libarchive - Project Hosting on Google Code

Today 4 hours ago r1767 (Eliminate the warnings which Borland C Compiler complains "U...) committed by ggcueroad - Eliminate the warnings which Borland C Compiler complains "Unreachable code". Eliminate the warnings which Borland C Compiler complains "Unreachable code". 5 hours ago r1766 (In libarchive_test, disable the warning which Borland C comp...) committed by ggcueroad - In libarchive_test, disable the warning which Borland C compiler complains "constant out of range in comparison". In libarchive_test, disable the warning which Borland C compiler complains "constant out of range in comparison". 21 hours ago r1765 (On Windows, don't use _strdup directly because Borland C doe...) committed by ggcueroad - On Windows, don't use _strdup directly because Borland C does not provide _strdup. On Windows, don't use _strdup directly because Borland C does not provide _strdup. Last 7 days Dec 18, 2009 r1764 (_chmod is obsolete on the Borland C environment. Use Win32 A...) committed by ggcueroad - _chmod is obsolete on the Borland C environment. Use Win32 API instead of _chmod/_wchmod. _chmod is obsolete on the Borland C environment. Use Win32 API instead of _chmod/_wchmod. Dec 18, 2009 r1763 (Eliminate warnings that some function parameters are not use...) committed by ggcueroad - Eliminate warnings that some function parameters are not used at libarchive/test/main.c. Eliminate warnings that some function parameters are not used at libarchive/test/main.c. Dec 18, 2009 r1762 (Don't use _getcwd and _strdup on Borland C Compiler. This is...) committed by ggcueroad - Don't use _getcwd and _strdup on Borland C Compiler. This is covering r1757. Don't use _getcwd and _strdup on Borland C Compiler. This is covering r1757. Dec 18, 2009 r1761 (On Windows, make sure not to leave empty directories when te...) committed by ggcueroad - On Windows, make sure not to leave empty directories when tests are success. On Windows, make sure not to leave empty directories when tests are success. Dec 17, 2009 LibarchiveFormats (Summary of formats supported by the library and command-line...) Wiki page commented on by 0rrm1k - Sexy stuff. Will use! Sexy stuff. Will use! Dec 17, 2009 issue 58 (error compiling in windows + mingw + msys) reported by gabrielnn77 - version of svn, compiling in mingw+msys, using cmake $ make .... [ 46%] Building C object libarchive/test/CMakeFiles/libarchive_test.dir/test_read_format_cpio_svr4_bzip2_rpm.c.obj c:/msys/1.0/opt/src/libarchive-svn/libarchive/test/test_read_format_cpio_svr4_bzip2_rpm.c: In function `test_read_format_cpio_svr4_bzip2_rpm': c:/msys/1.0/opt/src/libarchive-svn/libarchive/test/test_read_format_cpio_svr4_bzip2_rpm.c:100: warning: implicit declaration of function `archive_read_support_compression_rpm' make[2]: * [libarchive/test/CMakeFiles/libarchive_test.dir/test_read_format_cpio_svr4_bzip2_rpm.c.obj] Error 1 make[1]: * [libarchive/test/CMakeFiles/libarchive_test.dir/all] Error 2 make: * [all] Error 2 version of svn, compiling in mingw+msys, using cmake $ make .... [ 46%] Building C object libarchive/test/CMakeFiles/libarchive_test.dir/test_read_format_cpio_svr4_bzip2_rpm.c.obj c:/msys/1.0/opt/src/libarchive-svn/libarchive/test/test_read_format_cpio_svr4_bzip2_rpm.c: In function `test_read_format_cpio_svr4_bzip2_rpm': c:/msys/1.0/opt/src/libarchive-svn/libarchive/test/test_read_format_cpio_svr4_bzip2_rpm.c:100: warning: implicit declaration of function `archive_read_support_compression_rpm' make[2]: * [libarchive/test/CMakeFiles/libarchive_test.dir/test_read_format_cpio_svr4_bzip2_rpm.c.obj] Error 1 make[1]: * [libarchive/test/CMakeFiles/libarchive_test.dir/all] Error 2 make: * [all] Error 2 Dec 17, 2009 issue 57 (mingw : redefinition of "typedef int pid_t;") reported by gabrielnn77 - What steps will reproduce the problem? 