COIN-OR::LEMON - Graph Library

source: lemon-project-template-glpk/deps/glpk/NEWS @ 9:33de93886c88

subpack-glpk
Last change on this file since 9:33de93886c88 was 9:33de93886c88, checked in by Alpar Juttner <alpar@…>, 13 years ago

Import GLPK 4.47

File size: 71.6 KB
Line 
1GLPK 4.47 (release date: Sep 09, 2011)
2
3        The new API routine glp_intfeas1 was added to the package.
4        This routine is a tentative implementation of the integer (0-1)
5        feasibility solver based on the CNF-SAT solver (which currently
6        is MiniSat). It may be used in the same way as glp_intopt to
7        find either any integer feasible solution or a solution, for
8        which the objective function is not worse than the specified
9        value. Detailed description of this routine can be found in the
10        document "CNF Satisfiability Problem", which is included in the
11        distribution (see doc/cnfsat.pdf).
12
13        The following two options were added to glpsol:
14
15        --minisat       translate 0-1 feasibility problem to CNF-SAT
16                        problem and solve it with glp_intfeas1/MiniSat
17                        (if the problem instance is already in CNF-SAT
18                        format, no translation is performed)
19
20        --objbnd bound  add inequality obj <= bound (minimization) or
21                        obj >= bound (maximization) to 0-1 feasibility
22                        problem (this option assumes --minisat)
23
24        The paint-by-numbers puzzle model (pbn.mod) included in the
25        distribution is a nice example of the 0-1 feasibility problem,
26        which can be efficiently solved with glp_intfeas1/MiniSat. This
27        model along with a brief instruction (pbn.pdf) and benchmark
28        examples from <webpbn.com> encoded in GNU MathProg (*.dat) can
29        be found in subdirectory examples/pbn/.
30
31        The glpsol lp/mip solver was modified to bypass postprocessing
32        of MathProg models if the solution reported is neither optimal
33        nor feasible.
34
35        A minor bug in examples/Makefile.am was fixed to correctly
36        build glpk in a separate directory. Thanks to Marco Atzeri
37        <marco.atzeri@gmail.com> for bug report and patch.
38
39GLPK 4.46 (release date: Aug 09, 2011)
40
41        The following new API routines were added:
42
43        glp_read_cnfsat    read CNF-SAT problem data in DIMACS format
44        glp_check_cnfsat   check for CNF-SAT problem instance
45        glp_write_cnfsat   write CNF-SAT problem data in DIMACS format
46        glp_minisat1       solve CNF-SAT problem instance with MiniSat
47
48        The routine glp_minisat1 is a driver to MiniSat, a CNF-SAT
49        solver developed by Niklas Een and Niklas Sorensson, Chalmers
50        University of Technology, Sweden. This routine is similar to
51        the routine glp_intopt, however, it is intended to solve a 0-1
52        programming problem instance, which is the MIP translation of
53        a CNF-SAT problem instance.
54
55        Detailed description of these new API routines can be found in
56        the document "CNF Satisfiability Problem", which is included in
57        the distribution (see files doc/cnfsat.tex and doc/cnfsat.pdf).
58
59        The following new glpsol command-line options were added:
60
61        --cnf filename     read CNF-SAT problem instance in DIMACS
62                           format from filename and translate it to MIP
63        --wcnf filename    write CNF-SAT problem instance in DIMACS
64                           format to filename
65        --minisat          solve CNF-SAT problem instance with MiniSat
66                           solver
67
68        The zlib compression library (version 1.2.5) was ANSIfied,
69        modified according to GLPK requirements and included in the
70        distribution as an external software module. Thus, now this
71        feature is platform independent.
72
73        Some bugs were fixed in the SQL table driver. Thanks to Xypron
74        <xypron.glpk@gmx.de>.
75
76GLPK 4.45 (release date: Dec 05, 2010)
77
78        This is a bug-fix release.
79
80        Several bugs/typos were fixed. Thanks to
81        Xypron <xypron.glpk@gmx.de>,
82        Robbie Morrison <robbie@actrix.co.nz>, and
83        Ali Baharev <ali.baharev@gmail.com> for reports.
84
85        Some glpk documents was re-formatted and merged into a single
86        document. Now the glpk documentation consists of the following
87        three main documents (all included in the distribution):
88
89        GLPK: Reference Manual
90
91        GLPK: Graph and Network Routines
92
93        Modeling Language GNU MathProg: Language Reference
94
95GLPK 4.44 (release date: Jun 03, 2010)
96
97        The following suffixes for variables and constraints were
98        implemented in the MathProg language:
99
100        .lb     (lower bound),
101        .ub     (upper bound),
102        .status (status in the solution),
103        .val    (primal value), and
104        .dual   (dual value).
105
106        Thanks to Xypron <xypron.glpk@gmx.de> for draft implementation
107        and testing.
108
109        Now the MathProg language allows comment records (marked by
110        '#' in the very first position) in CSV data files read with the
111        table statements. Note that the comment records may appear only
112        in the beginning of a CSV data file.
113
114        The API routine glp_cpp to solve the Critical Path Problem was
115        added and documented.
116
117GLPK 4.43 (release date: Feb 20, 2010)
118
119        This is a maintainer release.
120
121        `configure.ac' was changed to allow building the package under
122        Mac OS and Darwin with ODBC support.
123        Thanks to Xypron <xypron.glpk@gmx.de> for suggestions and Noli
124        Sicad <nsicad@gmail.com> for testing.
125
126        The SQL table driver was improved to process NULL data. Thanks
127        to Xypron <xypron.glpk@gmx.de>.
128
129        Some bugs were fixed in the LP/MIP preprocessor.
130
131GLPK 4.42 (release date: Jan 13, 2010)
132
133        The following new API routines were added:
134
135        glp_check_dup         check for duplicate elements in sparse
136                              matrix
137        glp_sort_matrix       sort elements of the constraint matrix
138        glp_read_prob         read problem data in GLPK format
139        glp_write_prob        write problem data in GLPK format
140        glp_analyze_bound     analyze active bound of non-basic variable
141        glp_analyze_coef      analyze objective coefficient at basic
142                              variable
143        glp_print_ranges      print sensitivity analysis report (this
144                              routine replaces lpx_print_sens_bnds and
145                              makes it deprecated)
146
147        For description of these new routines and the GLPK LP/MIP
148        format see a new edition of the reference manual included in
149        the distribution. (Chapter "Graph and network API routines" was
150        carried out from the main reference manual and included in the
151        distribution as a separate document.)
152
153        The following new command-line options were added to the stand-
154        alone solver glpsol:
155        --glp filename        read problem data in GLPK format
156        --wglp filename       write problem data in GLPK format
157        --ranges filename     print sensitivity analysis report (this
158                              option replaces --bounds)
159
160        Now all GLPK routines performing file I/O support special
161        filenames "/dev/stdin", "/dev/stdout", and "/dev/stderr", which
162        can be specified in the same way as regular filenames. This
163        feature is plaform-independent.
164
165GLPK 4.41 (release date: Dec 21, 2009)
166
167        The following new API routies were added:
168
169        glp_transform_row     transform explicitly specified row
170        glp_transform_col     transform explicitly specified column
171        glp_prim_rtest        perform primal ratio test
172        glp_dual_rtest        perform dual ratio test
173
174        For description of these new routines see a new edition of the
175        reference manual included in the distribution.
176
177        The following API routines are deprecated: lpx_transform_row,
178        lpx_transform_col, lpx_prim_ratio_test, lpx_dual_ratio_test.
179
180        Some improvements were made in the MIP solver (glp_intopt).
181
182        The SQL table driver used to read/write data in MathProg models
183        was changed to allow multiple arguments separated by semicolon
184        in SQL statements. Thanks to Xypron <xypron.glpk@gmx.de>.
185
186        Two new options were added to the glpsol stand-alone solver:
187        --seed value (to initialize the pseudo-random number generator
188        used in MathProg models with specified value), and
189        --ini filename (to use a basis previously saved with -w option
190        as an initial basis on solving similar LP's).
191
192        Two new MathProg example models were included. Thanks to
193        Nigel Galloway <nigel_galloway@operamail.com> and Noli Sicad
194        <nsicad@gmail.com> for contribution.
195
196        Scripts to build GLPK with Microsoft Visual Studio 2010 for
197        both 32-bit and 64-bit Windows were included. Thanks to Xypron
198        <xypron.glpk@gmx.de> for contribution and testing.
