Age | Commit message (Collapse) | Author | Files | Lines | |
---|---|---|---|---|---|
2013-06-13 | solver: prune broken world dependencies with --force | Timo Teräs | 1 | -22/+54 | |
mostly useful for reboot, when all packages are not available. | |||||
2013-06-13 | solver: corner case fixes | Timo Teräs | 1 | -3/+19 | |
- try harder to not change anything during self-upgrade - also honor locking to packages that where earlier used in merging common dependencies - clarify upgrade applet help messages | |||||
2013-06-13 | solver: rewrite as deductive solver -- pinning support | Timo Teräs | 1 | -26/+135 | |
Fix also pinning test cases to be more sane. | |||||
2013-06-13 | solver: rewrite as deductive solver -- per name flags | Timo Teräs | 1 | -8/+24 | |
Handle properly per-name preference flags, and add test cases for testing those via fix applet. | |||||
2013-06-13 | solver: rewrite as deductive solver -- core features | Timo Teräs | 1 | -1830/+418 | |
Implementing basic dependency handling, install_if and awareness of pinning. | |||||
2013-06-12 | Revert "Revert "solver: increase score fields to 32-bits (from 16-bits)"" | William Pitcock | 1 | -21/+27 | |
This reverts commit a2d873a77c6dd2e7f6219e6941af796e1f904e69. | |||||
2013-06-11 | Revert "solver: increase score fields to 32-bits (from 16-bits)" | root | 1 | -27/+21 | |
This reverts commit 84bfef1a6b587a7da7d12fb701ab0d1d5d6ce2a9. | |||||
2013-05-30 | db: unify handling of special packages | Timo Teräs | 1 | -8/+2 | |
make cache a special kind of repository, and automatically cache special packages (virtual packages, or ones installed from command line). add test cases for handling virtual packages. fixes #1617. | |||||
2013-04-29 | solver: increase score fields to 32-bits (from 16-bits) | Timo Teräs | 1 | -21/+27 | |
We are having so many packages that they might overflow otherwise. "ERROR: Preference overflow" was already reported. | |||||
2013-01-28 | solver: remove backjumping by name | Timo Teräs | 1 | -28/+3 | |
It is incorrect optimization causing valid solutions to be skipped. Any performance it might've gained, should be fixed by reintroduction of the minimum penalty logic added in previous commit. | |||||
2013-01-28 | solver: reintroduce minimum penalty logic | Timo Teräs | 1 | -49/+55 | |
Basic per-name per-package specific scoring added. | |||||
2013-01-22 | solver: do not backtrack over "exclude none" | Timo Teräs | 1 | -1/+8 | |
Otherwise we can return no solution at all instead of partially satisfied solution (and give proper error message). | |||||
2013-01-21 | solver: simplify strength evaluation in constraint handling | Timo Teräs | 1 | -20/+8 | |
2013-01-21 | solver: check if solution is not found at all | Timo Teräs | 1 | -0/+6 | |
This is an internal error. | |||||
2013-01-18 | solver: fix reset of last_touched to zero | Timo Teräs | 1 | -2/+2 | |
2013-01-18 | solver: some code clean ups | Timo Teräs | 1 | -5/+5 | |
2013-01-18 | solver: build fix for debug print on x86_64 | Natanael Copa | 1 | -1/+1 | |
use %zu for size_t | |||||
2013-01-18 | solver: fix name assignment logic to not mess up internal state | Timo Teräs | 1 | -11/+10 | |
Should get rid of error messages like: ERROR: Saved_score {4/0/0,74} != score {6/0/0,74} | |||||
2012-11-04 | solver: fix package prefence calculation a bit | Timo Teräs | 1 | -1/+3 | |
2012-10-08 | solver: additional backjumping scheme | Timo Teräs | 1 | -11/+48 | |
Enabled when all attempts to satisfy a name failed, we know that we can ignore all decisions until we find a decision affecting the name we wanted to satisfy. | |||||
2012-10-08 | solver: optimize backjumping | Timo Teräs | 1 | -7/+7 | |
to be functional when backtracking | |||||
2012-10-08 | solver: properly calculate decision "strength" with provides | Timo Teräs | 1 | -17/+13 | |
2012-10-05 | solver: fix back jumping once more | Timo Teräs | 1 | -17/+33 | |
2012-10-05 | solver: record dependency apk_name in apk_decision | Timo Teräs | 1 | -51/+19 | |
We can't just use the primary name, as that would mess up backtracking. We need to record the name which caused the name to get considered - that way the right last_touched_decision is used on backtracking. | |||||
2012-10-03 | solver: various fixes | Timo Teräs | 1 | -14/+58 | |
