SCM

SCM Repository

[matrix] Diff of /pkg/ChangeLog
ViewVC logotype

Diff of /pkg/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1403, Sun Aug 20 16:15:59 2006 UTC revision 1852, Mon May 21 16:42:35 2007 UTC
# Line 1  Line 1 
1    2007-05-21  Martin Maechler  <maechler@stat.math.ethz.ch>
2    
3            * R/sparseMatrix.R (spMatrix): utility (T)sparse Matrix constructor;
4            * man/spMatrix.Rd: docu., including examples
5    
6            * R/Auxiliaries.R (sp.colMeans): etc, using a patched callGeneric(),
7            in order to make colMeans() etc fast *and* correct.
8            * R/sparseVector.R (replSPvec): "[<-" functionality for
9            sparseVectors; tested in
10            * tests/simple.R:
11    
12    2007-05-19  Martin Maechler  <maechler@stat.math.ethz.ch>
13    
14            * R/sparseMatrix.R (print.sparseSummary): and summary() method for
15            (very) sparse Matrices; output similar to Matlab's print().
16    
17    2007-05-17  Douglas Bates  <bates@stat.wisc.edu>
18    
19            * src/HBMM.c (Matrix_writeMatrixMarket): Write 1-based, not
20            0-based, indices (Jose Quesada <quesada@gmail.com>).
21    
22    2007-05-16  Douglas Bates  <bates@stat.wisc.edu>
23    
24            * R/CHMfactor.R: Added solve methods for a CHMfactor object.
25    
26    2007-05-16  Martin Maechler  <maechler@stat.math.ethz.ch>
27    
28            * R/Auxiliaries.R (sparsapply): new utility, much faster than
29            tapply1() for large sparse matrices.
30    
31    2007-05-15  Martin Maechler  <maechler@stat.math.ethz.ch>
32    
33            * R/Matrix.R (dim<-): reshape now via sparseVector.
34    
35            * R/sparseVector.R: methods and function for
36            * R/AllClass.R: new "sparseVector" class and daughters.
37            * NAMESPACE: export new classes
38    
39    2007-05-14  Martin Maechler  <maechler@stat.math.ethz.ch>
40    
41            * DESCRIPTION (Version): 0.99875-1
42            * src/Makefile.win: also remove Lapack code from here (cf. 04-25).
43    
44    2007-05-11  Martin Maechler  <maechler@stat.math.ethz.ch>
45    
46            * R/Tsparse.R ([, Tsparse): fix last case: *duplicated*, symmetric
47            indexing
48            * tests/indexing.R: test set for that.
49    
50    2007-05-08  Martin Maechler  <maechler@stat.math.ethz.ch>
51    
52            * R/Tsparse.R (replTmat): fix the case of *duplicated* index
53            entries.
54            * tests/indexing.R(out): add regression test for it
55    
56    2007-04-30  Martin Maechler  <maechler@stat.math.ethz.ch>
57    
58            * R/(l(dense|sparse))?Matrix.R (!): use 'x', not 'e1' as argument
59            name for "!" method definitions.
60    
61    2007-04-26  Martin Maechler  <maechler@stat.math.ethz.ch>
62    
63            * R/Tsparse.R (intI): new utility, used for "[" :
64                    Cleanup up there, and fixes for duplicated indices - more TODO!
65    
66            * tests/indexing.R(out): more tests
67    
68    2007-04-25  Douglas Bates  <bates@stat.wisc.edu>
69    
70            * DESCRIPTION,src/Makefile: require R>= 2.5.0 and remove Lapack
71            code that is now part of the R Lapack library.
72    
73            * src/init.c,inst/include/{Matrix_stubs.c,cholmod.h}:export
74            cholmod_factorize_p (used in lme4 for GLMMs and NLMMs).
75    
76    2007-04-21  Martin Maechler  <maechler@stat.math.ethz.ch>
77    
78            * R/Matrix.R (image): method for all Matrices, not just sparse ones.
79    
80    2007-04-17  Martin Maechler  <maechler@stat.math.ethz.ch>
81    
82            * R/Auxiliaries.R (tapply1): unname(.) -> colSums() etc don't end
83            up with extraneous names '0'...'<n-1>'
84    
85    2007-04-12  Martin Maechler  <maechler@stat.math.ethz.ch>
86    
87            * R/dgTMatrix.R (mat2dgT): care about NAs
88    
89    2007-04-11  Martin Maechler  <maechler@stat.math.ethz.ch>
90    
91            * R/kronecker.R: triangularity preserving methods
92    
93    2007-03-27  Martin Maechler  <maechler@stat.math.ethz.ch>
94    
95            * R/kronecker.R: new file collecting kronecker() methods in one
96                    place.  Goal: become much faster!
97    
98    2007-03-23  Martin Maechler  <maechler@stat.math.ethz.ch>
99    
100            * src/dtCMatrix.c (dtCMatrix_solve): use the new code from Kasper
101            Kristensen based cs_spsolve() instead of _lsolve & _usolve which
102            can be much faster.
103    
104            * tests/matprod.R: add regression tests for these (upper & lower).
105    
106    2007-03-19  Martin Maechler  <maechler@stat.math.ethz.ch>
107    
108            * R/Matrix.R (diff): method for our Matrices.
109    
110            * R/sparseMatrix.R (isDiagonal): check dim()!