199
200GLPK 4.40 (release date: Nov 03, 2009)
201
202        The following new API routines were added:
203
204        glp_del_vertices      remove vertices from graph
205        glp_del_arc           remove arc from graph
206        glp_wclique_exact     find maximum weight clique with the exact
207                              algorithm developed by Prof. P. Ostergard
208        glp_read_ccdata       read graph in DIMACS clique/coloring
209                              format
210        glp_write_ccdata      write graph in DIMACS clique/coloring
211                              format
212
213        For description of these new routines see a new edition of the
214        reference manual included in the distribution.
215
216        The hybrid pseudocost branching heuristic was included in the
217        MIP solver. It is available on API level (iocp.br_tech should
218        be set to GLP_BR_PCH) and in the stand-alone solver glpsol
219        (via the command-line option --pcost). This heuristic may be
220        useful on solving hard MIP instances.
221
222        The branching heuristic by Driebeck and Tomlin (used in the
223        MIP solver by default) was changed to switch to branching on
224        most fractional variable if an lower bound of degradation of
225        the objective is close to zero for all branching candidates.
226
227        A bug was fixed in the LP preprocessor (routine npp_empty_col).
228        Thanks to Stefan Vigerske <stefan@math.hu-berlin.de> for the
229        bug report.
230
231        A bug was fixed and some improvements were made in the FPUMP
232        heuristic module. Thanks to Xypron <xypron.glpk@gmx.de>.
233
234        A bug was fixed in the API routine glp_warm_up (dual
235        feasibility test was incorrect in maximization case). Thanks to
236        Uday Venkatadri <Uday.Venkatadri@dal.ca> for the bug report.
237
238GLPK 4.39 (release date: Jul 26, 2009)
239
240        The following new API routines were added:
241
242        glp_warm_up           "warm up" LP basis
243        glp_set_vertex_name   assign (change) vertex name
244        glp_create_v_index    create vertex name index
245        glp_find_vertex       find vertex by its name
246        glp_delete_v_index    delete vertex name index
247        glp_read_asnprob      read assignment problem data in DIMACS
248                              format
249        glp_write_asnprob     write assignment problem data in DIMACS
250                              format
251        glp_check_asnprob     check correctness of assignment problem
252                              data
253        glp_asnprob_lp        convert assignment problem to LP
254        glp_asnprob_okalg     solve assignment problem with the
255                              out-of-kilter algorithm
256        glp_asnprob_hall      find bipartite matching of maxumum
257                              cardinality with Hall's algorithm
258
259        Also were added some API routines to read plain data files.
260
261        The API routines glp_read_lp and glp_write_lp to read/write
262        files in CPLEX LP format were re-implemented. Now glp_write_lp
263        correctly writes double-bounded (ranged) rows by introducing
264        slack variables rather than by duplicating the rows.
265
266        For description of these new routines see a new edition of the
267        reference manual included in the distribution.
268
269        The 'xfree(NULL)' bug was fixed in the AMD routines. Thanks to
270        Niels Klitgord <niels@bu.edu> for bug report.
271
272        The message "Crashing..." was changed to "Constructing initial
273        basis..." due to suggestion by Thomas Kahle <tom111@gmx.de>.
274
275        Some typos were corrected in glpsol output messages. Thanks to
276        Xypron <xypron.glpk@gmx.de> for patch.
277
278GLPK 4.38 (release date: May 02, 2009)
279
280        API routines glp_read_mps and glp_write_mps were improved.
281
282        Some improvements were made in the dual simplex routines.
283
284        Two external software modules AMD and COLAMD were included in
285        the distribution (for more details please see src/amd/README
286        and src/colamd/README). Now they are used in the interior-point
287        solver to reorder the matrix prior to Cholesky factorization.
288
289        API routine glp_ipt_status may return two new statuses due to
290        changes in the routine glp_interior. For details please see the
291        reference manual included in the distribution.
292
293        A minor bug was fixed in the graph/network routines. Thanks to
294        Nelson H. F. Beebe <beebe@math.utah.edu> for bug report.
295
296GLPK 4.37 (release date: Mar 29, 2009)
297
298        The 0-1 Feasibility Pump heuristic was included in the GLPK
299        integer optimizer glp_intopt. On API level the heuristic can be
300        enabled by setting the parameter fp_heur in glp_iocp to GLP_ON.
301        This feature is also available in the solver glpsol through
302        command-line option '--fpump'. For more details please see the
303        reference manual included in the distribution.
304
305        The following new API routines were added:
306
307        glp_print_sol         write basic solution in printable format
308        glp_print_ipt         write interior-point solution in printable
309                              format
310        glp_print_mip         write MIP solution in printable format
311        glp_read_graph        read (di)graph from plain text file
312        glp_write_graph       write (di)graph to plain text file
313        glp_weak_comp         find all weakly connected components
314        glp_strong_comp       find all strongly connected components
315
316        The following API routines are deprecated: lpx_print_sol,
317        lpx_print_ips, lpx_print_mip, lpx_print_prob (the latter is
318        equivalent to glp_write_lp).
319
320        A bug was fixed in the interior-point solver (glp_interior) to
321        correctly compute dual solution components when the problem is
322        scaled.
323
324        The files configure.ac and Makefile.am were changed:
325        (a) to allow using autoreconf/autoheader;
326        (b) to allow building the package in a directory other than its
327            source directory.
328        Thanks to Marco Atzeri <marco_atzeri@yahoo.it> for bug report.
329
330        An example model in the GNU MathProg language was added.
331        Thanks to Larry D'Agostino <Larry.D'Agostino@gmacrescap.com> for
332        contribution.
333
334GLPK 4.36 (release date: Feb 06, 2009)
335
336        The following new API routines were added to the package:
337
338        glp_mincost_okalg     find minimum-cost flow with out-of-kilter
339                              algorithm
340        glp_maxflow_ffalg     find maximal flow with Ford-Fulkerson
341                              algorithm
342
343        For detailed description of these new routines and related data
344        structures see chapter "Graph and Network API Routines" in a new
345        edition of the reference manual included in the distribution.
346
347        The following two new command-line options were added to the
348        solver glpsol:
349
350        --mincost             read min-cost flow data in DIMACS format
351        --maxflow             read maximum flow data in DIMACS format
352
353        Duplicate symbols in the header glpk.h were removed to allow
354        using swig.
355        Thanks to Kelly Westbrooks <kellywestbrooks@yahoo.com> and
356        Nigel Galloway <nigel_galloway@operamail.com> for suggestion.
357
358        A minor defect was fixed in the routine glp_write_lp.
359        Thanks to Sebastien Briais <sbriais@free.fr> for bug report.
360
361        A minor bug was fixed in the SQL module.
362        Thanks to Xypron <xypron.glpk@gmx.de> for patch.
363
364        Some new example models in the GNU MathProg modeling language
365        were added. Thanks to Sebastian Nowozin <nowozin@gmail.com> and
366        Nigel Galloway <nigel_galloway@operamail.com> for contribution.
367
368GLPK 4.35 (release date: Jan 09, 2009)
369
370        The following new API routines were added to the package:
371
372        glp_create_graph      create graph
373        glp_set_graph_name    assign (change) graph name
374        glp_add_vertices      add new vertices to graph
375        glp_add_arc           add new arc to graph
376        glp_erase_graph       erase graph content
377        glp_delete_graph      delete graph
378        glp_read_mincost      read minimum cost flow problem data in
379                              DIMACS format
380        glp_write_mincost     write minimum cost flow problem data in
381                              DIMACS format
382        glp_mincost_lp        convert minimum cost flow problem to LP
383        glp_netgen            Klingman's network problem generator
384        glp_gridgen           grid-like network problem generator
385        glp_read_maxflow      read maximum flow problem data in DIMACS
386                              format
387        glp_write_maxflow     write maximum flow problem data in DIMACS
388                              format
389        glp_maxflow_lp        convert maximum flow problem to LP
390        glp_rmfgen            Goldfarb's maximum flow problem generator
391
392        For detailed description of these new routines and related data
393        structures see chapter "Graph and Network API Routines" in a new
394        edition of the reference manual included in the distribution.
395
396        A minor change were made in the internal routine xputc. Thanks
397        to Luiz Bettoni <bettoni@cpgei.ct.utfpr.edu.br> for suggestion.
