Mon Feb 26 00:19:28 2001 Darrell Brunsch <brunsch@uci.edu> * tests/Upgradable_RW_Test.dsp: Removed the Alpha configurations. I don't think we've actually - looked at these for over a year, and they are missing from a lot - of projects. Thankfully, they are easy to recreate if they are - needed. + looked at these for over a year, and they are missing from a lot + of projects. Thankfully, they are easy to recreate if they are + needed. * ACE-INSTALL.html: Changed the Windows NT Alpha section to mention that we no - longer have configurations out of the box for Win32/Alpha, they - must be recreated if needed. + longer have configurations out of the box for Win32/Alpha, they + must be recreated if needed. * WindozeCE/ACEApp.cpp: (removed) * WindozeCE/ACEApp.h: (removed) @@ -2032,11 +2038,11 @@ Sat Feb 24 14:07:11 2001 Darrell Brunsch <brunsch@uci.edu> * tests/Thread_Pool_Reactor_Test.cpp: Did a careful study of what this test was doing in regards to - sending wide-character strings, and made it consistent. The - most troublesome aspect was the fact that the length is encoded - in the shutdown string ("\011shutdown") which meant that I had - to send the length in characters, not in bytes. This should - really fix [Bug 537]. + sending wide-character strings, and made it consistent. The + most troublesome aspect was the fact that the length is encoded + in the shutdown string ("\011shutdown") which meant that I had + to send the length in characters, not in bytes. This should + really fix [Bug 537]. Sat Feb 24 14:37:00 2001 Craig Rodrigues <crodrigu@bbn.com> diff --git a/ace/Select_Reactor_Base.h b/ace/Select_Reactor_Base.h index 3ce734b6a7b..2ec0874db1e 100644 --- a/ace/Select_Reactor_Base.h +++ b/ace/Select_Reactor_Base.h @@ -28,6 +28,10 @@ #include "ace/Pipe.h" #include "ace/Reactor_Impl.h" +#if defined (ACE_HAS_REACTOR_NOTIFICATION_QUEUE) +#include "ace/Unbounded_Queue.h" +#endif /* ACE_HAS_REACTOR_NOTIFICATION_QUEUE */ + // Add useful typedefs to simplify the following code. typedef void (ACE_Handle_Set::*ACE_FDS_PTMF) (ACE_HANDLE); typedef int (ACE_Event_Handler::*ACE_EH_PTMF) (ACE_HANDLE); |