111    
112    2007-03-17  Martin Maechler  <maechler@stat.math.ethz.ch>
113    
114            * R/Matrix.R (dim<-): new method for "reshape()" built on a
115            proposal from Tamas Papp.
116    
117    2007-03-16  Martin Maechler  <maechler@stat.math.ethz.ch>
118    
119            * R/AllGeneric.R: remove all  if(!isGeneric(.))  clauses
120    
121            * R/zzz.R (.onLoad, .onUnload): do *not* leave bind_activation(TRUE);
122            rather define and export  cBind() and rBind() only.
123            --> useRs *must* change code that used to have cbind()/rbind() !!
124    
125            * R/bind.R: change tests from cbind() to cBind() and similarly to rBind()
126            * R/bind.Rout.save: ditto
127    
128    2007-02-16  Douglas Bates  <bates@r-project.org>
129    
130            * DESCRIPTION (Date, Version): 0.9975-11 with new date
131            * src/dgCMatrix.c (R_to_CMatrix, compressed_to_TMatrix): remove
132            const modifier on declaration of the array 'valid'
133    
134    2007-02-12  Douglas Bates  <bates@stat.wisc.edu>
135    
136            * R/CHMfactor.R: Add image method (coercion to sparseMatrix).
137    
138    2007-02-05  Martin Maechler  <maechler@stat.math.ethz.ch>
139    
140            * DESCRIPTION (Date, Version): 0.9975-10 with new date.
141    
142            * R/Ops.R (Arith): make sure Csparse o Csparse also works for e.g. ntCMatrix
143            * tests/simple.R: test the above and some of these coercions:
144    
145            * R/nsparseMatrix.R: coercing "Csparse" to "lsparseMatrix";
146                    be careful to avoid infinite recursion, using new coercions in
147            * R/ngCMatrix.R and nsC... and ntC...
148            * R/lsparseMatrix.R: ditto
149    
150            * R/SparseM-conv.R: more conversion, notably for triplet matrices.
151    
152            * src/dgCMatrix.c (R_to_C_Matrix): port Doug's fix and
153            * R/Rsparse.R: reactivate .Call()s
154    
155            * tests/Class+Meth.R: a bit more on actual classes
156    
157    2007-02-04  Douglas Bates  <bates@stat.wisc.edu>
158    
159            * src/dgCMatrix.c (compressed_to_TMatrix): fix memory bug using strdup()
160    
161    2007-02-03  Martin Maechler  <maechler@stat.math.ethz.ch>
162    
163            * DESCRIPTION (Version): 0.9975-10 to upload
164    
165            * tests/Class+Meth.R (tstMatrixClass): require coercions to
166            specific classes less unconditionally.
167    
168            * R/Auxiliaries.R: get rid of as_Tsparse() and as_Rsparse()
169    
170            * R/Tsparse.R (triu): etc, use as(*, "TsparseMatrix") instead of as_Tsparse()
171    
172            * R/Rsparse.R (.R.2.T): R-level workaround using compressed_to_TMatrix.
173            * R/Rsparse.R (.R.2.C): R-level workaround since C-level
174            R_to_CMatrix segfaults on one platform.
175            Eliminate most coercion method to *specific* classes, and replace
176            with virtual classes coercions.
177    
178    
179    2007-02-01  Martin Maechler <maechler@stat.math.ethz.ch>
180    
181            * src/init.c: export the CHM...._validate() placeholders, since
182            they *are* called.
183    
184            * tests/Class+Meth.R (classCanCoerce): and starting to test
185            all as(<from>, <to>)
186    
187    2007-01-30  Martin Maechler  <maechler@stat.math.ethz.ch>
188    
189            * R/Tsparse.R ([): more care when subsetting triangular Tsparse
190            * tests/indexing.R: tested now
191            * tests/indexing.Rout.save: updated
192    
193            * src/Csparse.c (Csparse_to_dense): use Rkind = -1 for PATTERN to
194            * src/chm_common.c (chm_dense_to_SEXP): return "ngeMatrix" when
195            appropriate.
196    
197            * NAMESPACE: export a trivial
198            * R/Matrix.R: drop() Matrix-method
199    
200            * R/AllClass.R: moved all prototypes to virtual super classes.
201    
202            * R/Rsparse.R: many more coercions to have less exceptions in
203            * tests/Class+Meth.R:
204    
205            * R/Ops.R (Compare): tweak for case with NA
206            * tests/simpl.R: hence another 'FIXME' eliminated
207    
208    2007-01-29  Martin Maechler  <maechler@stat.math.ethz.ch>
209    
210            * R/diagMatrix.R (solve): the obvious methods for diagonalMatrix
211            objects.
212    
213            * tests/Class+Meth.R (tstMatrixClass): now testing diag(), nnzero(),
214              and more of "!", "&", "|", all, any; coercions
215    
216            * R/Rsparse.R: many coercions (which enable quite a few other
217            methods), thanks to enhancements in
218            * src/dgCMatrix.c (R_to_CMatrix): new, and
219            * src/dgCMatrix.c (compressed_to_TMatrix): now for (d,l,n) ,
220            symmetric & triangular and ..RMatrix objects.
221    
222            * src/TMatrix_as.c (Matrix_T_as_DENSE,Matrix_T_as_GENERAL):
223            renamed file from src/dsTMatrix.c;
224            now dealing with symmetric and triangular Tsparse coercions, both
225            to dense and general.
226    
227    2007-01-27  Martin Maechler  <maechler@stat.math.ethz.ch>
228    
229            * src/dsTMatrix.c: has now "l" and "n" methods besides the "d" ones.