1.cmake -G "MSYS Makefiles" -DCMAKE_PREFIX_PATH=/mingw /opt/src/libarchive-svn 2.make 3. What is the expected output? What do you see instead? $ make Scanning dependencies of target archive [ 0%] Building C object libarchive/CMakeFiles/archive.dir/archive_check_magic.c.obj In file included from c:/msys/1.0/opt/src/libarchive-svn/libarchive/archive_platform.h:60, from c:/msys/1.0/opt/src/libarchive-svn/libarchive/archive_check_magic.c:26: c:/msys/1.0/opt/src/libarchive-svn/libarchive/archive_windows.h:302: error: redefinition of typedef 'pid_t' c:/MinGW/include/sys/types.h:75: error: previous declaration of 'pid_t' was here make[2]: * [libarchive/CMakeFiles/archive.dir/archive_check_magic.c.obj] Error 1 make[1]: * [libarchive/CMakeFiles/archive.dir/all] Error 2 make: * [all] Error 2 What version are you using? svn, 2009-12-17 On what operating system? windows, mingw, msys Please provide any additional information below. What steps will reproduce the problem? 1.cmake -G "MSYS Makefiles" -DCMAKE_PREFIX_PATH=/mingw /opt/src/libarchive-svn 2.make 3. What is the expected output? What do you see instead? $ make Scanning dependencies of target archive [ 0%] Building C object libarchive/CMakeFiles/archive.dir/archive_check_magic.c.obj In file included from c:/msys/1.0/opt/src/libarchive-svn/libarchive/archive_platform.h:60, from c:/msys/1.0/opt/src/libarchive-svn/libarchive/archive_check_magic.c:26: c:/msys/1.0/opt/src/libarchive-svn/libarchive/archive_windows.h:302: error: redefinition of typedef 'pid_t' c:/MinGW/include/sys/types.h:75: error: previous declaration of 'pid_t' was here make[2]: * [libarchive/CMakeFiles/archive.dir/archive_check_magic.c.obj] Error 1 make[1]: * [libarchive/CMakeFiles/archive.dir/all] Error 2 make: * [all] Error 2 What version are you using? svn, 2009-12-17 On what operating system? windows, mingw, msys Please provide any additional information below. Dec 17, 2009 r1760 (Disable two warnings on Borland C Compiler. - Constant out ...) committed by ggcueroad - Disable two warnings on Borland C Compiler. - Constant out of range in comparison(W8068). - Suspicious pointer arithmetic(W8072). Disable two warnings on Borland C Compiler. - Constant out of range in comparison(W8068). - Suspicious pointer arithmetic(W8072). Dec 17, 2009 r1759 (Remove an unneeded pragma that disable signed/unsigned misma...) committed by ggcueroad - Remove an unneeded pragma that disable signed/unsigned mismatch warning. Remove an unneeded pragma that disable signed/unsigned mismatch warning. Dec 17, 2009 r1758 (Add IO_REPARSE_TAG_SYMLINK macro if not defined. Old Windows...) committed by ggcueroad - Add IO_REPARSE_TAG_SYMLINK macro if not defined. Old Windows SDKs do not provide IO_REPARSE_TAG_SYMLINK. Add IO_REPARSE_TAG_SYMLINK macro if not defined. Old Windows SDKs do not provide IO_REPARSE_TAG_SYMLINK. Dec 17, 2009 r1757 (Borland C does not provide _access,_chdir,_getcwd,_strdup an...) committed by ggcueroad - Borland C does not provide _access,_chdir,_getcwd,_strdup and _umask. Borland C does not provide _access,_chdir,_getcwd,_strdup and _umask. Dec 16, 2009 r1756 (Fix indents. ) committed by ggcueroad - Fix indents. Fix indents. Dec 16, 2009 r1755 (Avoid segv on linux64. It caused at second calling vfprintf(...) committed by ggcueroad - Avoid segv on linux64. It caused at second calling vfprintf() of vlogprintf() in /test/main.c. Use a copy of ap for second calling vfprintf(). Avoid segv on linux64. It caused at second calling vfprintf() of vlogprintf() in /test/main.c. Use a copy of ap for second calling vfprintf(). Dec 16, 2009 r1754 (bzstream_valid is not presented if bzip2 is not available. ) committed by ggcueroad - bzstream_valid is not presented if bzip2 is not available. bzstream_valid is not presented if bzip2 is not available. Dec 16, 2009 r1753 (All of invalid_parameter_handler() are used only when HAVE__...) committed by ggcueroad - All of invalid_parameter_handler() are used only when HAVE__CrtSetReportMode is defined. All of invalid_parameter_handler() are used only when HAVE__CrtSetReportMode is defined. Dec 16, 2009 r1752 (Borland C++ does not provide sys/utime.h. Use macro HAVE_SYS...) committed by ggcueroad - Borland C++ does not provide sys/utime.h. Use macro HAVE_SYS_UTIME_H to include sys/utime.h and also use macro HAVE_UTIME_H to include utime.h. Borland C++ does not provide sys/utime.h. Use macro HAVE_SYS_UTIME_H to include sys/utime.h and also use macro HAVE_UTIME_H to include utime.h. Dec 16, 2009 r1751 (Isolate libxml header check The check HAVE_LIBXML_XMLREADER...) committed by brad.k...