398
399        A minor bug was fixed in the internal routine mpl_fn_time2str.
400        Thanks to Stefan Vigerske <stefan@vigerske.de> for bug report.
401
402GLPK 4.34 (release date: Dec 04, 2008)
403
404        The GNU MathProg modeling language was supplemented with three
405        new built-in functions:
406
407        gmtime    obtaining current calendar time
408        str2time  converting character string to calendar time
409        time2str  converting calendar time to character string
410
411        (Thanks to Xypron <xypron.glpk@gmx.de>.)
412
413        For detailed description of these functions see Appendix A in
414        the document "Modeling Language GNU MathProg", a new edition of
415        which was included in the distribution.
416
417        A bug was fixed in the MIP solver. Thanks to Nigel Galloway
418        <nigel_galloway@operamail.com> for bug report.
419
420        A new makefile was added to build the GLPK DLL with Microsoft
421        Visual Studio Express 2008 for 64-bit Windows. Thanks to Xypron
422        <xypron.glpk@gmx.de> for contribution and testing.
423
424GLPK 4.33 (release date: Oct 30, 2008)
425
426        The following new API routines were added to the package:
427        glp_copy_prob         copy problem object content
428        glp_exact             solve LP in exact arithmetic
429                              (makes lpx_exact deprecated)
430        glp_get_unbnd_ray     determine variable causing unboundedness
431                              (makes lpx_get_ray_info deprecated)
432        glp_interior          solve LP with interior-point method
433                              (makes lpx_interior deprecated)
434
435        The following new API routines for processing models written in
436        the GNU Mathprog language were added to the package:
437        glp_mpl_alloc_wksp    allocate the translator workspace
438        glp_mpl_read_model    read and translate model section
439        glp_mpl_read_data     read and translate data section
440        glp_mpl_generate      generate the model
441        glp_mpl_build_prob    build LP/MIP instance from the model
442        glp_mpl_postsolve     postsolve the model
443        glp_mpl_free_wksp     deallocate the translator workspace
444        (These routines make lpx_read_model deprecated.)
445
446        For description of all these new API routines see the reference
447        manual included in the distribution.
448
449        A crude implementation of CPLEX-like interface to GLPK API was
450        added to the package. Currently it allows using GLPK as a core
451        LP solver for Concorde, a well known computer code for solving
452        the symmetric TSP. For details see examples/cplex/README.
453
454        Some bugs were fixed in the SQL table driver. Thanks to Xypron
455        <xypron.glpk@gmx.de>.
456
457GLPK 4.32 (release date: Oct 03, 2008)
458
459        The following new features were included in the MIP solver
460        (the API routine glp_intopt):
461
462        *  MIP presolver
463        *  mixed cover cut generator
464        *  clique cut generator
465        *  Euclidean reduction of the objective value
466
467        Due to changes the routine glp_intopt may additionally return
468        GLP_ENOPFS, GLP_ENODFS, and GLP_EMIPGAP.
469
470        The API routines lpx_integer are lpx_intopt are deprecated,
471        since they are completely superseded by glp_intopt.
472
473        The following new branch-and-cut API routines were added:
474        glp_ios_row_attr      determine additional row attributes
475        glp_ios_pool_size     determine current size of the cut pool
476        glp_ios_add_row       add constraint to the cut pool
477        glp_ios_del_row       delete constraint from the cut pool
478        glp_ios_clear_pool    delete all constraints from the cut pool
479
480        For description of these new routines see the reference manual
481        included in the distribution.
482
483        The stand-alone solver glpsol was changed to allow multiple
484        data files.
485
486        A new edition of the supplement "Using Data Tables in the GNU
487        MathProg Modeling Language" was included.
488
489        As usual, some bugs were fixed (in the MathProg translator).
490        Thanks to Xypron <xypron.glpk@gmx.de>.
491
492GLPK 4.31 (release date: Sep 02, 2008)
493
494        The core LP solver based on the dual simplex method was
495        re-implemented and now it provides both phases I and II.
496
497        The following new API routines were added:
498        glp_scale_prob  automatic scaling of problem data
499        glp_std_basis   construct standard initial LP basis
500        glp_adv_basis   construct advanced initial LP basis
501        glp_cpx_basis   construct Bixby's initial LP basis
502
503        For description of these new routines see the reference manual
504        included in the distribution.
505
506        The following API routines are deprecated:
507        lpx_scale_prob, lpx_std_basis, lpx_adv_basis, lpx_cpx_basis.
508
509        Necessary changes were made in memory allocation routines to
510        resolve portability issues for 64-bit platforms.
511
512        New version of the routine lpx_write_pb to write problem data
513        in OPB (pseudo boolean format) was added to the package. Thanks
514        to Oscar Gustafsson <oscarg@isy.liu.se> for the contribution.
515
516        Two new makefiles were added to build the package for 32- and
517        64-bit Windows with Microsoft Visual Studio Express 2008.
518        Thanks to Heinrich Schuchardt <heinrich.schuchardt@gmx.de> (aka
519        Xypron) for the contribution and testing.
520
521        Two new makefiles were added to build the package with Digital
522        Mars C/C++ 8.50 and Open Watcom C/C++ 1.6 (for 32-bit Windows).
523
524GLPK 4.30 (release date: Aug 13, 2008)
525
526        The core LP solver based on the primal simplex method was
527        re-implemented to allow its further improvements. Currently the
528        new version provides the same features as the old one, however,
529        it is a bit faster and more numerically stable.
530
531        Some changes were made in the MathProg translator to allow <,
532        <=, >=, and > on comparing symbolic values. Thanks to Heinrich
533        Schuchardt <heinrich.schuchardt@gmx.de> for patches.
534
535        Internal routine set_d_eps in the exact LP solver was changed
536        to prevent approximation errors in case of integral data.
537        Thanks to Markus Pilz <pilz@cs.uni-bonn.de> for bug report.
538
539GLPK 4.29 (release date: Jul 06, 2008)
540
541        The configure script was changed to disable all optional
542        features by default. For details please see file INSTALL.
543
544        The following new API routines were added:
545        glp_erase_prob  erase problem object content
546        glp_read_mps    read problem data in MPS format
547        glp_write_mps   write problem data in MPS format
548        glp_read_lp     read problem data in CPLEX LP format
549        glp_write_lp    write problem data in CPLEX LP format
550
551        For description of these new routines see the reference manual
552        included in the distribution.
553
554        The following API routines are deprecated:
555        lpx_read_mps, lpx_read_freemps, lpx_write_mps,
556        lpx_write_freemps, lpx_read_cpxlp, and lpx_write_cpxlp.
557
558        Two bugs were fixed. Thanks to
559        Anne-Laurence Putz <anne-laurence.putz@eurodecision.com> and
560        Xypron <xypron.glpk@gmx.de> for bug report.
561
562GLPK 4.28 (release date: Mar 25, 2008)
563
564        The iODBC and MySQL table drivers, which allows transmitting
565        data between MathProg model objects and relational databases,
566        were re-implemented to replace a static linking by a dynamic
567        linking to corresponding shared libraries.
568        Many thanks to Heinrich Schuchardt <heinrich.schuchardt@gmx.de>
569        for the contribution, Rafael Laboissiere <rafael@debian.org>
570        for useful advices concerning the shared library support under
571        GNU/Linux, and Vijay Patil <vijay.patil@gmail.com> for testing
572        this feature under Windows XP.
573
574        A new optional feature was added to the package. This feature
575        is based on the zlib data compression library and allows GLPK
576        API routines and the stand-alone solver to read and write
577        compressed data files performing compression/decompression "on
578        the fly" (compressed data files are recognized by suffix `.gz'
579        in the file name). It may be useful in case of large MPS files
580        to save the disk space (up to ten times).
581
582        The `configure' script was re-implemented. Now it supports the
583        following specific options:
584
585        --with-gmp           Enable using the GNU MP bignum library
586        --without-gmp        Disable using the GNU MP bignum library
587        --with-zlib          Enable using the zlib data compression
588                             library
589        --without-zlib       Disable using the zlib data compression
590                             library
591        --enable-dl          Enable shared library support (auto check)
592        --enable-dl=ltdl     Enable shared library support (GNU)
593        --enable-dl=dlfcn    Enable shared library support (POSIX)
594        --disable-dl         Disable shared library support
595        --enable-odbc        Enable using ODBC table driver
596        --disable-odbc       Disable using ODBC table driver
597        --enable-mysql       Enable using MySQL table driver
598        --disable-mysql      Disable using MySQL table driver
599
600        For more details please see file INSTALL.
601
602GLPK 4.27 (release date: Mar 02, 2008)
603
604        Three new table drivers were added to the MathProg translator:
605
606        xBASE built-in table driver, which allows reading and writing
607        data in .dbf format (only C and N fields are supported);
608
609        MySQL table driver, which provides connection to a MySQL
610        database;
611
612        iODBC table driver, which provides connection to a database
613        through ODBC.