230    
231            * R/Ops.R (Arith): <dgCMatrix> o <numeric> now remains sparse
232            where sensible when the <numeric> is of length > 1.
233    
234    2007-01-26  Martin Maechler  <maechler@stat.math.ethz.ch>
235    
236            * R/Matrix.R ([<-): for M[] <- value: fix length
237    
238    2007-01-25  Martin Maechler  <maechler@stat.math.ethz.ch>
239    
240            * R/Auxiliaries.R (n2l_Matrix): new, to be used in
241            * R/ndenseMatrix.R: new coercions n* -> l*
242    
243    2007-01-22  Martin Maechler  <maechler@stat.math.ethz.ch>
244    
245            * R/triangularMatrix.R: new file; simple triu() and tril() methods.
246    
247            * R/Ops.R ("Logic"): and other "Ops", many updates
248    
249    2007-01-18  Martin Maechler  <maechler@stat.math.ethz.ch>
250    
251            * src/Mutils.h (SET_DimNames): new utility
252    
253            * R/Auxiliaries.R (nnzero): improved and now exported via
254    
255            * NAMESPACE: + nnzero(); length() == prod(dim(.)) method for all "Matrix" objects
256    
257    2007-01-17  Martin Maechler  <maechler@stat.math.ethz.ch>
258    
259            * R/diagMatrix.R (!): fix typo.
260    
261    2007-01-16  Martin Maechler  <maechler@stat.math.ethz.ch>
262    
263            * R/Auxiliaries.R (as_Csparse): and quite a few others:
264                    allow to pass class definition --> speedup
265            * R/sparseMatrix.R: apply the above
266    
267            * R/Csparse.R: coercion Csparse* to dense* now preserves shape properties.
268    
269            * src/Mutils.h (mMatrix_as_geMatrix): new, based on
270            * src/Mutils.c (dup_mMatrix_as_geMatrix): new; generalization of
271            old dup_mMatrix_as_dgeMatrix), eliminating a long-standing "FIXME".
272    
273            * src/dense.c (dense_to_Csparse): use new mMatrix_as_geMatrix()
274    
275            * R/denseMatrix.R (.dense2C): based on dense_to_Csparse: name it,
276            and use it for "sparse*" as well, since it's faster than the
277            as_Csparse(.) way.
278    
279    2007-01-15  Martin Maechler  <maechler@stat.math.ethz.ch>
280    
281            * R/Ops.R ("Logic"): more methods, notably an  <lgC> o <lgC> one.
282    
283    2007-01-12  Martin Maechler  <maechler@stat.math.ethz.ch>
284    
285            * R/Tsparse.R (.TM.repl.i.2col): new internal function to be used
286            as method for   M[ ij ] <- v
287            * R/Csparse.R:: go via Tsparse for "M[ij] <- v"
288    
289            * R/Ops.R: "Compare" for (C|R)sparse: need pointer slot for all
290            FALSE answer
291    
292            * R/Csparse.R (replCmat): fix the "all non-zero" case with reordering
293            * tests/indexing.R: test it, and some of the above
294    
295    2007-01-05  Martin Maechler  <maechler@stat.math.ethz.ch>
296    
297            * R/Auxiliaries.R (is_duplicatedT): new utility
298    
299    2007-01-05  Douglas Bates  <bates@stat.wisc.edu>
300    
301            * src/init.c (R_init_Matrix): export cholmod_scale
302    
303    2006-12-30  Martin Maechler  <maechler@stat.math.ethz.ch>
304    
305            * R/zzz.R (tmp): for R >= 2.5.0, extend formals of our
306            base::as.matrix to (x, ...)
307    
308    2006-12-28  Martin Maechler  <maechler@stat.math.ethz.ch>
309    
310            * R/Ops.R ("Arith" etc): move almost all "Ops" methods to new R
311            file; start using "Logic", hence
312    
313            * DESCRIPTION (Depends): R >= 2.4.1 (since we want "Logic")
314            * NAMESPACE: import and export "Logic"
315    
316    2006-12-27  Martin Maechler  <maechler@stat.math.ethz.ch>
317    
318            * src/zpotfr.f and dependencies: use LAPACK 3.1 version
319            only needed previously to R version 2.5.0.
320    
321    2006-12-26  Martin Maechler  <maechler@stat.math.ethz.ch>
322    
323            * DESCRIPTION (Date, Version): 0.9975-8, ready for release
324    
325            * R/Tsparse.R (replTmat): fix subassignment of triangular
326            * R/Csparse.R (replCmat): ditto
327            * tests/indexing.R: more tests, incl the above fix
328    
329    2006-12-23  Martin Maechler  <maechler@stat.math.ethz.ch>
330    
331            * R/Auxiliaries.R (drop0): extend for non CSparse
332            * R/Auxiliaries.R (diagU2N): should work for all sparseMatrix
333    
334            * src/Csparse.c (Csparse_to_Tsparse, Csparse_general_to_symmetric):
335                    use uplo correctly (!); other places: use uplo_P() macro
336    
337            * R/Csparse.R (replCmat): call diagU2N() when needed
338            * R/Tsparse.R (replTmat): ditto
339    
340            * src/dtCMatrix.c (tCMatrix_validate): new
341            * src/dtTMatrix.c (tTMatrix_validate): new, used in
342            * R/AllClass.R: for validity of dtC, dsC, and dtT, dsT.