@kitware.com - Isolate libxml header check The check HAVE_LIBXML_XMLREADER_H should not use LA_CHECK_INCLUDE_FILE because it sets CMAKE_REQUIRED_INCLUDES for only the one check. The LA_CHECK_INCLUDE_FILE macro accumulates headers that exist for use in future checks, but including libxml/xmlreader.h breaks later checks because they do not set CMAKE_REQUIRED_INCLUDES. Instead we use CHECK_INCLUDE_FILES directly for this check. Isolate libxml header check The check HAVE_LIBXML_XMLREADER_H should not use LA_CHECK_INCLUDE_FILE because it sets CMAKE_REQUIRED_INCLUDES for only the one check. The LA_CHECK_INCLUDE_FILE macro accumulates headers that exist for use in future checks, but including libxml/xmlreader.h breaks later checks because they do not set CMAKE_REQUIRED_INCLUDES. Instead we use CHECK_INCLUDE_FILES directly for this check. Dec 16, 2009 issue 22 (Conditionally disable openssl requirement for mtree format i...) commented on by brad.k...@kitware.com - I've committed CMake build support as r1750: cmake -DENABLE_OPENSSL=OFF ... I've committed CMake build support as r1750: cmake -DENABLE_OPENSSL=OFF ... Dec 16, 2009 r1750 (Create ENABLE_OPENSSL CMake option This option tells libarc...) committed by brad.k...@kitware.com - Create ENABLE_OPENSSL CMake option This option tells libarchive whether to check for OpenSSL and use it. We enable it by default since we previously did this unconditionally. This provides the CMake equivalent to the --without-openssl option, spelled as -DENABLE_OPENSSL=OFF. See issue #22. Create ENABLE_OPENSSL CMake option This option tells libarchive whether to check for OpenSSL and use it. We enable it by default since we previously did this unconditionally. This provides the CMake equivalent to the --without-openssl option, spelled as -DENABLE_OPENSSL=OFF. See issue #22. Dec 16, 2009 r1749 (Avoid warnings with bcc that are comparing signed and unsign...) committed by ggcueroad - Avoid warnings with bcc that are comparing signed and unsigned values. Avoid warnings with bcc that are comparing signed and unsigned values. Dec 16, 2009 r1748 (Explicit unused variables at __archive_check_child(). ) committed by ggcueroad - Explicit unused variables at __archive_check_child(). Explicit unused variables at __archive_check_child(). Dec 16, 2009 r1747 (Eliminate a paramter 'sym' of __link(), which is no longer n...) committed by ggcueroad - Eliminate a paramter 'sym' of __link(), which is no longer needed. Eliminate a paramter 'sym' of __link(), which is no longer needed. Dec 16, 2009 r1746 (Correct a wide character buffer size of MultiByteToWideChar(...) committed by ggcueroad - Correct a wide character buffer size of MultiByteToWideChar() in __link(). Correct a wide character buffer size of MultiByteToWideChar() in __link(). Dec 16, 2009 r1745 (Remove unreachable statement in archive_read_support_format_...) committed by ggcueroad - Remove unreachable statement in archive_read_support_format_raw.c Remove unreachable statement in archive_read_support_format_raw.c Dec 16, 2009 r1744 (Remove unreachable statement. ) committed by ggcueroad - Remove unreachable statement. Remove unreachable statement. Dec 16, 2009 r1743 (Remove unused variable. ) committed by ggcueroad - Remove unused variable. Remove unused variable. Dec 16, 2009 r1742 (On Windows, do not define and use WIFSIGNALED because condit...) committed by ggcueroad - On Windows, do not define and use WIFSIGNALED because condition of WIFSIGNALED was always false and bcc complained. On