614
615        The MySQL and iODBC table drivers were contributed to GLPK by
616        Heinrich Schuchardt <heinrich.schuchardt@gmx.de>.
617
618        The table driver is a program module which allows transmitting
619        data between MathProg model objects and external data tables.
620
621        For detailed description of the table statement and table
622        drivers see the document "Using Data Tables in the GNU MathProg
623        Modeling Language" (file doc/tables.txt) included in the
624        distribution. Some examples which demonstrate using MySQL and
625        iODBC table drivers can be found in subdirectory examples/sql.
626
627GLPK 4.26 (release date: Feb 17, 2008)
628
629        The table statement was implemented in the GNU MathProg
630        modeling language. This new feature allows reading data from
631        external tables into model objects such as sets and parameters
632        as well as writing results of computations to external tables.
633
634        A table is a (unordered) set of records, where each record
635        consists of the same number of fields, and each field is
636        provided with a unique symbolic name called the field name.
637
638        Currently the GLPK package has the only built-in table driver,
639        which supports tables in the CSV (comma-separated values) file
640        format. This format is very simple and supported by almost all
641        spreadsheets and database management systems.
642
643        Detailed description of the table statement and CSV format can
644        be found in file doc/tables.txt, included in the distribution.
645
646GLPK 4.25 (release date: Dec 19, 2007)
647
648        A tentative implementation of Gomory's mixed integer cuts was
649        included in the branch-and-cut solver. To enable generating
650        Gomory's cuts the control parameter gmi_cuts passed to the
651        routine glp_intopt should be set to GLP_ON. This feature is
652        also available in the solver glpsol through command-line option
653        '--gomory'. For more details please see the reference manual
654        included in the distribution.
655
656GLPK 4.24 (release date: Nov 21, 2007)
657
658        A tentative implementation of MIR (mixed integer rounding) cuts
659        was included in the MIP solver. To enable generating MIR cuts
660        the control parameter mir_cuts passed to the routine glp_intopt
661        should be set to GLP_ON. This feature is also available in the
662        stand-alone solver glpsol via command-line option '--mir'. For
663        more details please see the reference manual included in the
664        distribution.
665
666        The implementation is mainly based on the following two papers:
667
668        1. H. Marchand and L. A. Wolsey. Aggregation and mixed integer
669           rounding to solve MIPs. CORE discussion paper 9839, CORE,
670           Universite catholique de Louvain, June 1998.
671
672        2. G. Andreello, A. Caprara, and M. Fischetti. Embedding cuts
673           in a Branch&Cut framework. Preliminary draft, October 2003.
674
675        MIR cuts can be generated on any level of the search tree that
676        makes the GLPK MIP solver to be a real branch-and-cut solver.
677
678        A bug was fixed in the routine lpx_write_cpxlp. If a variable
679        x has upper bound and no lower bound, it should appear in the
680        bounds section as "-inf <= x <= u", not as "x <= u". Thanks to
681        Enric Rodriguez <erodri@lsi.upc.edu> for the bug report.
682
683GLPK 4.23 (release date: Oct 28, 2007)
684
685        The following new API routines were added:
686
687        glp_read_sol    read basic solution from text file
688        glp_write_sol   write basic solution to text file
689        glp_read_ipt    read interior-point solution from text file
690        glp_write_ipt   write interior-point solution to text file
691        glp_read_mip    read MIP solution from text file
692        glp_write_mip   write MIP solution to text file
693
694        For description of these routines and corresponding file
695        formats see Chapter "API Routines", Section "Utility routines"
696        in the reference manual included in the distribution.
697
698        Advanced API routine glp_free_env was added. It may be used by
699        the application program to free all resources allocated by GLPK
700        routines.
701
702        The following three new command-line options were added to the
703        solver glpsol:
704
705        --mipgap tol    set relative MIP gap tolerance
706        -r filename     read solution from filename
707        -w filename     write solution to filename
708
709GLPK 4.22 (release date: Sep 19, 2007)
710
711        This is a maintainer release.
712
713        A bug was fixed in the MIP preprocessor (ios_preprocess_node).
714        Thanks to Roberto Bagnara <bagnara@cs.unipr.it> (Department of
715        Mathematics, University of Parma, Italy) for the bug report.
716
717        A bug was fixed in the MIP preprocessor (col_implied_bounds),
718        due to which constraint coefficients with small magnitude could
719        lead to wrong implied bounds of structural variables.
720
721        A similar bug was fixed in the routine reduce_bounds.
722
723        A bug was fixed in the routines glp_set_mat_row and
724        glp_set_mat_col. (The bug appeared due to incorrect removing
725        zero elements from the row/column lists.)
726
727        A bug was fixed in the API routines lpx_read_mps and
728        lpx_read_freemps, due to which bounds of type LI specified in
729        BOUNDS section were incorrectly processed.
730
731        A call to standard function vsprintf was replaced by a call to
732        vsnprintf for security reasons. Many thanks to Peter T. Breuer
733        <ptb@inv.it.uc3m.es> and Rafael Laboissiere <rafael@debian.org>.
734
735GLPK 4.21 (release date: Aug 28, 2007)
736
737        Additional reasons for calling the callback routine used in the
738        MIP solver (glp_intopt) were introduced. Currently the following
739        reasons are supported:
740
741        * request for subproblem selection
742        * request for preprocessing
743        * request for row generation
744        * request for heuristic solution
745        * request for cut generation
746        * request for branching
747        * better integer solution found
748
749        A basic preprocessing component used to improve subproblem
750        formulations by tightening bounds of variables was included in
751        the MIP solver. Depending on the control parameter pp_tech
752        passed to the routine glp_intopt the preprocessing can be
753        performed either on the root level or on all levels (default)
754        or can be disabled.
755
756        Backtracking heuristic used by default in the MIP solver was
757        changed to the "best local bound".
758
759        For more details see Chapter "Advanced API routines", Section
760        "Branch-and-bound interface routines" in a new edition of the
761        reference manual included in the distribution.
762
763GLPK 4.20 (release date: Jul 26, 2007)
764
765        API routine lpx_integer was replaced by API routine glp_intopt,
766        which provides equivalent functionality and additionally allows
767        the application to control the solution process by means of the
768        user-written callback routine, which is called by the solver at
769        various points of the branch-and-bound algorithm. Besides, the
770        new MIP solver allows generating "lazy" constraints and cutting
771        planes on all levels of the branch-and-bound tree, not only on
772        the root level. The routine lpx_integer is also still available
773        for the backward compatibility.
774
775        The following new advanced API routines, which may be called
776        from the B&B callback routine, were included in the package:
777
778        glp_ios_reason     determine reason for calling callback
779                           routine
780        glp_ios_get_prob   access the problem object
781        glp_ios_tree_size  determine size of the branch-and-bound tree
782        glp_ios_curr_node  determine current active subproblem
783        glp_ios_next_node  determine next active subproblem
784        glp_ios_prev_node  determine previous active subproblem
785        glp_ios_up_node    determine parent subproblem
786        glp_ios_node_level determine subproblem level
787        glp_ios_node_bound determine subproblem local bound
788        glp_ios_mip_gap    compute relative MIP gap
789        glp_ios_heur_sol   provide solution found by heuristic
790        glp_ios_terminate  terminate the solution process
791
792        For description of these routines see Chapter "Advanced API
793        routines", Section "Branch-and-bound interface routines" in a
794        new edition of the reference manual, which was included in the
795        distribution.
796
797        Old version of the integer optimization suite (IOS) as well as
798        TSP solver tspsol based on it are no longer supported and were
799        removed from the package.
800
801        A minor error in the MIP presolver was fixed; thanks to Graham
802        Rockwell <bionomicron@gmail.com> for the bug report.
803
804GLPK 4.19 (release date: Jul 05, 2007)
805
806        The principal change is upgrading to GPLv3.
807
808        A serious bug in the routine glp_del_cols was fixed; thanks to
809        Cedric[FR] <fox2113@wanadoo.fr> for the bug report. The bug
810        appeared because on deleting non-basic columns the basis header
811        remained valid, however, contained invalid (old) column ordinal
812        numbers.