343    
344            * R/diagMatrix.R (replDiag): to use in [<-
345    
346    2006-12-22  Martin Maechler  <maechler@stat.math.ethz.ch>
347    
348            * R/Auxiliaries.R (as_Csparse2, as_geSimpl): new functions;
349                    also more general diagU2N().
350    
351    2006-12-21  Martin Maechler  <maechler@stat.math.ethz.ch>
352    
353            * R/bind2.R: new file for all cbind2(), rbind() methods moved here
354            from R/<foo>Matrix.R files. Better diagonal & improved sparse methods.
355    
356    2006-12-20  Martin Maechler  <maechler@stat.math.ethz.ch>
357    
358            * tests/bind.R: a few more cases
359            * R/Auxiliaries.R (.M.kind): also work for atomic vectors
360    
361            * R/denseMatrix.R (cbind2/rbind2): moved here (and generalized) from
362            * R/ddenseMatrix.R (cbind2/rbind2)
363            * R/Tsparse.R (replTmat): final(?!) fix for "[<-" ..
364            * tests/indexing.R
365            * tests/indexing.Rout.save: updated
366    
367    2006-12-18  Martin Maechler  <maechler@stat.math.ethz.ch>
368    
369            * R/Tsparse.R (replTmat): fixed a remaining "[<-" bug in
370            * tests/indexing.R
371    
372    2006-12-15  Martin Maechler  <maechler@stat.math.ethz.ch>
373    
374            * R/sparseMatrix.R (prSpMatrix): "." alignment much improved:
375            align with proper position of "0", i.e., right for integers.
376            argument 'align' by default is "fancy".
377    
378    2006-12-14  Martin Maechler  <maechler@stat.math.ethz.ch>
379    
380            * R/sparseMatrix.R: delegate "Compare" to "Csparse.."
381            * R/Csparse.R: and fix "Compare" for more cases.
382            * tests/Class+Meth.R: test some of these (m == m, m != m)
383    
384    2006-12-13  Martin Maechler  <maechler@stat.math.ethz.ch>
385    
386            * R/lsparseMatrix.R: all() and any() methods
387            * R/ldenseMatrix.R:      ditto
388            * NAMESPACE, R/Matrix.R: ditto
389    
390            * man/all-methods.Rd: document them minimally
391    
392            * tests/simple.R: add a few examples for these
393    
394    2006-12-11  Martin Maechler  <maechler@stat.math.ethz.ch>
395    
396            * R/Tsparse.R ([): fix long standing typo in symmetric case
397            * man/dsCMatrix-class.Rd: add example exhibiting the above case
398    
399    2006-12-10  Douglas Bates  <bates@stat.wisc.edu>
400    
401            * src/CHMfactor.c (CHMfactor_to_sparse): change LDL factorization
402            to LL before converting to a sparse matrix. (The LDL form can be
403            converted to a sparse matrix but it is implicitly a unit
404            triangular matrix and a diagonal matrix overwritten on the diagonal.)
405    
406    2006-12-09  Douglas Bates  <bates@stat.wisc.edu>
407    
408            * src/chm_common.c (chm_factor_to_SEXP): allocate and fill the Dim slot.
409    
410    2006-12-08  Douglas Bates  <bates@stat.wisc.edu>
411    
412            * DESCRIPTION (Version): updated -> release 0.9975-7
413            * src/{init.c,chm_common.c}, inst/include/*: export cholmod_analyze_p
414    
415    2006-11-30  Martin Maechler  <maechler@stat.math.ethz.ch>
416    
417            * R/diagMatrix.R (%*%): write a direct [diag o Csparse] method
418    
419    2006-11-29  Douglas Bates  <bates@stat.wisc.edu>
420    
421            * src/dgeMatrix.c (dgeMatrix_solve): Check error code from dgetri.
422            * tests/dg_Matrix.R: Add Barry Rowlingson's test of a matrix that
423            is exactly singular.
424    
425    2006-11-07  Martin Maechler  <maechler@stat.math.ethz.ch>
426    
427            * DESCRIPTION (Date): updated -> release 0.9975-6
428    
429    2006-11-06  Martin Maechler  <maechler@stat.math.ethz.ch>
430    
431            * R/Csparse.R (replCmat): symmetric indexing of symmetric matrix
432            now returns symmetric.
433    
434            * R/zzz.R ("diag<-"): replace "diag<-" in base for R <= 2.4.x
435    
436            * R/Matrix.R (.M.sub.i.2col):  new, for  M[ cbind(i,j) ] indexing.
437            * R/Matrix.R (.M.repl.i.2col): new, for  M[ cbind(i,j) ] <- value
438    
439            * R/Auxiliaries.R (.type.kind): added
440    
441    2006-11-04  Martin Maechler  <maechler@stat.math.ethz.ch>
442    
443            * src/cs.[ch]: updated to CSparse Version 2.0.3 by simply
444            running src/CSparse_install.sh
445    
446            * R/denseMatrix.R: "[": keep symmetric on symmetric indexing.
447    
448    2006-11-03  Martin Maechler  <maechler@stat.math.ethz.ch>
449    
450            * src/dsCMatrix.c (dsCMatrix_Csparse_solve): new
451            * R/dsCMatrix.R (solve): "fully-sparse" using the above.