Windows, do not define and use WIFSIGNALED because condition of WIFSIGNALED was always false and bcc complained. Dec 15, 2009 issue 56 (Crash on mtree archive with contents=.) Status changed by kientzle - Fixed in r1741. I used the fix provided as-is. I put the regression test in libarchive_test since this is a libarchive bug and not a tar bug. Thank you! Status: Fixed Fixed in r1741. I used the fix provided as-is. I put the regression test in libarchive_test since this is a libarchive bug and not a tar bug. Thank you! Status: Fixed Dec 15, 2009 r1741 (Fix Issue 56: Don't crash when reading an mtree input with ...) committed by kientzle - Fix Issue 56 : Don't crash when reading an mtree input with "content=." Fix Issue 56 : Don't crash when reading an mtree input with "content=." Dec 15, 2009 issue 56 (Crash on mtree archive with contents=.) reported by zwe...@xs4all.nl - What steps will reproduce the problem? 1. { echo '#mtree';echo 'd type=dir content=.';}\|bsdtar cf d.tar @- What is the expected output? d.tar created without error containing single directory called d What do you see instead? Segmentation violation What version are you using? subversion trunk revision 1737 On what operating system? GNU/linux, Debian testing Please provide any additional information below. Special handling of contents=. is missing a NULL check. Attaching a patch with fix and regression test. Please notice added file in patch: tar/test/test_mtree_cwd.c What steps will reproduce the problem? 1. { echo '#mtree';echo 'd type=dir content=.';}\|bsdtar cf d.tar @- What is the expected output? d.tar created without error containing single directory called d What do you see instead? Segmentation violation What version are you using? subversion trunk revision 1737 On what operating system? GNU/linux, Debian testing Please provide any additional information below. Special handling of contents=. is missing a NULL check. Attaching a patch with fix and regression test. Please notice added file in patch: tar/test/test_mtree_cwd.c Dec 15, 2009 r1740 (Eliminate test_write_compress_program failur on Visual Studi...) committed by ggcueroad - Eliminate test_write_compress_program failur on Visual Studio 10. Eliminate test_write_compress_program failur on Visual Studio 10. Dec 15, 2009 r1739 (Remove unneeded including windows.h at tar/write.c. It's inc...) committed by ggcueroad - Remove unneeded including windows.h at tar/write.c. It's included through bsdtar_platform.h. Remove unneeded including windows.h at tar/write.c. It's included through bsdtar_platform.h. Dec 15, 2009 r1738 (Avoid build error on Visual Studio 10. ) committed by ggcueroad - Avoid build error on Visual Studio 10. Avoid build error on Visual Studio 10. Dec 14, 2009 issue 48 (Use CMAKE_CURRENT_SOURCE_DIR) commented on by brad.k...@kitware.com - >> 2.) The RUNTIME_OUTPUT_DIRECTORY properties can be set to ${libarchive_BINARY_DIR}, > It would be nice to allow individual directories here to be incorporated into other > projects, so relying on libarchive_BINARY_DIR within those directories would be > wrong. Why wouldn't we just leave those as ${CMAKE_BINARY_DIR}, putting the output > into the top level of the incorporating project, whatever that might be. The best approach is to let the outside projects configure the output location. Perhaps: if(NOT libarchive_RUNTIME_OUTPUT_DIRECTORY) set(libarchive_RUNTIME_OUTPUT_DIRECTORY ${libarchive_BINARY_DIR}) endif() then use ${libarchive_RUNTIME_OUTPUT_DIRECTORY} to set the property. Another approach is if(NOT CMAKE_RUNTIME_OUTPUT_DIRECTORY) set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${libarchive_BINARY_DIR}) endif() at the top, and then do not set the property. CMake will use this as the property default when it creates the target. If an outside user project sets the value first, libarchive simply won't override it. This approach is a little simpler but provides slightly less flexibility. >> 2.) The RUNTIME_OUTPUT_DIRECTORY properties