aboutsummaryrefslogtreecommitdiff
path: root/build2/cc
AgeCommit message (Collapse)AuthorFilesLines
2019-08-23dry_runBoris Kolpackov6-25/+33
2019-08-23Get rid of var_extension char arrayBoris Kolpackov1-4/+4
2019-08-23Minor fixKaren Arutyunov1-1/+1
2019-08-23VariablesBoris Kolpackov1-4/+9
2019-08-23Meta/operations and countsBoris Kolpackov2-7/+8
2019-08-22Run phaseBoris Kolpackov3-5/+5
2019-08-22Targets, scopes, varsBoris Kolpackov9-64/+88
2019-08-22Make use of butl::sanitize_identifier()Karen Arutyunov1-9/+2
2019-08-21Cleanup context.hxx and its usageBoris Kolpackov5-5/+0
2019-08-17Revert testing code (link name on Windows)Boris Kolpackov1-2/+1
2019-08-16Fix bugs in derive_libs_paths() logicBoris Kolpackov1-23/+29
2019-08-16Cosmetic changesBoris Kolpackov1-2/+2
2019-08-15Redo Windows implementation of install_l() to use uninstall_f()Boris Kolpackov1-2/+2
2019-08-15Fix bug in mkanylink() usageBoris Kolpackov1-5/+5
2019-08-15Minor naming improvementBoris Kolpackov2-6/+6
2019-08-15Add support for bin.lib.load_suffixBoris Kolpackov3-8/+34
This allow the creation of yet another symlink to the shared library that is meant to be used for dynamic loading. For example, we may want to embed the main program interface number into its plugins to make sure that we only load compatible versions.
2019-08-15Fix bug in bin.lib.suffix handlingBoris Kolpackov1-19/+13
2019-08-15Implement libs_paths symlinking support on WindowsBoris Kolpackov1-8/+26
Also, temporarily enable libs_paths::link on Windows for testing.
2019-08-14Remove no longer used lambda captureBoris Kolpackov1-1/+1
2019-08-14Use new mkanylink() from libbutlBoris Kolpackov1-52/+20
2019-08-14Enable libs_paths symlinking logic on WindowsBoris Kolpackov2-25/+28
This should currently be noop since all the paths are set to empty.
2019-08-14Make libs_path::link empty instead of import library path for WindowsBoris Kolpackov2-21/+24
We don't really use this anywhere (in particular, the import library name cannot possibly match the clean pattern).
2019-08-06Improve module name to file name heuristicsBoris Kolpackov1-25/+101
2019-07-26Use .gcm extension for both modules and headers in GCCBoris Kolpackov1-12/+1
2019-07-11Improve diagnosticsBoris Kolpackov1-4/+5
2019-07-05Fix incorrect traceBoris Kolpackov1-1/+1
2019-07-05Move config, dist, test, and install modules into libraryKaren Arutyunov5-6/+7
2019-07-04Remove libu{} target groupBoris Kolpackov4-32/+24
The semantics provided by libu{} is rarely required and as result was not yet documented. However, if you are using it, the new way to achieve the same result is to use both libue{} and libul{} explicitly, for example: exe{foo}: libue{foo} lib{foo}: libul{foo} {libue libul}{foo}: cxx{*}
2019-07-02Minor improvementsBoris Kolpackov1-1/+1
2019-07-01Split build system into library and driverBoris Kolpackov28-108/+108
2019-06-27Fix member-group linkup issue in previous fixBoris Kolpackov2-58/+87
We cannot linkup prerequisite targets since we are not matching them.
2019-06-26Fix C/C++ link rule matching ambiguity by seeing-through utility librariesBoris Kolpackov7-45/+151
2019-06-24Constrain access to options to build system driver main() onlyBoris Kolpackov1-2/+2
2019-06-21Add support for [config.]{cc,c,cxx}.aoptions (archive options)Boris Kolpackov4-11/+23
In particular, this can be used to suppress lib.exe warnings, for example: cc.aoptions += /IGNORE:4221
2019-06-19Improve diagnosticsBoris Kolpackov4-13/+17
2019-06-18Fix crashing on unhandled invalid_path thrown by extract_library_dirs()Karen Arutyunov1-23/+31
2019-06-07Fix /usr/local/{include,lib} logic for Mac OS 10.14Boris Kolpackov1-0/+45
Turns out there /usr/include is no more.
2019-06-07Update module/header unit-related notesBoris Kolpackov1-2/+5
2019-06-07Reply with immediate BMI in case of include translationBoris Kolpackov1-31/+22
2019-06-07Complete and normalize paths in cxx.importable_headersBoris Kolpackov1-22/+37
Also add more header unit include/import tests.
2019-06-06Fix cc link rule matching ambiguity wrt to C/C++ sourcesBoris Kolpackov3-19/+30
Considering a C header as C++ source was definitely a bad idea.
2019-06-06Make preprocessing options order consistent across preprocessing and ↵Karen Arutyunov1-5/+5
compilation compiler runs
2019-06-06Add noteBoris Kolpackov1-0/+5
2019-06-06Redo header path normalization/realization logicBoris Kolpackov1-18/+68
We now try to use the normalized path (which preserves symlinks) if possible and fall back to realized otherwise.
2019-06-05Cosmetic changeBoris Kolpackov1-4/+2
2019-06-05Tweak generated header heuristics, add more diagnosticsBoris Kolpackov1-1/+37
2019-06-05Adjust GCC module mapper to new protocolBoris Kolpackov1-44/+79
2019-06-04Fix header dependency generating with ClangKaren Arutyunov1-1/+1
2019-06-03Adapt to renaming traits alias to traits_type for basic_path, basic_url, and ↵Karen Arutyunov6-22/+23
string_table class templates
2019-06-03Add note on GCC optionsBoris Kolpackov1-0/+2
generated by cgit v1.1 at 2024-04-16 22:31:34 +0000