1<?xml version="1.0" encoding="utf-8"?> 2<explicit-failures-markup> 3 <!-- spirit (v2) --> 4 <library name="spirit"> 5 <mark-unusable> 6 <toolset name="sun-5.7"/> 7 <toolset name="sun-5.8"/> 8 <toolset name="sun-5.9"/> 9 <toolset name="sun-5.10"/> 10 <toolset name="vacpp*"/> 11 <toolset name="borland-*"/> 12 <toolset name="cray-8.0"/> 13 <toolset name="msvc-7.1*"/> 14 <toolset name="clang-*-3.1*11*"/> 15 <toolset name="gcc-mngw-gnu-4.3c+"/> 16 </mark-unusable> 17 <test name="karma_*" category="Karma" /> 18 <test name="lex_*" category="Lex" /> 19 <test name="qi_*" category="Qi" /> 20 <test name="support_*" category="Support" /> 21 <test name="x3_*" category="X3" /> 22 </library> 23 24 <!-- spirit (v2) repository --> 25 <library name="spirit/repository"> 26 <mark-unusable> 27 <toolset name="borland-cb2009"/> 28 <toolset name="borland-cb2010"/> 29 <toolset name="msvc-7.1*"/> 30 <toolset name="clang-*-3.1*11*"/> 31 <toolset name="gcc-mngw-gnu-4.3c+"/> 32 </mark-unusable> 33 <test name="karma_*" category="Karma" /> 34 <test name="qi_*" category="Qi" /> 35 </library> 36 37 <!-- spirit (classic) --> 38 <library name="spirit/classic"> 39 <mark-unusable> 40 <toolset name="msvc-6.5*"/> 41 <toolset name="borland-5.5*"/> 42 <toolset name="borland-5.6*"/> 43 <toolset name="borland-5.8*"/> 44 <toolset name="msvc-7.0"/> 45 <toolset name="gcc-2.95.3-linux"/> 46 <toolset name="gcc-2.95.3-stlport-4.5.3-linux"/> 47 <toolset name="gcc-2.95.3-stlport-4.6.2-linux"/> 48 <toolset name="sunpro-5_3-sunos"/> 49 50 <note> 51 <p> 52 Historically, Spirit supported a lot of compilers, including (to some 53 extent) poorly conforming compilers such as VC6. Spirit v1.6.x will be 54 the last release that will support older poorly conforming compilers. 55 Starting from Spirit v1.8.0, ill conforming compilers will not be 56 supported. If you are still using one of these older compilers, you can 57 still use Spirit v1.6.x. 58 </p> 59 <p> 60 The reason why Spirit v1.6.x worked on old non-conforming compilers is 61 that the authors laboriously took the trouble of searching for 62 workarounds to make these compilers happy. The process takes a lot of 63 time and energy, especially when one encounters the dreaded ICE or 64 "Internal Compiler Error". Sometimes searching for a single workaround 65 takes days or even weeks. Sometimes, there are no known workarounds. This 66 stifles progress a lot. And, as the library gets more progressive and 67 takes on more advanced C++ techniques, the difficulty is escalated to 68 even new heights. 69 </p> 70 <p> 71 Spirit v1.6.x will still be supported. Maintenance and bug fixes will 72 still be applied. There will still be active development for the back- 73 porting of new features introduced in Spirit v1.8.0 (and Spirit 1.9.0) 74 to lesser able compilers; hopefully, fueled by contributions from the 75 community. For instance, there is already a working AST tree back-port 76 for VC6 and VC7 by Peder Holt. 77 </p> 78 </note> 79 </mark-unusable> 80 <mark-expected-failures> 81 <test name="action_tests*"/> 82 <toolset name="iw-7_1-vc6"/> 83 <note author="Aleksey Gurtovoy" refid="4"/> 84 </mark-expected-failures> 85 <mark-expected-failures> 86 <test name="ast_calc_tests*"/> 87 <test name="closure_tests*"/> 88 <test name="multi_pass_compile_tests"/> 89 <test name="repeat_ast_tests*"/> 90 <toolset name="intel-8.0-linux"/> 91 <toolset name="intel-8.1-linux"/> 92 <note author="Aleksey Gurtovoy"> 93 This failure is caused by a compiler bug that manifests itself in the 94 particular environment/hardware configuration the test has been run in. 95 You may or may not experience this issue in your local setup. 96 </note> 97 </mark-expected-failures> 98 <mark-expected-failures> 99 <test name="escape_char_parser_tests*"/> 100 <toolset name="intel-7.1-linux"/> 101 <toolset name="intel-7.1-stdlib-default-linux"/> 102 <note author="Aleksey Gurtovoy" refid="19"/> 103 </mark-expected-failures> 104 <mark-expected-failures> 105 <test name="escape_char_parser_tests*"/> 106 <toolset name="iw-7_1-vc6*"/> 107 <note author="Aleksey Gurtovoy" refid="28"/> 108 </mark-expected-failures> 109 <mark-expected-failures> 110 <test name="chset_tests*"/> 111 <toolset name="iw-7_1-vc6-stlp-4_5_3"/> 112 <note author="Aleksey Gurtovoy" refid="28"/> 113 </mark-expected-failures> 114 <mark-expected-failures> 115 <test name="int_numerics"/> 116 <test name="karma_pattern*"/> 117 <test name="karma_sequence"/> 118 <test name="rule"/> 119 <test name="sequence"/> 120 <toolset name="acc"/> 121 <note author="Boris Gubenko" refid="47"/> 122 </mark-expected-failures> 123 <mark-expected-failures> 124 <test name="lexertl3"/> 125 <test name="lexertl4"/> 126 <test name="lexertl5"/> 127 <toolset name="gcc-3.4.6_linux_ia64"/> 128 <note author="Boris Gubenko"> 129 With GCC 3.4.6 the test fails with ICE: internal compiler error. 130 The footprint is similar to that in GCC Bugzilla Bug 34950 131 except 34950 is a regression introduced in GCC 4.2.3. In any 132 case, whatever the problem is, the GCC 4.x series does not seem 133 to have it: the test compiles just fine with GCC 4.x compiler. 134 </note> 135 </mark-expected-failures> 136 </library> 137</explicit-failures-markup> 138