can be set to ${libarchive_BINARY_DIR}, > It would be nice to allow individual directories here to be incorporated into other > projects, so relying on libarchive_BINARY_DIR within those directories would be > wrong. Why wouldn't we just leave those as ${CMAKE_BINARY_DIR}, putting the output > into the top level of the incorporating project, whatever that might be. The best approach is to let the outside projects configure the output location. Perhaps: if(NOT libarchive_RUNTIME_OUTPUT_DIRECTORY) set(libarchive_RUNTIME_OUTPUT_DIRECTORY ${libarchive_BINARY_DIR}) endif() then use ${libarchive_RUNTIME_OUTPUT_DIRECTORY} to set the property. Another approach is if(NOT CMAKE_RUNTIME_OUTPUT_DIRECTORY) set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${libarchive_BINARY_DIR}) endif() at the top, and then do not set the property. CMake will use this as the property default when it creates the target. If an outside user project sets the value first, libarchive simply won't override it. This approach is a little simpler but provides slightly less flexibility. Dec 14, 2009 issue 48 (Use CMAKE_CURRENT_SOURCE_DIR) commented on by brad.k...@kitware.com - >> 4.) Hunks like this ... are not necessary. > Do you mean the change isn't needed or the logging itself isn't needed? The logging is necessary, the change isn't. Logging to the top level is fine. >> 4.) Hunks like this ... are not necessary. > Do you mean the change isn't needed or the logging itself isn't needed? The logging is necessary, the change isn't. Logging to the top level is fine. Dec 14, 2009 r1737 (Don't check whether bsdxml is available or not. The bsdxml i...) committed by ggcueroad - Don't check whether bsdxml is available or not. The bsdxml is intended for use in FreeBSD base system. Don't check whether bsdxml is available or not. The bsdxml is intended for use in FreeBSD base system. Dec 13, 2009 issue 55 (FreeBSD 6 build failure) reported by kientzle - FreeBSD 6.4 has an older version of OpenSSL that lacks the SHA256, SHA384, and SHA512 implementations but does have a sha.h header. The configure.ac needs to explicitly test for SHA256, SHA384, and SHA512 support (as documented in a TODO in that file). FreeBSD 6.4 has an older version of OpenSSL that lacks the SHA256, SHA384, and SHA512 implementations but does have a sha.h header. The configure.ac needs to explicitly test for SHA256, SHA384, and SHA512 support (as documented in a TODO in that file). Dec 13, 2009 r1736 (Refactor the release-building scripts just a bit: * bump-ve...) committed by kientzle - Refactor the release-building scripts just a bit: * bump-version.sh actually bumps the 'version' value * clean.sh attempts to clean the local directory as thoroughly as possible * autogen.sh runs clean.sh, edits the current version value into configure.ac, archive.h, and archive_entry.h, re-runs autoconf/automake/etc, and does a ./configure && make distcheck to verify the distribution on the local system. Refactor the release-building scripts just a bit: * bump-version.sh actually bumps the 'version' value * clean.sh attempts to clean the local directory as thoroughly as possible * autogen.sh runs clean.sh, edits the current version value into configure.ac, archive.h, and archive_entry.h, re-runs autoconf/automake/etc, and does a ./configure && make distcheck to verify the distribution on the local system. Last 30 days Dec 13, 2009 r1735 (We no longer use %jd or %lld from the system printf, so we d...) committed by kientzle - We no longer use %jd or %lld from the system printf, so we don't need to test for them. We no longer use %jd or %lld from the system printf, so we don't need to test for them. Dec 13, 2009 r1734 (Correct the spelling of RIPEMD160. ) committed by kientzle - Correct the spelling of RIPEMD160. Correct the spelling of RIPEMD160. Dec 13, 2009 issue 48 (Use CMAKE_CURRENT_SOURCE_DIR) Labels changed by kientzle - Labels: Milestone-2.8 Labels: Milestone-2.8 Dec 13, 2009 issue 22 (Conditionally disable openssl requirement for mtree format i...) Labels changed by kientzle - Labels: Milestone-2.8 Labels: Milestone-2.8 Dec 13, 2009 issue 22 (Conditionally disable openssl requirement for mtree format i...) Cc changed by kientzle - Brad: Do you have time to take a look at this? Cc: brad.k...