Commit Graph

159 Commits

Author SHA1 Message Date
Yves Brise 56221d9fd8 taking care of some more size_t vs. int warnings 2010-11-03 07:58:08 +00:00
Yves Brise a6d426f99b Putting int as the difference type in all counting_iterator transformations 2010-10-27 07:09:33 +00:00
Yves Brise 03f90cf9b8 Counting iterator with non-default template arguments in QP_models.h 2010-10-15 09:28:49 +00:00
Laurent Rineau 9277a77b72 Fix all svn:keywords in */include/ 2010-10-04 11:00:27 +00:00
Yves Brise abfd45595c reverting to counting_iterator<int> in QP_models.h 2010-09-30 17:32:05 +00:00
Yves Brise 34b8544354 trying to solve 64bit unsigned int problem again 2010-09-29 21:05:51 +00:00
Yves Brise 40a939272d 1. some more static_casts regarding the vector.size() problem. 2. using generic size_type in counting_iterators (just testing one place in QP_models.h) 2010-09-26 14:55:11 +00:00
Yves Brise fe42440b0f Start resolving the size_t vs. int comparison problems... (part 1) 2010-09-22 14:14:42 +00:00
Yves Brise 3b04af9287 fix index out of bounds bug in Initialization.h 2010-09-17 08:09:20 +00:00
Yves Brise df051d8a83 minor changes in comments, and some loops (signed vs unsigned warnings ) 2010-09-16 09:04:31 +00:00
Yves Brise e2bc8436ec signed vs unsigned warning fixed 2010-09-15 06:41:54 +00:00
Yves Brise 8a795d3a57 Implemented BLAND rule for leaving variable 2010-09-13 15:04:45 +00:00
Laurent Rineau a31222bd4c Fix a bug: non-inline non-template functions were defined in headers. 2010-09-06 11:25:20 +00:00
Yves Brise 4446bc60bf revert QP_solver_impl.h because of Bland vs. artificial problem 2010-07-19 21:10:45 +00:00
Yves Brise e640b46ac0 Fixed the Bland rule in the QP_solver. In particular, the leaving variable is now also chosed according to the Bland rule. 2010-07-15 10:22:45 +00:00
Sébastien Loriot 52317dd49f add python script (replace_CGAL_NAMESPACE.py) to replace CGAL_BEGIN_NAMESPACE and CGAL_END_NAMESPACE
by namespace CGAL { and } //namespace CGAL. in all .h and .cpp files
in a directory.
Apply it to all packages in the trunk
Remove macro definition from the config.h file.
2010-06-09 07:37:13 +00:00
Andreas Fabri de2da9d4e6 Add 'using' for ansi compliance 2010-05-17 07:19:52 +00:00
Andreas Fabri aa28266015 Add 'using' for ansi compliance 2010-05-16 13:35:52 +00:00
Andreas Fabri e4c680cba4 Add using and this for ansi-compliance 2010-05-13 13:47:44 +00:00
Andreas Fabri ceaf59ea56 Add 'this->' for ansi-compliance 2010-05-10 07:38:17 +00:00
Sylvain Pion 85dc12f2a7 Remove empty lines at beginning and end of files
(apply Scripts/developer_scripts/remove_empty_lines.pl).
2010-02-01 12:55:28 +00:00
Bernd Gärtner 4cdb07ecc3 last homegrown iterator replaced by boost-based solution 2008-10-23 14:31:10 +00:00
Bernd Gärtner b5a2bb1fec qp_x[i] -> *(qp_x+i) throughout, for all user-duplied iterators 2008-10-09 13:07:49 +00:00
Bernd Gärtner 3c686beb36 some more operator[]-related ad-hoc fixes for Polytope_distance_d 2008-10-09 10:06:21 +00:00
Bernd Gärtner c522b485c5 -some fixes to account for result of operator[] only being convertible to value type (at some point, this should sytematically be addressed by removing all occurrences of [] in the QP_solver code) 2008-10-09 09:42:02 +00:00
Bernd Gärtner 907223b777 First Boost 1.33.1 workaround (maybe others will be necessary as well) 2008-09-29 09:03:20 +00:00
Bernd Gärtner 081110184f - some homemade iterators replaced by boost variants 2008-07-31 20:55:38 +00:00
Bernd Gärtner 8612c935be - fixes to cope with removal of CGAL's funtional.h 2008-07-31 18:56:21 +00:00
Sylvain Pion 79864c0c8e Remove Arity_tag. 2008-07-22 12:29:01 +00:00
Sylvain Pion 2eb0d8f9e9 certainly() -> possibly(). 2008-07-14 11:16:41 +00:00
Sylvain Pion 9d7a609ff2 Make assertions trigger an error only for *certain* conditions (using CGAL::certainly()).