452    
453            * R/AllClass.R: "pMatrix" now also inherits from "generalMatrix"
454    
455            * tests/Class+Meth.R (tstMatrixClass): now assure
456            the (important in method programming) property :
457    
458                    ###>>  Every "Matrix" is either
459                    ###>>  "general*", "symmetric*", "triangular*" or "diagonal*"
460    
461                    (where "*" stands for "Matrix")
462    
463            * R/Auxiliaries.R (diagU2N): now .Call()s Csparse_diagU2N for <Csparse>
464    
465            * R/dMatrix.R (Compare(<dMatrix>,<dMatrix>): update and
466            * tests/validObj.R: checks for "comparison"
467    
468            * R/sparseMatrix.R ([): improved indexing for sparse;
469                    trying to keep <symmetric>[ n, n] symmmetric
470    
471            * tests/indexing.R: indexing for logical sparse now ok
472    
473    
474    2006-11-02  Martin Maechler  <maechler@stat.math.ethz.ch>
475    
476            * src/Tsparse.c: use xTsparse_validate() , and hence remove
477    
478            * src/{ltC,lsC,lgT}Matrix.[ch]: removed
479    
480    2006-11-02  Martin Maechler  <maechler@stat.math.ethz.ch>
481    
482            * R/AllClass.R (Matrix-class): check length of dimnames in validity.
483    
484            * tests/simple.R: validObject() checking the above.
485    
486            * src/dgCMatrix.c (xCMatrix_validate): new, small and simple,
487            replacing both dgCMatrix_validate and lgCM*.
488    
489            * src/Csparse.c (Csparse_dense_prod, etc): do not lose dimnames;
490            fix dimnames setting in other places.
491            * src/chm_common.c (chm_dense_to_SEXP): now can pass dimnames
492    
493    2006-11-01  Martin Maechler  <maechler@stat.math.ethz.ch>
494    
495            * R/Csparse.R,src/Csparse.c, etc: tcrossprod(<CsparseM>,<CsparseM>)
496    
497            * R/sparseMatrix.R (isSymmetric): drop 'factors' slot for
498            symmetry test, via
499            * R/Auxiliaries.R (.as.dgC.0.factors):
500    
501    2006-11-01  Douglas Bates  <bates@r-project.org>
502    
503            * R/Csparse.R,src/Csparse.c,tests/matprod.R,
504            man/CsparseMatrix-class.Rd: crossprod(<CsparseMatrix>,
505            <CsparseMatrix>) added
506    
507    2006-10-30  Martin Maechler  <maechler@stat.math.ethz.ch>
508    
509            * tests/matprod.R: add a variation of Harri's example
510    
511            * R/dsparseMatrix.R: fix crossprod(<dsparse>, <dge>) to *not*
512            recursive infinitely.
513    
514            * R/dgCMatrix.R: + solve(<sparse>, <sparse>)
515    
516            * tests/indexing.R: add test for the "<" bug fixed 10-27 in R/dMatrix.R
517    
518    2006-10-28  Martin Maechler  <maechler@stat.math.ethz.ch>
519    
520            * tests/Class+Meth.R (tstMatrixClass): more: use non-trivial
521            matrix if possible; test m+m == 2*m; now test dgRMatrix.
522            * R/dgRMatrix.R (.to.dgR): a few more coercions, in order to
523            satisfy the above test.
524    
525    2006-10-27  Martin Maechler  <maechler@stat.math.ethz.ch>
526    
527            * R/Matrix.R (Ops):  <Matrix> o <matrix>  method added
528    
529            * R/dgCMatrix.R: solve(a, b="missing") based on
530            * src/dgCMatrix.c (dgCMatrix_matrix_solve): extend to work with
531            RHS = NULL.
532    
533            * R/diagMatrix.R (diagdiagprod): extend %*% etc to ldiMatrix;
534            add more (needed) [t]crossprod() methods.
535    
536            * man/ddiMatrix-class.Rd: more info, notably on 'diag'
537    
538            * R/Auxiliaries.R (as_CspClass): cleanup
539            (drop0): internal utility for "Csparse_drop(*, 0)"
540            (.bail.out.2): encourage active feedback
541    
542    2006-10-26  Martin Maechler  <maechler@stat.math.ethz.ch>
543    
544            * R/dMatrix.R(Compare): new(), then slots [no validity check]
545    
546            * src/Csparse.c (Csparse_validate): fixed (and more efficient in
547            non-valid or 'sorted' case).
548    
549            * R/dsparseMatrix.R: add "chol" method.
550            * R/ddenseMatrix.R: ditto
551    
552            * R/diagMatrix.R (Ops): group methods for  <diagonal> o <sparse>
553            * NAMESPACE (Ops)
554            * R/diagMatrix.R (diag2T): simple utility used "higher level"
555            coercion; deprecating direct lower level coercions.
556    
557            * R/*.R (seq): use seq_len() and seq_along() where possible.
558    
559    
560    2006-10-23  Martin Maechler  <maechler@stat.math.ethz.ch>
561    
562            * DESCRIPTION (Version): 0.9975-5 ready for release
563    
564    2006-10-20  Douglas Bates  <bates@stat.wisc.edu>
565    
566            * src/init.c (R_init_Matrix): export more cholmod CCallable functions.
567    
568    2006-10-20  Martin Maechler  <maechler@stat.math.ethz.ch>
569    
570            * R/AllClass.R (corMatrix): add 'validity' check;
571                    comment out unused "LDL" class definition
572            * NAMESPACE: mention, but do not export "LDL" class
573            * R/corMatrix.R: new (simple), needed for R-devel with
574            * tests/Class+Meth.R (tstMatrixClass): 1 exception for corMatrix
575              coerce and t() exceptions for all 5 'Mat.MatFact' classes.