@kitware.com Brad: Do you have time to take a look at this? Cc: brad.k...@kitware.com Dec 13, 2009 issue 48 (Use CMAKE_CURRENT_SOURCE_DIR) Status changed by kientzle - Changing status to "Started" since some work has been committed. Status: Started Changing status to "Started" since some work has been committed. Status: Started Dec 13, 2009 issue 29 (Minor build problem on QNX) Status changed by kientzle - Since it has been a few months, I'm going to close this as "WontFix". I'll happily reopen it if there's additional information. Status: WontFix Since it has been a few months, I'm going to close this as "WontFix". I'll happily reopen it if there's additional information. Status: WontFix Dec 13, 2009 issue 38 (Truncated archive extraction error message) Status changed by kientzle - With 2.8 coming out soon, I'm going to just close this. Status: Fixed With 2.8 coming out soon, I'm going to just close this. Status: Fixed Dec 13, 2009 issue 46 ([patch] add support for Haiku OS) Status changed by kientzle - I believe all of the issues raised here have been addressed, so I'm closing this as "Fixed." Please let me know if there are any further issues. Status: Fixed I believe all of the issues raised here have been addressed, so I'm closing this as "Fixed." Please let me know if there are any further issues. Status: Fixed

issue 48 (Use CMAKE_CURRENT_SOURCE_DIR) commented on by brad.k...@kitware.com - >> 2.) The RUNTIME_OUTPUT_DIRECTORY properties can be set to ${libarchive_BINARY_DIR}, > It would be nice to allow individual directories here to be incorporated into other > projects, so relying on libarchive_BINARY_DIR within those directories would be > wrong. Why wouldn't we just leave those as ${CMAKE_BINARY_DIR}, putting the output > into the top level of the incorporating project, whatever that might be. The best approach is to let the outside projects configure the output location. Perhaps: if(NOT libarchive_RUNTIME_OUTPUT_DIRECTORY) set(libarchive_RUNTIME_OUTPUT_DIRECTORY ${libarchive_BINARY_DIR}) endif() then use ${libarchive_RUNTIME_OUTPUT_DIRECTORY} to set the property. Another approach is if(NOT CMAKE_RUNTIME_OUTPUT_DIRECTORY) set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${libarchive_BINARY_DIR}) endif() at the top, and then do *not* set the property. CMake will use this as the property default when it creates the target. If an outside user project sets the value first, libarchive simply won't override it. This approach is a little simpler but provides slightly less flexibility.

>> 2.) The RUNTIME_OUTPUT_DIRECTORY properties can be set to ${libarchive_BINARY_DIR}, > It would be nice to allow individual directories here to be incorporated into other > projects, so relying on libarchive_BINARY_DIR within those directories would be > wrong. Why wouldn't we just leave those as ${CMAKE_BINARY_DIR}, putting the output > into the top level of the incorporating project, whatever that might be. The best approach is to let the outside projects configure the output location. Perhaps: if(NOT libarchive_RUNTIME_OUTPUT_DIRECTORY) set(libarchive_RUNTIME_OUTPUT_DIRECTORY ${libarchive_BINARY_DIR}) endif() then use ${libarchive_RUNTIME_OUTPUT_DIRECTORY} to set the property. Another approach is if(NOT CMAKE_RUNTIME_OUTPUT_DIRECTORY) set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${libarchive_BINARY_DIR}) endif() at the top, and then do *not* set the property. CMake will use this as the property default when it creates the target. If an outside user project sets the value first, libarchive simply won't override it. This approach is a little simpler but provides slightly less flexibility.

Today

Last 7 days

Last 30 days