813
814        A new advanced API routine glp_mem_limit was added.
815
816        The case GLP_EBOUND was added to the routine lpx_simplex.
817        Thanks to Cameron Kellough <Cameron.Kellough@sri.com> for the
818        bug report.
819
820        An API routine lpx_write_pb to write the problem instance in
821        OPB (pseudo boolean) format format was added. Thanks to Oscar
822        Gustafsson <oscarg@isy.liu.se> for the contribution.
823
824        Two new options --wpb and --wnpb were added to glpsol to write
825        the problem instance in OPB format.
826
827GLPK 4.18 (release date: Jun 25, 2007)
828
829        The following new API routines were added:
830
831        glp_set_rii        set (change) row scale factor
832        glp_set_sjj        set (change) column scale factor
833        glp_get_rii        retrieve row scale factor
834        glp_get_sjj        retrieve column scale factor
835        glp_simplex        solve LP problem with the simplex method
836                           (this routine replaces lpx_simplex, which is
837                           also available for backward compatibility)
838        glp_init_smcp      initialize simplex method control params
839        glp_bf_exists      check if the basis factorization exists
840        glp_factorize      compute the basis factorization
841        glp_bf_updated     check if the basis factorization has been
842                           updated
843        glp_get_bfcp       retrieve basis factorization control params
844        glp_set_bfcp       change basis factorization control params
845        glp_get_bhead      retrieve the basis header information
846        glp_get_row_bind   retrieve row index in the basis header
847        glp_get_col_bind   retrieve column index in the basis header
848        glp_ftran          perform forward transformation
849        glp_btran          perform backward transformation
850
851        For description of all these routines see a new edition of the
852        reference manual included in the distribution.
853
854        Type names ulong_t and uldiv_t were changed to glp_ulong and
855        glp_uldiv to avoid conflicts with standard type names on some
856        platforms. Thanks to Boris Wirtz <Boris.Wirtz@uni-oldenburg.de>
857        for the bug report.
858
859        Some new examples in the MathProg language were added. Thanks
860        to Sebastian Nowozin <nowozin@gmail.com>.
861
862GLPK 4.17 (release date: May 26, 2007)
863
864        API routines glp_set_mat_row, glp_set_mat_col, and glp_load_mat
865        were modified to allow zero constraint coefficients (which are
866        not stored in the constraint matrix). Note that constraint
867        coefficients with duplicate row/column indices are not allowed.
868
869        Another form of LP basis factorization was implemented in the
870        package. It is based on LU-factorization of an initial basis
871        and Schur complement to reflect changes in the basis. Currently
872        the implementation is incomplete and provides only updating the
873        factorization on replacing a column of the basis matrix. On API
874        level the user can set the control parameter LPX_K_BFTYPE to
875        choose between the folloiwng forms of LP basis factorization to
876        be used in the simplex method routines:
877        1) LU + Forrest-Tomlin update;
878        2) LU + Schur complement + Bartels-Golub update;
879        3) LU + Schur complement + Givens rotation update.
880        The GLPK implementation is similar to LUSOL/LUMOD developed by
881        Michael A. Saunders.
882
883        The user can choose the form of LP basis factorzation used by
884        the simplex method routines by specifying the folloiwng options
885        of glpsol: --luf, --cbg, --cgr.
886
887GLPK 4.16 (release date: May 05, 2007)
888
889        A number of basic GLPK API routines, which now are in the
890        stable stable, were renamed to be prefixed with 'glp_'. Note
891        that all these routines are available via their old names
892        prefixed with 'lpx_' that keeps the downward compatibility with
893        older versions of the package.
894
895        Three new GLPK API routines were added to the package:
896        glp_version, glp_term_hook, and glp_mem_usage; for more details
897        see a new edition of the GLPK reference manual included in the
898        distribution. The routine glp_version reports the actual version
899        of the GLPK library and also can be used (along with the header
900        glpk.h) in Autotools specification files to check if the GLPK
901        library has been installed.
902
903        The header glpk.h was changed to conform to C++ environment.
904
905GLPK 4.15 (release date: Feb 18, 2007)
906
907        Autotools specification files (configure.ac, Makefile.am) were
908        changed to use GNU Libtool. This allows building the static as
909        well as shared GLPK library.
910
911GLPK 4.14 (release date: Feb 05, 2007)
912
913        Now GLPK conforms to ILP32, LLP64, and LP64 programming models
914        (the latter seems to be the ultimate choice regarding 64-bit
915        architectures). Note that GLPK itself is a 32-bit application,
916        and the conformity only means that the package works correctly
917        on all these arenae. Nevertheless, on 64-bit platforms it is
918        possible to use more than 4GB of memory, if necessary.
919
920GLPK 4.13 (release date: Nov 13, 2006)
921
922        A tentative implementation of the "exact" simplex method based
923        on bignum (rational) arithmetic was included in the package.
924
925        On API level this new feature is available through the routine
926        lpx_exact, which is similar to the routine lpx_simplex.
927
928        In the solver glpsol this feature is available through two new
929        command-line options: --exact and --xcheck. If the '--exact'
930        option is specified, glpsol solves LP instance using the exact
931        simplex method; in case of MIP it is used to obtain optimal
932        solution of LP relaxation. If the --xcheck option is specified,
933        LP instance (or LP relaxation) is solved using the standard
934        (floating-point) simplex method, however, then glpsol calls the
935        exact simplex routine to make sure that the final LP basis is
936        exactly optimal, and if it is not, to perform some additional
937        simplex iterations in exact arithmetic.
938
939GLPK 4.12 (release date: Nov 08, 2006)
940
941        A tentative implementation of some simplex method routines
942        based on exact (bignum) arithmetic was included in the package.
943        Currently these routines provide computing LU-factorization of
944        the basis matrix and computing components of basic solution.
945
946        These routines were used to implement a routine, which checks
947        primal and dual feasibility of basic solution exactly, i.e. in
948        rational numbers, without round-off errors. In glpsol this
949        feature is available through the command-line option --xcheck.
950
951        GLPK has its own low-level routines implementing operations on
952        integer and rational numbers that makes it independent on other
953        software packages. However, to attain a much better performance
954        it is highly recommended to install (before configuring GLPK)
955        the GNU Multiple Precision Arithmetic Library (GMP). Using GMP
956        makes computations 100-200 times faster.
957
958GLPK 4.11 (release date: Jul 25, 2006)
959
960        Three new built-in functions in the modeling language were
961        implemented: card (cardinality of set), length (length of
962        character string), and substr (substring of character string).
963        Another improvement concerns the printf statement which now
964        allows redirecting its output to a specified file. These new
965        features are illustrated in example models crypto.mod and
966        graph.mod included in the distribution. For more details see
967        the document "Modeling Language GNU MathProg".
968
969        Four batch files (along with corresponding makefiles) were
970        included in the distribution to simplify building GLPK under
971        MS Windows; see them in subdirectory 'w32'.
972
973GLPK 4.10 (release date: May 11, 2006)
974
975        Cutting planes of two new classes were implemented: mixed cover
976        cuts and clique cuts. On API level this feature can be enabled
977        by setting control parameter LPX_K_USECUTS passed to the routine
978        lpx_intopt. In glpsol this feature is available through the
979        command-line options --cover and --clique. For more details see
980        the reference manual.
981
982        Now the routines lpx_read_mps and lpx_read_freemps support LI
983        bound type. It is similar to LO, however, indicates the column
984        as of integer kind.
985
986GLPK 4.9 (release date: Jan 17, 2006)
987
988        An advanced MIP solver was implemented. It includes:
989
990        - basic presolving technique (removing free, singleton and
991          redundant rows, improving bounds of columns, removing fixed
992          columns, reducing constraint coefficents);
993
994        - generating cutting planes to improve LP relaxation (currently
995          only Gomory's mixed integer cuts are implemented);
996
997        - using the branch-and-bound method to solve resultant MIP;
998
999        - recovering solution of the original MIP.
1000
1001        The solver is available on API level via the routine lpx_intopt
1002        (see the reference manual). It is similar to the routine
1003        lpx_integer, however, does not require initial solution of LP
1004        relaxation.