576    
577    2006-10-19  Douglas Bates  <bates@stat.wisc.edu>
578    
579            * src/chm_common.h: Add R_cholmod_start to initialize cholmod to
580            use Rprintf and R's error handling.
581    
582    2006-10-17  Martin Maechler  <maechler@stat.math.ethz.ch>
583    
584            * R/diagMatrix.R (%*%): rep(*, each = .) in Matrix %*% diagonal.
585            * tests/matprod.R: add tests for the bug fixed.
586    
587    2006-10-11  Douglas Bates  <bates@stat.wisc.edu>
588    
589            * src/HBMM.[ch]: remove HarwellBoeing format for writing.
590            * src/SOURCES_C.mkf (SOURCES_C): no longer compile iohb.c
591    
592    2006-10-06  Douglas Bates  <bates@stat.wisc.edu>
593    
594            * R/d[gs]CMatrix.R: deprecate the writeHB function.  Use writeMM instead.
595    
596    2006-10-06  Martin Maechler  <maechler@stat.math.ethz.ch>
597    
598            * DESCRIPTION (Version): 0.9975-3
599    
600            * R/diagMatrix.R (bdiag): new function constructing block diagonal
601            (sparse) matrices.
602            * man/bdiag.Rd: docu + examples
603    
604            * R/Csparse.R (replCmat): calling new Csparse_drop() now.
605    
606            * src/Csparse.c (Csparse_general_to_symmetric, Csparse_drop): new functions
607            * R/lsCMatrix.R: three more coercions to lsC (thanks to the above)
608    
609            * R/diagMatrix.R (Diagonal): '[<-' method for diag.matrices such
610            that result is sparse or diagonal (and not dense).
611    
612            * man/Subassign-methods.Rd: fix examples
613    
614            * R/Matrix.R (Matrix):  Matrix(0, *) or Matrix(*, sparse=TRUE)
615            should always return a sparse (and not sometimes a diagonal) matrix.
616    
617    2006-10-05  Martin Maechler  <maechler@stat.math.ethz.ch>
618    
619            * R/Matrix.R ([<-): also for value "Matrix" or "matrix"
620    
621    2006-10-04  Douglas Bates  <bates@stat.wisc.edu>
622    
623            * DESCRIPTION (Version): 0.9975-2
624    
625            * inst/include/Matrix_stubs.c (M_cholmod_sparse_to_triplet): export more symbols
626    
627    2006-10-02  Douglas Bates  <bates@R-project.org>
628    
629            * tests/dg_Matrix.R: Simplify test taking into account new code.
630    
631    2006-09-29  Martin Maechler  <maechler@stat.math.ethz.ch>
632    
633            * R/Csparse.R (replCmat): improve for missing i / j in non-simple cases
634    
635            * R/lsTMatrix.R: new files w/ missing methods
636            * R/nsTMatrix.R:  "  'for completeness'
637    
638            * tests/Class+Meth.R: a bit less 'not.ok.classes'
639    
640            * R/Tsparse.R (t): generalized "t" method from "dgT*" to "Tsparse*"
641    
642    
643    2006-09-28  Douglas Bates  <bates@R-project.org>
644    
645            * src/dppMatrix.h: Ensure definition of dspMatrix_validate is
646            included.
647    
648            * src/init.c, inst/include/{Matrix.h,Matrix_stubs.h,cholmod.h}:
649            Export C-callable functions used in Zt_create in lme4.
650    
651    2006-09-28  Martin Maechler  <maechler@stat.math.ethz.ch>
652    
653            * DESCRIPTION (Version): 0.9975-1
654    
655            * tests/simple.R: less checks fail; using NA, found that our
656                    kronecker() is not base-compatible with NA's.
657    
658            * R/dMatrix.R: "Compare" method now implemented for all cases
659    
660            * R/Auxiliaries.R (indTri): == which([lower/upper].tri( * )) new utility
661    
662            * man/dtpMatrix-class.Rd: mention length of 'x' slot
663    
664            * src/dtpMatrix.c (dtpMatrix_validate): fix check
665            * src/dspMatrix.c (dspMatrix_validate): ditto
666    
667            * R/dtTMatrix.R (gt2tT): fix ("l" -> "n")
668              including coercion to [nl]tTMatrix.
669    
670            * R/diagMatrix.R (show): print a header line as for other classes.
671    
672    2006-09-27  Martin Maechler  <maechler@stat.math.ethz.ch>
673    
674            * src/Makefile.win (SUBDIRS): fix typo
675    
676    2006-09-19  Martin Maechler  <maechler@stat.math.ethz.ch>
677    
678            * DESCRIPTION (Date): ready to release 0.9975-0 to CRAN
679    
680    2006-09-18  Douglas Bates  <bates@stat.wisc.edu>
681    
682            * R/[CT]sparse.R (crossprod and tcrossprod): Handle the cases for x
683            symmetric and y missing in R code using %*% (cholmod_aat doesn't
684            accept a symmetric matrix).
685            * tests/group-methods.R: Uncomment test of crossprod applied to
686            lsCMatrix objects.
687    
688    2006-09-18  Martin Maechler  <maechler@stat.math.ethz.ch>
689    
690            * R/AllClass.R (symmetricMatrix): add validity method (available
691            in C for a long time). Many "n..Matrix": drop (wrong) validity arg.