This is useful for interval arithmetic code.
(replay of r44002 with the header order hopefully fixed)
2008-07-10 21:49:17 +00:00
Sylvain Pion 2b4ebf01f7 Undo r44002 as it's buggy 2008-07-06 20:47:29 +00:00
Sylvain Pion 514a17ad4f Improve the interaction of assertions and interval computations.
Things like "CGAL_assertion(denominator != 0)" produced assertion failures for no gain.
So now, the assertion is triggered only if the condition is certain, using CGAL::certainly().
That is, change the following in all assertion files :
  ((EX)?(static_cast<void>(0)): ::CGAL::assertion_fail( # EX , __FILE__, __LINE__))
to :
  (CGAL::certainly(EX)?(static_cast<void>(0)): ::CGAL::assertion_fail( # EX , __FILE__, __LINE__))
2008-07-06 17:05:47 +00:00
Sylvain Pion d49e006c53 Remove default argument "0" to calls to assertion_fail() et al,
as the default is now the empty string "".
It should fix the problem that we have lost the assertion messages
(seeing "what():  basic_string::_S_construct NULL not valid" instead),
for packages that use package-specific assertion macros.
2007-12-21 21:58:27 +00:00
Bernd Gärtner b0cbc0defa - cosmetic output fix 2007-05-05 07:57:23 +00:00
Bernd Gärtner 5424f71850 - minor error/warning fixes 2007-05-01 15:55:12 +00:00
Bernd Gärtner a6aba7c40a - fixed file location in include
- renamed ...validation_flag() to ...auto_validation()
2007-04-30 10:16:02 +00:00
Bernd Gärtner 25a03549d4 - copyright years brought up-to-date
- (outdated) emails of everbody except me removed
2007-04-27 00:34:44 +00:00
Bernd Gärtner d7a098033c - added another set of random tests (with double input, and using
a different model)
- fixed memory leak in QP_solution
- refactored validity checks a little
2007-04-26 23:56:44 +00:00
Bernd Gärtner a9be004d5a - added the interface validation methods to examples,
and referred to them from the doc (+ typo fix)
2007-04-25 01:33:32 +00:00
Bernd Gärtner 0bc86dea9d - moved certificate check out of the solver and into solution
- at the same time, we don't check anymore by default, but there
  is an option to trigger a self-check
2007-04-25 01:19:38 +00:00
Bernd Gärtner 060f6bf342 - minor refactoring; test_solver now also calls the interface
methods (solve...) instead of the QP_solver directly
2007-04-23 09:12:34 +00:00
Bernd Gärtner 6a79d4f45d - some more manual fixes
- made partial pricing also work for high constraint/variable ratio (needs to
  be properly tested)
- added QSCORPIO as interesting additional example 
- switched to quotient output in iteration log to avoid nans
- removed filter_failure example (after all, this particular example
  was just an artifact of the bad to_double() routine of Gmpzf that I
  fixed earlier
2007-04-20 12:31:57 +00:00
Bernd Gärtner 49ca59733c - user manual:
- added remark about possible filter failures
    - added example where they occur, and how to fix them
    - added benchmark section
- ref manual:
    - documented pricing strategies
- code:
    - made partial_filtered_dantzig the default for double input
    - added timer output to test_solver
2007-04-19 14:36:10 +00:00
Bernd Gärtner 4f4a806c0f - tested name <-> index mapping in mps reader
- fix in handling names in ranges section
2007-04-18 19:23:21 +00:00
Bernd Gärtner 7f04b33cdb - minor cleanup:
- non-debugged vout in pricing strategies removed
        - debug.h adapted to react to other things than D_NDEBUG as well
2007-04-17 20:52:11 +00:00
Bernd Gärtner 558c2e3c1e - Ron's fixes
- some debug output tweaking
2007-04-17 17:54:17 +00:00
Bernd Gärtner 159679d480 - QP_options put into manual, with cycling example 2007-04-15 19:36:07 +00:00
Bernd Gärtner 4559516680 - user manual rewritten and almost complete
- added class Quadratic_program_options to customize behavior, still
  need to document it
2007-04-15 16:49:56 +00:00
Bernd Gärtner 8bf9b65886 - QP_status -> Quadratic_program_status 2007-04-14 18:32:37 +00:00