1005
1006        The solver is also available in the command-line utility glpsol
1007        via two options: --intopt (only presolving) and --cuts (assumes
1008        --intopt plus generating cuts).
1009
1010        Note that efficiency of the MIP solver strongly depends on the
1011        internal structure of the problem to be solved. For some hard
1012        instances it is very efficient, but for other instances it may
1013        be significantly worse than the standard branch-and-bound.
1014
1015        For some comparative benchmarks see doc/bench1.txt.
1016
1017        Well, what else...
1018
1019        Three built-in functions were added to MathProg: sin, cos, and
1020        atan (the latter allows one or two arguments).
1021
1022        Some bugs were fixed.
1023
1024        Several new examples in MathProg were included: color.mod
1025        (graph coloring problem), tsp.mod (traveling salesman problem),
1026        and pbn.mod (paint-by-numbers puzzle).
1027
1028GLPK 4.8 (release date: Jan 12, 2005)
1029
1030        Core simplex method and interior-point method routines were
1031        re-implemented and now they use a new, "storage-by-rows" sparse
1032        matrix format (unlike previous versions where linked lists were
1033        used to represent sparse matrices). For details see ChangeLog.
1034
1035        Also a minor bug was fixed in API routine lpx_read_cpxlp.
1036
1037GLPK 4.7 (release date: Aug 23, 2004)
1038
1039        Now GLPK supports free MPS format. Two new API routines
1040        lpx_read_freemps (to read problem data in free MPS format) and
1041        lpx_write_freemps (to write problem data in free MPS format)
1042        were added. This feature is also available in the solver glpsol
1043        via new command-line options --freemps and --wfreemps. For more
1044        details see the GLPK reference manual.
1045
1046        API routines lpx_read_cpxlp and lpx_write_cpxlp for reading and
1047        writing problem data in CPLEX LP format were re-implemented to
1048        allow long symbolic names (up to 255 characters).
1049
1050        The following three modules were temporarily removed from the
1051        GLPK distribution due to licensing problems: DELI (an interface
1052        module to Delphi), GLPKMEX (an interface module to Matlab), and
1053        JNI (an interface module to Java).
1054
1055GLPK 4.6 (release date: Aug 04, 2004)
1056
1057        Three new statements were implemented in the GNU MathProg
1058        language: solve, printf, and for. Their detailed description can
1059        be found in the GLPK documentation included in the distribution.
1060        (See also a sample model, examples/queens.mod, which illustrates
1061        using these new statements.)
1062
1063        Two new API routines were added to the package: lpx_read_prob
1064        and lpx_write_prob. They allow reading/writing problem data in
1065        GNU LP low-level text format.
1066
1067        Three new command-line options were implemented in the LP/MIP
1068        solver glpsol: --glp (to read problem data in GNU LP format),
1069        --wglp (to write problem data in GNU LP format), and --name (to
1070        change problem name). Now glpsol also supports processing models
1071        where the new statements (see above) are used.
1072
1073        A new version of GLPKMEX, a Matlab MEX interface to GLPK, was
1074        included. For more details see contrib/glpkmex/ChangeLog.
1075
1076GLPK 4.5 (release date: Jul 19, 2004)
1077
1078        The branch-and-bound solver was completely re-implemented.
1079
1080        Some modifications were made in memory allocation routines that
1081        allows using the package on 64-bit platforms.
1082
1083        For more details see ChangeLog.
1084
1085GLPK 4.4 (release date: Jan 17, 2004)
1086
1087        All API routines were re-implemented using new data structures.
1088        The new implementation provides the same specifications and
1089        functionality of API routines as the old one, however, it has
1090        some important advantages, in particular:
1091        * linked lists are used everywhere that allows creating and
1092          modifying the problem object as efficiently as possible
1093        * all data stored in the problem object are non-scaled (even if
1094          the internal scaling is used) that prevents distortion of the
1095          original problem data
1096        * solution components obtained by the solver remain available
1097          even if the problem object has been modified
1098        * no solver-specific data are used in the new data structures
1099          that allows attaching any external lp/mip solver using GLPK
1100          API as an uniform interface
1101        Note that some API routines became obsolete being replaced by
1102        new, more convenient routines. These obsolete routines are kept
1103        for backward compatibility, however, they will be removed in
1104        the future. For more details please see ChangeLog and the GLPK
1105        Reference Manual.
1106
1107        New edition of the GLPK Reference Manual was included in the
1108        distribution.
1109
1110        GLPKMEX, a Matlab MEX interface to GLPK package, contributed by
1111        Nicolo Giorgetti <giorgetti@dii.unisi.it> was included in the
1112        distribution.
1113
1114        GLPK FAQ contributed by Harley Mackenzie <hjm@bigpond.com> was
1115        included in the distribution.
1116
1117GLPK 4.3 (release date: Dec 12, 2003)
1118
1119        The bug, due to which the standard math library is not linked
1120        on building the package on some platforms, was fixed.
1121
1122        The following new built-in functions were added to the MathProg
1123        language: round, trunc, Irand224, Uniform01, Uniform, Normal01,
1124        Normal. For details see the language description.
1125
1126        The MathProg syntax was changed to allow writing 'subj to' that
1127        means 'subject to'.
1128
1129        The new api routine lpx_get_ray_info was added. It is intended
1130        to determine which (non-basic) variable causes unboundness. For
1131        details see the reference manual.
1132
1133        The module glpmps.c was changed to avoid compilation errors on
1134        building the package on Mac OS X.
1135
1136        Several typos was fixed and some new material was added to the
1137        GLPK documentation.
1138
1139GLPK 4.2 (release date: Nov 14, 2003)
1140
1141        A preliminary implementation of the Integer Optimization Suite
1142        (IOS) was included in the package. The Branch-and-Cut Framework
1143        being completely superseded by IOS was removed from the package.
1144
1145        New API routine lpx_print_sens_bnds intended for bounds
1146        sensitivity analysis was contributed to GLPK by Brady Hunsaker
1147        <hunsaker@engr.pitt.edu>. This function is also available in
1148        the solver glpsol (via command-line option --bounds).
1149
1150        An improved version of GLPK JNI (Java Native Interface) was
1151        contributed by Chris Rosebrugh <cpr@pobox.com>.
1152
1153        GLPK DELI (Delphi Interface) was contributed by Ivo van Baren
1154        <i.van.baren@freeler.nl>.
1155
1156        Several makefiles were added to allow compiling GLPK on some
1157        non-GNU 32-bit platforms:
1158        * Windows single-threaded static library, Visual C++ 6.0
1159        * Windows multi-threaded dynamic library, Visual C++ 6.0
1160        * Windows single-threaded static library, Borland C++ 5.2
1161        * DOS single-threaded static library, Digital Mars C++ 7.50
1162
1163        And, of course, some bugs were fixed.
1164
1165        For more details see ChangeLog.
1166
1167GLPK 4.1 (release date: Aug 23, 2003)
1168
1169        Some improvements were made in the lp/mip solver routines and
1170        several bugs were fixed in the model translator.
1171
1172        For more details see ChangeLog.
1173
1174GLPK 4.0 (release date: May 06, 2003)
1175
1176        Now GLPK supports the GNU MathProg modeling language, which is
1177        a subset of the AMPL modeling language.
1178
1179        The document "GLPK: Modeling Language GNU MathProg" included in
1180        the distribution is a complete description of GNU MathProg. (See
1181        the files lang.latex, lang.dvi, and lang.ps in the subdirectory
1182        'doc'. See also some examples in the subdirectory 'sample'.)
1183
1184        New version of the solver glpsol, which supports models written
1185        in GNU MathProg, was implemented. (Brief instructions how to use
1186        glpsol can be found in the GNU MathProg documentation.)
1187
1188        The GLPK/L modeling language is no more supported. The reason is
1189        that GNU MathProg being much more powerful completely supersedes
1190        all features of GLPK/L.
1191
1192GLPK 3.3 (release date: Mar 25, 2003)
1193
1194        LP PRESOLVER
1195        ------------
1196
1197        Now the routine lpx_simplex (which is a driver to the simplex
1198        method for solving LP) is provided with the built-in LP
1199        presolver, which is a program that transforms the original LP
1200        problem to an equivalent LP problem, which may be easier for
1201        solving with the simplex method than the original one. Once the
1202        transformed LP has been solver, the presolver transforms its
1203        basic solution back to a corresponding basic solution of the
1204        original problem. For details about this feature please see the
1205        GLPK reference manual.
1206
1207        Currently the LP presolver implements the following features:
1208        * removing empty rows;
1209        * removing empty columns;
1210        * removing free rows;
1211        * removing fixed columns;
1212        * removing row singletons, which have the form of equations;
1213        * removing row singletons, which have the form of inequalities;
1214        * removing column singletons, which are implied slack variables;
1215        * fixing and removing column singletons, which are implied free
1216          variables;
1217        * removing forcing rows that involves fixing and removing the
1218          corresponding columns;
1219        * checking for primal and dual infeasibilities.
