aboutsummaryrefslogtreecommitdiffstats
path: root/configure.py
Commit message (Collapse)AuthorAgeFilesLines
* Further fixes to ci build scriptJack Lloyd2017-08-291-1/+1
|
* Create new fuzzer build modeJack Lloyd2017-08-251-32/+77
|
* Fix generating amalgamation under python3Jack Lloyd2017-08-251-3/+11
| | | | It complains about UTF8 chars unless we tell it that really it's UTF8
* Revert "configure.py: raise error when --prefix is not absolute path"Simon Warta2017-08-041-4/+0
| | | | | | | This reverts commit b6b575d961b15d59f9382086fd59d1e5062a3cdb. In order to disallow prepend_prefix for relative paths, we do not need to disallow a relativ --refix for users that do not use DESTDIR
* configure.py: raise error when --prefix is not absolute pathZoltan Gyarmati2017-08-041-0/+4
|
* Turn --destdir option into an environment variable, fixes #1101, fixes #996, ↵Zoltan Gyarmati2017-08-041-2/+6
| | | | | | | | | | | | | supersedes #997 This commit removes the (broken) --destdir command line option and introduces the DESTDIR environment variable for the 'make install' target. When it's set, the installation will take place in the DESTDIR directory, under the prefix set by --prefix, with all of the internal references (pkgconfig file, etc) containing only the prefix. This behavior is more standard (as in autotools and co.), so makes packaging easier. Based on work of danimo and zgyarmati
* Merge GH #1094 Add initial BearSSL providerJack Lloyd2017-08-021-1/+1
|\
| * BearSSL: Initial support and hash testsPatrick Wildt2017-07-051-1/+1
| | | | | | | | | | | | | | | | | | BearSSL is an implementation of the SSL/TLS protocol in C aiming to be correct and secure, small and highly portable. Thus making it nicer to be included in a rather sparse bootloader. This commit adds support for BearSSL's hash routines only, with more stuff coming up in following commits. The goal is to be able to test BearSSL using Botan's extensive testsuite.
* | Log success/failure without early return in version_from_compiler_output()Simon Warta2017-07-311-3/+3
| |
* | Use clang/mac code branch when cc_version is unsetSimon Warta2017-07-311-3/+3
| | | | | | | | This is easier to extend than the elif construction
* | Remove special case and early return for msvc matchingSimon Warta2017-07-311-8/+5
| |
* | Use native compiler versioning of MSVCSimon Warta2017-07-311-9/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | While using marketing names like 2013, 2015 etc. is more convenient at first sight, it requires keeping a table about all supported compiler versions, as there is no formular to calculate between the representations. Keeping a list of compilers leads to the following issue: if one version of Botan is released in 2017, requiring MSVS 2015 for one module, this source can be compiled using MSVS 2015 and 2017 but not a future version like 2019. Also preview/development versions of MSVC that may use an intermediate version number cannot be handled with the marketing name table because they may be unknown to the general public.
* | Explicitly handle negative cases in supported_compiler()Simon Warta2017-07-311-8/+13
| | | | | | | | | | | | | | | | | | | | Two cases only returned a falsy None implicitly by reaching the end of the function: - when user's compiler is not listed in the list of compiler choices - when user's compiler version is None and the module requires a min version This changes no behavior.
* | Refactor supported_compilerSimon Warta2017-07-311-5/+12
| |
* | Use _MSC_VER to get MSVC versionSimon Warta2017-07-301-4/+6
| |
* | Adapt MSVC version regexp to non-english outputsSimon Warta2017-07-301-1/+1
| |
* | Fix compiler version matching patternsSimon Warta2017-07-301-9/+7
| |
* | Refactor and test compiler version detectionSimon Warta2017-07-301-34/+46
| |
* | configure: explicitly concatenate stdout+stdderr for cc_putputSimon Warta2017-07-051-5/+6
|/ | | | | | This prevents a lint error in newer pylint versions ("R:2600, 8: Redefinition of cc_output type from tuple to str (redefined-variable-type)") and makes code clearer.