692    
693            * src/lgCMatrix.c (lgCMatrix_validate): check 'x' slot (!)
694    
695            * tests/indexing.Rout.save:
696            * tests/indexing.R: additions, mainly for 'lsparse'
697    
698            * R/diagMatrix.R (Diagonal) & coercion to lgTMatrix: fixes for NA
699            case.
700    
701            * R/Auxiliaries.R (nz.NA): new utility now used in nnzero()
702    
703    2006-09-16  Martin Maechler  <maechler@stat.math.ethz.ch>
704    
705            * R/sparseMatrix.R (prSpMatrix): print logical NAs "visibly"
706                    as 'N' (1-letter - Ok?)
707    
708            * tests/group-methods.R: add test for logical + NAs
709    
710            * R/dMatrix.R ("Compare"): fix to work with NA's
711    
712            * R/AllClass.R: "Cholesky" etc now inherit from MatrixFactorization.
713    
714            * src/lgCMatrix.c (ncsc_to_matrix): renamed from lcsc_to_matrix()
715                    which is implemented.
716    
717    2006-09-15  Martin Maechler  <maechler@stat.math.ethz.ch>
718    
719            * src/chm_common.c: coerce logical <-> double instead of
720            typecasting;  needed for "l" matrix handling in cholmod.
721    
722            * tests/other-pkgs.R (graph): small extension in "graph" checks.
723    
724            * R/sparseMatrix.R (graphNEL -> Tsparse): method for weight case.
725            (Tsp2grNEL): other fixes needed
726    
727    2006-09-11  Martin Maechler  <maechler@stat.math.ethz.ch>
728    
729            * R/AllClass.R ("nMatrix"): and subclasses for "nonzero pattern"
730                    Matrices, since "lMatrix", also "lsparseM" can have NA
731            * R/ndenseMatrix.R, etc: new source files
732            * man/nsparseMatrix-classes.Rd, etc: new help files
733            * tests/: adaptions
734            * src/chm_common.c (chm_dense_to_SEXP): and others:
735                    new 'Rkind' argument: "l*" and "d*" both use CHOLMOD_REAL
736            * src/Csparse.c, etc: ditto
737    
738    2006-09-11  Douglas Bates  <bates@R-project.org>
739    
740            * src/Mutils.[ch],init.c inst/include/*.h: Move the
741            alloc_d**Matrix functions to the lme4 package.
742    
743    2006-09-09  Douglas Bates  <bates@R-project.org>
744    
745            * src/dsCMatrix.c (dsCMatrix_Cholesky): igoring LDL = FALSE now
746            fixed
747    
748    2006-09-09  Martin Maechler  <maechler@stat.math.ethz.ch>
749    
750            * R/lMatrix.R: new
751    
752            * R/sparseMatrix.R (Tsp2grNEL): do not yet use graph::foo()
753    
754            * R/dgeMatrix.R: do not define tcrossprod() methods for "matrix"
755            * man/tcrossprod.Rd: ditto
756    
757    2006-09-08  Douglas Bates  <bates@stat.wisc.edu>
758    
759            * inst/include/Matrix_stubs.c,Matrix.h: Add declarations and stubs
760            for exported functions
761            * src/Makefile, src/CHOLMOD/Lib/Makefile, src/Metis,CAMD,CCOLAMD:
762            Remove partitioning algorithms for sparse matrix reordering.  The
763            copyright on the Metis code was problematic and the methods were
764            rarely used.
765            * src/triplet_to_col.[ch],MMHB.[ch]: Remove triplet_to_col.  Such
766            operations are now done entirely in CHOLMOD code.
767    
768    2006-09-06  Douglas Bates  <bates@R-project.org>
769    
770            * src/Mutils.h: Remove functions that are no longer used.
771    
772    2006-09-04  Douglas Bates  <bates@R-project.org>
773    
774            * src/dtCMatrix.c (dtCMatrix_validate): rename functions.
775            * src/DEPS.mkf: update
776    
777    2006-09-02  Martin Maechler  <maechler@stat.math.ethz.ch>
778    
779            * created branches/Matrix-for-R-2.3.x;  on trunk: do
780            * DESCRIPTION (Version): 0.9975-0
781            (Depends): R (>= 2.4.0)
782    
783    2006-09-01  Douglas Bates  <bates@R-project.org>
784    
785            * R/sparseMatrix.R: Added direct method to CsparseMatrix from graphNEL
786    
787    2006-09-01  Martin Maechler  <maechler@stat.math.ethz.ch>
788    
789            * R/sparseMatrix.R: add coercion from "ANY" to "sparseMatrix"
790            * R/denseMatrix.R:  add coercion from "ANY" to "denseMatrix"
791    
792            * R/Matrix.R ([): use nargs() to disambiguate M[i] and M[i,]
793    
794    2006-08-31  Martin Maechler  <maechler@stat.math.ethz.ch>
795    
796            * R/sparseMatrix.R (Arith): moved Arith group method one-level up
797            from "dsparse" to "sparse" and now go via "Csparse" instead of "dgC"
798            * R/dsparseMatrix.R: ditto
799            * R/Csparse.R: ditto
800    
801    2006-08-31  Martin Maechler  <maechler@stat.math.ethz.ch>
802    
803            * R/dMatrix.R (Compare): improve availability of "<", etc
804            * R/Auxiliaries.R (asTuniq): new; also make use of R 2.4.x print(*,max)
805    
806    2006-08-30  Martin Maechler  <maechler@stat.math.ethz.ch>
807    
808            * R/dgCMatrix.R: aargh: "Arith(<dgC>, numeric)" was wrong because
809            of a 0-index which was used as 1-index (..hmm)
810    
811            * R/sparseMatrix.R (prSpMatrix): fix printing an all-0 sparse Matrix
812    
813            * R/Auxiliaries.R (all0, is0): for 0-testing in presence of NA's
814            * R/Auxiliaries.R (isTriMat): use all0() for 0-testing.