1220
1221        The LP presolver is also used by default in the stand-alone
1222        program glpsol. In order *not* to use it, the option --nopresol
1223        should be specified in the command-line.
1224
1225        CHANGES IN GLPK/L
1226        -----------------
1227
1228        The syntax and semantics of the GLPK/L modeling language was
1229        changed to allow declaration of "interval" sets. This means that
1230        now the user can declare a set, for example, as:
1231
1232           set task = [8:11];
1233
1234        that is exactly equivalent to the following declaration:
1235
1236           set task = (task_8, task_9, task_10, task_11);
1237
1238        For details see the language description.
1239
1240        JAVA INTERFACE
1241        --------------
1242
1243        Now GLPK includes the package GLPK JNI (Java Native Interface)
1244        that implements Java binding for GLPK. It allows Java programs
1245        to utilize GLPK in solving LP and MIP problems. For details see
1246        a brief user's guide in the subdirectory contrib/java-binding.
1247        This package was developed and programmed by Yuri Victorovich
1248        <yuri@gjt.org>, who contributed it to GLPK.
1249
1250GLPK 3.2.4 (release date: Feb 18, 2003)
1251
1252        This is a bug-fix release. For details see ChangeLog.
1253
1254GLPK 3.2.3 (release date: Nov 11, 2002)
1255
1256        A new implementation of the api routine lpx_integer which now
1257        is based on the b&b driver (which is based on the implicit
1258        enumeration suite) was included in the package. This new
1259        implementation has exactly the same functionality as the old
1260        version, so all changes are transparent to the api user.
1261
1262        Four new api routines were included in the package:
1263        lpx_check_kkt checks Karush-Kuhn-Tucker optmality conditions;
1264        lpx_read_bas reads predifined basis in MPS format;
1265        lpx_write_bas writes current basis in MPS format;
1266        lpx_write_lpt writes problem data in CPLEX LP format.
1267
1268        Also other minor improvements were made (for details see the
1269        file 'ChangeLog').
1270
1271GLPK 3.2.2 (release date: Oct 14, 2002)
1272
1273        The api routine lpx_read_lpt was included in the package. It
1274        is similar to the routine lpx_read_mps and intended to read
1275        LP/MIP data prepared in CPLEX LP format. Description of this
1276        format is given in the GLPK reference manual, a new edition of
1277        which was also included in the distribution (see the files
1278        'refman.latex', 'refman.dvi', 'refman.ps' in the subdirectory
1279        'doc'). In order to use data files in CPLEX LP format with the
1280        solver glpsol the option '--lpt' should be specified in the
1281        command line.
1282
1283        Several bugs were fixed and some minor improvements were made
1284        (for details see the file 'ChangeLog').
1285
1286GLPK 3.2.1 (release date: Aug 12, 2002)
1287
1288        Now GLPK includes a preliminary implementation of the
1289        branch-and-cut framework, which is a set of data structures and
1290        routines intended for developing branch-and-cut methods for
1291        solving mixed-integer and combinatorial optimization problems.
1292
1293        Detailed decsription of the branch-and-cut framework is given in
1294        the document "GLPK: A Preliminary Implementation of the
1295        Branch-And-Cut Framework" included in the distribution (see the
1296        file 'brcut.txt' in the subdirectory 'doc').
1297
1298        In order to illustrate how the GLPK branch-and-cut framework
1299        can be used for solving a particular optimization problem there
1300        is an example included in the package. This is a stand-alone
1301        program, TSPSOL, which is intended for solving to optimality the
1302        symmetric Traveling Salesman Problem (TSP), a classical problem
1303        of the combinatorial optimization (see the file 'tspsol.c' in
1304        the subdirectory 'sample').
1305
1306GLPK 3.2 (release date: Jul 15, 2002)
1307
1308        New edition of the document "GLPK: Reference Manual" was
1309        included (see the files 'refman.latex', 'refman.dvi', and
1310        'refman.ps' in the subdirectory 'doc').
1311
1312        New edition of the document "GLPK: Modeling Language GLPK/L" was
1313        included (see the files 'lang.latex', 'lang.dvi', and 'lang.ps'
1314        in the subdirectory 'doc').
1315
1316        The following new API routines were added to the package:
1317
1318        lpx_transform_row (transform explicitly specified row);
1319        lpx_transform_col (transform explicitly specified column);
1320        lpx_prim_ratio_test (perform primal ratio test);
1321        lpx_dual_ratio_test (perform dual ratio test);
1322        lpx_interior (solve LP problem using interior point method);
1323        lpx_get_ips_stat (query status of interior point solution);
1324        lpx_get_ips_row (obtain row interior point solution);
1325        lpx_get_ips_col (obtain column interior point solution);
1326        lpx_get_ips_obj (obtain interior point value of obj.func.);
1327        lpx_read_lpm (read LP/MIP model written in GLPK/L);
1328        lpx_write_mps (write problem data using MPS format);
1329        lpx_print_ips (print interior point solution).
1330
1331        Detailed description of all these new API routines are given in
1332        the new edition of the reference manual.
1333
1334        New version of the stand-alone solver glpsol (which is based on
1335        the new API) was implemented.
1336
1337        So long as the new API (introduced in glpk 3.0) now provides
1338        all the functions, which were provided by the old API, the old
1339        API routines were removed from the package at all.
1340
1341GLPK 3.1 (release date: May 27, 2002)
1342
1343        A preliminary implementation of new API routines was completed
1344        and included in the package.
1345
1346        These new API routines provide much more flexible interaction
1347        between the application program, LP/MIP problem instances, and
1348        solver routines. Based on completely changed data structures
1349        they are, however, similar to the API routines and provide the
1350        same functionality. Please note that three routines, namely,
1351        solving LPs using interior point method, reading model written
1352        in the GLPK/L modeling language, and writing problem data in
1353        the MPS format, are not implemented in the new API, however,
1354        these routines are planned to be implemented in the next version
1355        of the package.
1356
1357        A description of the new API routines is given in the document
1358        "GLPK Reference Manual", a draft edition of which is included
1359        in the package (see the files 'refman.latex', 'refman.dvi', and
1360        'refman.ps' in the subdirectory 'doc').
1361
1362        Although the old API routines are kept in the package, they are
1363        no longer supported and will be removed in the future.
1364
1365GLPK 3.0.8 (release date: May 13, 2002)
1366
1367        A preliminary implementation of new API routines was included
1368        in the package. These new API routines are intended to provide
1369        much more flexible interaction between the application program,
1370        LP/MIP problem and solver routines. See the document "New GLPK
1371        API Routines" (the file 'newapi.txt' in the subdirectory 'doc')
1372        also included in the package.
1373
1374        The api routines glp_simplex2, glp_call_ipm1, glp_call_bbm1 were
1375        renamed, respectively, to glp_simplex, glp_interior, glp_integer
1376        in order to reflect changes in implementation. The api routines
1377        glp_call_rsm1, glp_simplex1, glp_pivot_in, glp_pivout_out were
1378        removed from the package since they are completely supreseded by
1379        the new API routines (however, these routines still can be found
1380        in the subdirectory 'oldsrc'). Please consult a new edition of
1381        the document "GLPK User's Guide" about all these changes in the
1382        existing api routines.
1383
1384        The document "GLPK Library Reference" was removed from the
1385        package (into the subdirectory 'oldsrc') since it describes the
1386        obsolete library routines, most of which are no longer used.
1387
1388GLPK 3.0.7 (release date: Apr 22, 2002)
1389
1390        A new, more efficient implementation of the primal/dual simplex
1391        method was included in the package. Due to some improvements the
1392        simplex-based solver allows solving many LP problems faster and
1393        provides more reliable results. Note that the new implementation
1394        is currently incomplete and available only via the api routine
1395        glp_simplex2.