* Fixes for MSVC version detectionJack Lloyd2017-05-191-9/+23
| | | | | See comments in https://github.com/randombit/botan/commit/7f019174bfcf3c2b31bda74fe40ce35727db1cef
* Change clang regex format to just look for major.minorJack Lloyd2017-05-191-1/+1
| | | | Appears that Clang 3.7 doesn't include the patch in -v output
* Don't error out the build if compiler version does not match expected formatJack Lloyd2017-05-191-2/+3
|
* Merge GH #1043 Fix CMake generatorJack Lloyd2017-05-191-26/+59
|\
| * Fix StyleRyuhei Mori2017-05-071-32/+31
| |
| * Fix StylesRyuhei Mori2017-05-061-11/+13
| |
| * Fix CMake Generator for visual studioRyuhei Mori2017-05-061-16/+48
| | | | | | | | | | | | | | | | | | details: -Most of CMake commands accept only unix style path -Visual Stduio Generator allows mutiple configurations (debug;release) -Name collsion of static libary and dynamic libary and exe -Mismatch of dllexport/dllimport -Allows "x86_64:avx2" in isa flags
* | Fucking AppleJack Lloyd2017-05-181-4/+29
| | | | | | | | Map Apple's Clang version back to the LLVM equivalent
* | Let modules specify minimum compiler version, add runtime compiler version checkJack Lloyd2017-05-181-17/+79
| | | | | | | | | | | | | | | | | | | | | | | | | | This comes up recently to support CPU extensions by Intel (#826) and ARM (#844) since the intrinsic functions were only added in compiler versions more recent than the minimums we generally support, but might actually be useful for all kinds of things in the future since one can add a (new, otherwise optional) C++14 or C++17 module in 2.x branch while continuing to support the current set of supported compiler versions for the main library functionality. Uses RDSEED as the test case, to ensure version detection/comparisons work on all platforms.
* | Fix configure bug breaking --with-external-libdirJack Lloyd2017-05-131-0/+1
|/ | | | | | Reported by Marc Espie [ci skip]
* Error out if the configuration results in an empty libraryJack Lloyd2017-05-031-1/+4
| | | | | This can happen for example if a dependency of 'base' (block, hash, rng) is disabled. Noticed while testing #1034
* Rewrite dependency resolutionSimon Warta2017-04-301-28/+58
| | | | Closes #1028
* Improve debugging in ModulesChooser._validate_stateSimon Warta2017-04-301-2/+4
|
* Simplify implementation of makefile_listSimon Warta2017-04-301-2/+2
| | | | this makes it compatible with zero length collections
* Make configure script pylint2 complientSimon Warta2017-04-261-13/+12
|
* Refactor strip_header_goopSimon Warta2017-04-261-16/+24
|
* Simplify list emptyness checkSimon Warta2017-04-251-1/+1
|
* Let caller of _isa_specific_flags care for string representationSimon Warta2017-04-251-7/+10
|
* Remove extra code branch for house_ecc_curve_definesSimon Warta2017-04-251-5/+2
|
* Extract prepare_configure_build()Simon Warta2017-04-251-17/+23
|
* Split main method into main actionsSimon Warta2017-04-251-67/+78
|
* Rename info_* variables for consistencySimon Warta2017-04-251-11/+11
|
* Group info file loadingSimon Warta2017-04-251-12/+11
|
* Make argv non-optional in main()Simon Warta2017-04-251-8/+3
|
* Run --list-modules after options are checkedSimon Warta2017-04-251-5/+5
|
* Move Jython check to correct positionSimon Warta2017-04-251-3/+5
|
* Move os/cpu/cc detection to set_defaults_for_unset_optionsSimon Warta2017-04-251-38/+38
|
* Extract canonicalize_optionsSimon Warta2017-04-251-19/+22
|
* Move os and cc validation into validate_optionsSimon Warta2017-04-251-14/+12
|
* Move windows+gcc check to validate_optionsSimon Warta2017-04-251-3/+3
|
* Move policy option check into validate_optionsSimon Warta2017-04-251-10/+8
|