* push_decision expects to always get the package primary 'name' as apk_name. ASSERT that and fix problem cases. (though - this might need to be reverted, and store the non primary name in apk_decision instead to accomodate for better backtracking optimizations) * fix error reporting of virtual package names * make 'assign_name' errors soft. the incorrect packages just are no longer consider instead of aborting whole calculation. * fix backtracking of virtual packages that are not depended directly | |||||
2012-10-03 | solver: improve assert error messages | Natanael Copa | 1 | -2/+3 | |
Print name of package and void double "ERROR" | |||||
2012-10-03 | solver: fix compile with -Werror on 64bit | Natanael Copa | 1 | -1/+1 | |
2012-09-20 | apk: implement --progress-fd to write progress to a specified fd | Timo Teräs | 1 | -17/+27 | |
2012-04-27 | solver: fix install_if triggers on cyclic and early deps | Timo Teräs | 1 | -11/+15 | |
2012-04-27 | solver: fix backtracking with install_if dependencies | Timo Teräs | 1 | -17/+20 | |
2012-03-31 | solver: fix a type from previous commit | Timo Teräs | 1 | -1/+1 | |
2012-03-31 | solver: fix installation of non-repository packages | Timo Teräs | 1 | -1/+2 | |
2012-03-01 | solver: properly set allowed flag for installed packages | Timo Teräs | 1 | -3/+6 | |
2012-02-29 | solver: rename some fields | Timo Teräs | 1 | -24/+25 | |
2012-02-29 | solver: merge apk_name_state to apk_name | Timo Teräs | 1 | -223/+136 | |
apk_name_state is now quite small; and we avoid overhead of two pointers (+ malloc overhead) when we just make it part of apk_name. It also fixes some problems (that got introduced) where apk_name_state was not allocated. | |||||
2012-02-29 | solver, test: make conflicts unconditional | Timo Teräs | 1 | -13/+36 | |
Solver will now never report partial solution where a conflict constraint is not satisfied. The is because with --force we might install the partial solution; and if conflicted packages were to be installed we might have extra trouble. | |||||
2012-02-28 | solver: ask confirmation in interactive mode only if there's changes | Timo Teräs | 1 | -1/+2 | |
2012-02-28 | solver: do not consider non-allowed packages in main loop | Timo Teräs | 1 | -43/+32 | |
Instead cache the allowed pinning decision, and use it. Update install decision heuristic to also use this cached information. | |||||
2012-02-28 | solver: consider provided names also for preference | Timo Teräs | 1 | -2/+13 | |
ref #574 | |||||
2012-02-28 | solver: fix conflicting provides detection | Timo Teräs | 1 | -2/+15 | |
ref #574 | |||||
2012-02-28 | solver: allow multiple packages with same virtual provides | Timo Teräs | 1 | -8/+14 | |
ref #574 | |||||
2012-02-27 | solver, test: implements more provides things, add tests | Timo Teräs | 1 | -30/+92 | |
ref #574 | |||||
2012-02-27 | solver: have most inherited things per-package and clean ups | Timo Teräs | 1 | -156/+168 | |
Required for provides support as package might be pulled in via non-primary package name. This allows relatively easily to pass through inherited flags via the provided names. ref #574. | |||||
2012-02-27 | solver: remove minimum penalty logic | Timo Teräs | 1 | -49/+7 | |
Reasoning: - it is less useful now that we do not do common dependency merging - provides support would make the required logic overly complicated - callgrind reports that depending on the case it can improve or decrease performance (the overhead pays off only in some cases); the difference is not large either way | |||||
2012-02-24 | solver, dot: elementary provides fixes | Timo Teräs | 1 | -21/+57 | |
implementation is still not near finished, but now at least it can handle it to a minimum degree. many cases are not done right yet, though. ref #574. | |||||
2012-02-24 | all: introduce apk_provides and use it in apk_name | Timo Teräs | 1 | -73/+79 | |
in preparation for provides support. implements also some dependency satisfaction helper routines. ref #574. | |||||
2012-02-24 | solver: unallowed pinning is worse than changing installed package | Timo Teräs | 1 | -10/+7 | |
2012-02-24 | solver: non preferred actions are worse then non preferred pinning | Timo Teräs | 1 | -6/+16 | |
Otherwise we might start to change packages unexpectedly when not upgrading. This also fixes some other things the solver might've decided to do. Add also few test cases to detect bad behaviour. | |||||
2012-02-24 | test: improve pinning tests | Timo Teräs | 1 | -1/+4 | |
2012-02-23 | solver: fix output of broken dependencies | Timo Teräs | 1 | -1/+1 | |