815            (.is.diagonal): ditto
816    
817            * R/lgTMatrix.R: as("matrix", "lgTMatrix"): warn about NA's
818    
819            * R/Matrix.R (Matrix): also work for NA data
820    
821    2006-08-28  Martin Maechler  <maechler@stat.math.ethz.ch>
822    
823            * R/Matrix.R (Matrix): + 'forceCheck' argument; dimnames setting
824            in all cases.
825    
826    2006-08-27  Douglas Bates  <bates@R-project.org>
827    
828            * src/dense.[ch],init.c, R/ddenseMatrix.R, man/band.Rd
829            (ddense_band): Added triu, tril and band for ddenseMatrix objects.
830    
831    2006-08-25  Martin Maechler  <maechler@stat.math.ethz.ch>
832    
833            * src/Mutils.c (dup_mMatrix_as_dgeMatrix): added all subclasses of
834            subclasses of "ddenseMatrix"
835            * src/init.c et al: outcomment dtrM*_as_dge* and dsyM*_as_dge*_
836            * R/ddenseMatrix.R et al: setAs("ddenseMatrix", "dgeMatrix", ..)
837            instead of half dozen specialized ones.
838    
839    2006-08-25  Douglas Bates  <bates@R-project.org>
840    
841            * R/lmer.R (qqmath method): Bug fix provided by Emmanuel Tillard
842            <tillard@cirad.fr> - ordering standard errors to match effects.
843    
844    2006-08-24  Douglas Bates  <bates@R-project.org>
845    
846            * src/lsCMatrix.c (lsCMatrix_trans): Remove lsCMatrix_chol based
847            on R_ldl.
848            * R/lCholCMatrix.R,
849            src/[dl]CholCMatrix.[ch],R_ldl.[ch],Metis_utils.[ch]: removed
850            * src/dsCMatrix.c (dsCMatrix_to_dgTMatrix): use CHOLMOD
851            * many files in ./R and ./src: Use more general version of
852            dup_mMatrix_as_dgeMatrix to simplify method definitions.
853            * src/Mutils.c (dup_mMatrix_as_dgeMatrix): Add ddiMatrix,
854            dtpMatrix, dspMatrix and dppMatrix conversions.
855    
856    
857    2006-08-23  Douglas Bates  <bates@R-project.org>
858    
859            * R/AllClass.R,lCholCMatrix.R,src/SOURCES_C.mkf,init.c,NAMESPACE:
860            Remove classes lCholCMatrix and dCholCMatrix based on R_ldl code.
861            * src/dgeMatrix.c: Ensure 'factors' slot exists in result of
862            dgeMatrix_crossprod (may need to do this in other places).
863            * R/AllGeneric.R,dsCMatrix.R, src/dscMatrix.[ch]: Add Cholesky
864            generic and method for dsCMatrix.  Use CHOLMOD for chol().
865    
866    2006-08-22  Douglas Bates  <bates@R-project.org>
867    
868            * src/Mutils.c (dup_mMatrix_as_dgeMatrix): updated for general
869            types of classed Matrices, matrices or numeric or logical vectors.
870            * src/init.c: register dup_mMatrix_as_dgeMatrix for coercions.
871            * src/chm_common.c,Mutils.h (as_cholmod_factor): Move check_class
872            to Mutils.h (as an inline) and change name to Matrix_check_class;
873            fixes in as_cholmod_factor and chm_factor_to_SEXP.
874            * src/dsCMatrix.[ch]: Use CHOLMOD for dsCMatrix_chol and
875            dsCMatrix_matrix_solve.  Comment out vestigial functions.
876            * src/Csparse.c: use diag_P and uplo_P macros.
877    
878    2006-08-21  Douglas Bates  <bates@R-project.org>
879    
880            * src/lmer.c (internal_mer_RZXinv): Fix memory leak caught by
881            valgrind.
882            * tests/matprod.R: Add tests to verify that 'solve' and '%*%' are
883            inverses.
884            * src/sparseQR.c (sparseQR_validate): add new validation test, fix
885            -Wall warnings.
886            * src/dppMatrix.c,dtrMatrix.c,dgCMatrix.c,dgeMatrix.c,dspMatrix.c:
887            Use dup_mMatrix_as_dgeMatrix.
888    
889  2006-08-20  Douglas Bates  <bates@R-project.org>  2006-08-20  Douglas Bates  <bates@R-project.org>
890    
891          * src/sparseQR.c: Fix thinko in sparseQR_resid_fitted.          * src/sparseQR.c: Fix thinko in sparseQR_resid_fitted.

Legend:
Removed from v.1403  
changed lines
  Added in v.1852

root@r-forge.r-project.org
ViewVC Help
Powered by ViewVC 1.0.0  
Thanks to:
Vienna University of Economics and Business Powered By FusionForge