1396
1397        All the changes are transparent on API level.
1398
1399GLPK 3.0.6 (release date: Mar 28, 2002)
1400
1401        New version of LU-factorization and basis maintenance routines
1402        (based on Forrest-Tomlin updating technique) was implemented.
1403        Since these new routines functionally supersede some routines
1404        (which implement other forms of the basis matrix) and make them
1405        obsolete, the latter were removed from the package (they still
1406        can be found in the subdirectory 'oldsrc').
1407
1408        All the changes are transparent on API level.
1409
1410GLPK 3.0.5 (release date: Jan 29, 2002)
1411
1412        New edition of the document "GLPK User's Guide" was included in
1413        the distribution. Now it describes all additional API routines,
1414        which were recently added to the package.
1415
1416        Structure of the package was re-organized in order to make its
1417        maintenance easier (all small files in the subdurectory 'source'
1418        were merged in bigger units). These changes are transparent for
1419        the user.
1420
1421GLPK 3.0.4 (release date: Dec 10, 2001)
1422
1423        A new, more efficient implementation of the two-phase primal
1424        simplex method was included in the package. Due to some new
1425        features (an advanced initial basis, projected steepest edge,
1426        recursive updating values and reduced costs) the new LP solver
1427        is faster and numerically more stable than the old one.
1428
1429        The new LP solver is available as API routine glp_simplex2 and
1430        has the same purpose as API routine glp_call_rsm1. For detailed
1431        specification see the file 'newapi.txt' in the directory 'doc'.
1432
1433        Now the new LP solver is also used by default to solve an
1434        initial LP problem in the branch-and-bound routine glp_call_bbm1
1435        instead the routine rsm1_driver. Note that the branch-and-bound
1436        procedure itself is still based on rsm1_driver.
1437
1438        The new LP solver is also used as default solver in GLPSOL for
1439        solving LP and MIP problems. In order to choose the old solver
1440        the option '--old-sim' can be specified in the command line.
1441
1442GLPK 3.0.3 (release date: Oct 03, 2001)
1443
1444        Some minor changes were made in the simplex method routines in
1445        order to improve numerical stability of the method.
1446
1447GLPK 3.0.2 (release date: Sep 24, 2001)
1448
1449        A new implementation of the basis maintaining routines was
1450        included in the package. These routines, which are based on so
1451        called FHV-factorization (a variety of LU-factorization) of the
1452        basis matrix and Gustavson's data structures, allows performing
1453        the main operations faster at the expense of some worsening
1454        numerical accuracy.
1455
1456        AFI (Advanced Form of the Inverse), which is the form of the
1457        basis matrix based on FHV-factorization, is available via the
1458        parameter form = 3 (on API level) or via the option --afi (in
1459        GLPSOL solver).
1460
1461GLPK 3.0.1 (release date: Aug 01, 2001)
1462
1463        Old GLPK API routines have been removed from the package.
1464
1465        New GLPK API routines were added:
1466
1467        - scaling routines;
1468
1469        - a routine for writing problem data in MPS format;
1470
1471        - a comprehensive driver to the simplex method;
1472
1473        - basis maintaining routines.
1474
1475        A description of the new API routines is given in the document
1476        "Additional GLPK API Routines". This document is included into
1477        the distribution in plain text format (see the file 'newapi.txt'
1478        in the subdirectory 'doc').
1479
1480        Now the distribution includes a non-trivial example of using
1481        GLPK as a base LP solver for Concorde, a well known program that
1482        solves Traveling Salesman Problem (TSP). For further details see
1483        comments in the file 'sample/lpglpk30.c'.
1484
1485GLPK 3.0 (release date: Jul 19, 2001)
1486
1487        Now GLPK is provided with new API, which being more flexible
1488        can be used in more complex algorithmic schemes.
1489
1490        New edition of the document "GLPK User's Guide" is included in
1491        the distribution. Now it completely corresponds to the new GLPK
1492        API routines.
1493
1494        Old API routines are not removed yet from the package, however
1495        they became obsolete and therefore should not be used. Since now
1496        the header glpk.h corresponds to new API, in order to compile
1497        existing programs that use old GLPK API routines the statement
1498
1499        #define GLP_OLD_API
1500
1501        should be inserted before the statement
1502
1503        #include "glpk.h"
1504
1505GLPK 2.4.1 (release date: Jun 14, 2001)
1506
1507        The document "Modeling language GLPK/L" is included into the
1508        distribution in texinfo format.
1509
1510        New edition of the document "GLPK User's Guide" is included in
1511        the distribution. Now it describes all additional API routines
1512        which were recently added to the package.
1513
1514GLPK 2.4 (release date: May 10, 2001)
1515
1516        Now GLPK includes an implementation of a preliminary version
1517        of the GLPK/L modeling language. This language is intended for
1518        writing mathematcal programming models. The name GLPK/L is
1519        derived from GNU Linear Programming Kit Language.
1520
1521        A brief description of the GLPK/L language is given in the
1522        document "GLPK/L Modeling Language: A Brief Description". This
1523        document is included into the distribution in plain text format
1524        (see the file 'language.txt' in the subdirectory 'doc').
1525
1526        The language processor (which is a program that analyzes model
1527        description written in GLPK/L and translates it to internal data
1528        structures) is available as the GLPK API routine.
1529
1530        The stand-alone solver GLPSOL now is able: a) to process model
1531        descriptions written in the GLPK/L language; b) to solve pure LP
1532        problems using the interior point method (therefore the program
1533        GLPIPM was removed from the package).
1534
1535GLPK 2.3 (release date: Apr 09, 2001)
1536
1537        New edition of the document "GLPK User's Guide" is included in
1538        the distribution. Now it describes all additional API routines
1539        which were recently added to the package.
1540
1541        The MIP solver was fully re-programmed in order to improve its
1542        robustness and performance. In particular, a basis recovering
1543        procedure was implemented (this procedure allows switching to
1544        the primal simplex method in case when the dual simplex method
1545        fails).
1546
1547GLPK 2.2 (release date: Mar 15, 2001)
1548
1549        Now GLPK includes a tentative implementation of the
1550        branch-and-bound procedure based on the dual simplex method for
1551        mixed integer linear programming (MIP).
1552
1553        Complete description of this new feature of the package is given
1554        in the preliminary document "Mixed Integer Linear Programming
1555        Using GLPK Version 2.2 (Supplement to GLPK User's Guide)". This
1556        document is included into the distribution in plain text format
1557        (see the file 'mip.txt' in the subdirectory 'doc').
1558
1559        The MIP solver (glp_integer) can be used as GLPK API routine in
1560        the same way as the pure LP solver (glp_simplex).
1561
1562        The stand-alone program 'glpsol' is now able to solve LP as well
1563        as MIP problems.
1564
1565        Note that the current version of GLPK MIP solver is based on
1566        easiest heuristics for branching and backtrackng. Therefore the
1567        solver is fit mainly for MIP problems which are not very hard
1568        and have few integer variables.
1569
1570GLPK 2.1 (release date: Feb 19, 2001)
1571
1572        The document "GLPK Implementation of the Revised Simplex Method"
1573        is included into the distribution. This document describes most
1574        of routines related to the revised simplex method.
1575
1576GLPK 2.0 (release date: Jan 25, 2001)
1577
1578        Now GLPK includes a tentative implementation of the primal-dual
1579        interior point method for large-scale linear programming.
1580
1581        The interior point solver can be used as GLPK API routine in the
1582        same manner as the simplex method solver (glp_simplex):
1583
1584        ret = glp_interior();
1585
1586        Note that currently the interior point solver implemented in
1587        GLPK doesn't include many important features, in particular:
1588
1589        * it can't process dense columns; therefore if the problem has
1590          dense columns, the solving will be extremely inefficient;
1591
1592        * it has no special features against numerical unstability;
1593          some problems may cause premature termination of the solving
1594          when the matrix A*D*A' becomes ill-conditioned;
1595
1596        * it computes only values of primal (auxiliary and structural)
1597          variables and doesn't compute values of dual variables (i.e.
1598          reduced costs) which are just set to zero;
1599
1600        * it doesn't identify optimal basis corresponding to the found
1601          interior point solution; all variables in the found solution
1602          are just marked as basic variables.
1603
1604        GLPK also includes a stand-alone program 'glpipm' which is a
1605        demo based on the interior point method. It may be used in the
1606        same way as the program 'glpsol' that is based on the simplex
1607        method.
Note: See TracBrowser for help on using the repository browser.