aboutsummaryrefslogtreecommitdiff
path: root/libbuild2/algorithm.cxx
AgeCommit message (Collapse)AuthorFilesLines
2024-02-26Add ability to request serialization from schedulerBoris Kolpackov1-1/+1
In particular, this can be used to make sure no other recipe is being executed in parallel with the caller.
2024-02-22Diagnose instead of asserting ad hoc group member not already existingBoris Kolpackov1-21/+18
Fixes GH issue #365.
2024-02-22Deal with libs{} being member of group in windows_rpath_timestamp()Boris Kolpackov1-1/+0
Fixes GH issue #366.
2024-02-14Add search_prerequisite*() variants of match_prerequisite*() versionsBoris Kolpackov1-7/+26
2024-02-14Avoid duplicate fsdir{} in inject_fsdir(), match_prerequisite*() call sequencesBoris Kolpackov1-4/+27
2024-02-13Add ability to omit matching in inject_fsdir()Boris Kolpackov1-2/+4
2024-01-16Add add_adhoc_member_identity(), use to fix ad hoc pattern rule logicBoris Kolpackov1-0/+43
2024-01-11Properly split injected ad hoc group member name in regex pattern ruleBoris Kolpackov1-2/+3
2023-12-03Reimplement search_existing() functions via target_type::searchBoris Kolpackov1-9/+13
This allows us to automatically get the target type-specific behavior with regards to the out_only semantics (added in the previous commit) instead of passing it explicitly from each call site.
2023-12-03Search in src for existing prerequisites with unspecified outBoris Kolpackov1-8/+4
2023-11-07Account for match options re-locking when checking if target is matchedBoris Kolpackov1-0/+15
2023-11-03Fix data race in match options logicBoris Kolpackov1-1/+22
2023-11-01Tighten/optimize cleaning of fsdir{} during matchBoris Kolpackov1-1/+10
2023-11-01Add support for adjusting match options of post hoc prerequisitesBoris Kolpackov1-10/+69
2023-11-01Add notion of match optionsBoris Kolpackov1-102/+220
Now, when matching a rule, the caller may request a subset of the full functionality of performing an operation on a target. This is achieved with match options.
2023-10-26Add inject_fsdir_direct(), fsdir_rule::perform_clean_direct() functionsBoris Kolpackov1-1/+24
2023-10-26Add clean_during_match*() functionsBoris Kolpackov1-0/+180
2023-10-06Handle 0 mask in update_during_match_prerequisites()Boris Kolpackov1-4/+4
2023-07-20Change inner rule/prerequisites match order in install::file_ruleBoris Kolpackov1-0/+14
The old order messed up the for-install signaling logic. See the long comment in install::file_rule::apply_impl() for background and details.
2023-06-12Hack around GCC 13 -Wdangling-reference false positivesBoris Kolpackov1-1/+1
See GCC bugs 107532, 110213.
2023-06-08Add support for buildfile importationBoris Kolpackov1-2/+2
2023-06-01Resolve (but disable for now) target_count issue in resolve_members()Boris Kolpackov1-22/+50
2023-05-31Provide implied configure_* and dist_* action registration for ad hoc recipesBoris Kolpackov1-27/+73
This makes it consistent with the existing ad hoc rules semantics.
2023-05-29Extend special match_rule() logic to all groups with dynamic targetsBoris Kolpackov1-24/+27
2023-05-29Disable counts verification if matched but not executed in resolve_members()Boris Kolpackov1-0/+7
2023-05-29Explicit group: static membersBoris Kolpackov1-10/+71
2023-04-05Allow creating context with bare minimum of initializationsBoris Kolpackov1-25/+25
This is used by bpkg to detect forwarded configurations without incurring the full context creation overhead.
2022-12-12Adapt to dir_iterator API changeKaren Arutyunov1-2/+1
2022-11-29Fix backlink logic for target groupsBoris Kolpackov1-107/+266
We used to backlink ad hoc group members both via the group and as individual members. And for explicit groups it was done only via individual members, which means it only works correctly if every member is individually updated. Now both types of groups are backlinked from the group target.
2022-11-23Take into account ad hoc recipes in rule::sub_match() (fixed GH issue #227)Boris Kolpackov1-61/+69
2022-11-18Complete low verbosity diagnostics reworkBoris Kolpackov1-12/+41
2022-11-16Initial low verbosity diagnostics reworkBoris Kolpackov1-13/+50
2022-10-20Don't create targets for non-existent source filesBoris Kolpackov1-2/+14
2022-10-19Handle operation-specific variable values in post hoc logicBoris Kolpackov1-1/+19
2022-10-19Minor tweaks to target_lock passing semanticsBoris Kolpackov1-2/+8
This should also get rid of the bogus -Wdangling-pointer issued by GCC 12.
2022-10-19Add support for post hoc prerequisitesBoris Kolpackov1-50/+164
Unlike normal and ad hoc prerequisites, a post hoc prerequisite is built after the target, not before. It may also form a dependency cycle together with normal/ad hoc prerequisites. In other words, all this form of dependency guarantees is that a post hoc prerequisite will be built if its dependent target is built. See the NEWS file for details and an example.
2022-09-19Add target_lock::first to distinguish first lock of targetBoris Kolpackov1-7/+9
2022-09-12Do not treat primary ad hoc group member as group for variable lookupBoris Kolpackov1-4/+6
Note that we started with this semantics but it was changed in a commit on 2021-09-16 for reasons not entirely unclear but most likely due to target- specific variables specified for the group not being set on all the members. Which we have now addressed (see the previous commit). Note also that this new (old) semantics is not without its own drawbacks. Specifically, there is a bit of waste when the target-specific variable is really only meant for the recipe and thus setting it on all the members is unnecessary. For example: <{hxx ixx cxx}{options}>: cli{options} { options = ... } {{ # Use options. }} But this feels like a quality of implementation rather than conceptual issue. For example, we could likely one day address it by synthesizing a separate group target for ad hoc groups.
2022-09-07Fix assert after cycle detection in ad hoc groupsBoris Kolpackov1-31/+31
2022-09-07Fix fsdir{} handling corner cases in ad hoc buildscript recipes/rulesdist-remapBoris Kolpackov1-2/+2
2022-07-13Work around "storing address of local variable" warning issued by GCC 12Boris Kolpackov1-1/+1
The warning is issued inside the target_lock's move constructor so we sidestep the whole thing by avoiding the call to the move constructor.
2022-06-23Minor enhancements to algorithmsBoris Kolpackov1-10/+49
2022-06-21Add --trace-{match,execute} optionsBoris Kolpackov1-1/+95
These options can be used to understand which dependency chain causes matching or execution of a particular target.
2022-04-19Use target recipe for auxiliary data storage during match-applyBoris Kolpackov1-4/+5
In particular, we now have separate auxiliary data storage for inner and outer operations.
2022-04-19Skip find() inside target_set::insert*() if target is unlikely to be thereBoris Kolpackov1-1/+2
2022-04-19Cache "recipe is group_action" information, clear recipes after executionBoris Kolpackov1-6/+12
2022-04-18Avoid locking target set if in load phaseBoris Kolpackov1-16/+22
2022-04-08Optimize update_during_match()Boris Kolpackov1-7/+14
2022-04-08Add hint-less rule match pass for non-perform meta-operations as fallbackBoris Kolpackov1-116/+143
2022-04-07Rename {match,execute}() to *_sync(), add *_complete()Boris Kolpackov1-59/+35
In particular, the match() rename makes sure it doesn't clash with rule::match() which, after removal of the hint argument in simple_rule, has exactly the same signature, thus making it error-prone to calling recursively.