Debian Codegraph @

NEWS

raw
1 Changes in 3.0.1 (2024-10-22)
2 -----------------------------
3
4 Other changes:
5
6 - Changed ll() so it checks that all elements of a list-like object are uniquely
7 named.
8
9
10
11
12 Changes in 3.0.0 (2023-10-09)
13 -----------------------------
14
15 Other changes:
16
17 - Removed installXLSXsupport(), read.xls(), sheetCount(), sheetNames(),
18 xls2csv(), xls2tab(), xls2tsv(), xls2sep(), and xlsFormats(). See gdata 2.19.0
19 NEWS entry announcing their deprecation.
20
21 - Added aliases n_obs(), object_size(), and starts_with(). Deprecated
22 gdata::object.size().
23
24
25
26
27 Changes in 2.19.0 (2023-05-06)
28 ------------------------------
29
30 New functions:
31
32 - Added update.data.frame() to replace rows of a data frame.
33
34 New features:
35
36 - Added argument 'add.col.nums' to left() and right().
37
38 Other changes:
39
40 - Improved ll() so it handles S4 objects. Also accepts unit="GB" (contributed by
41 Michael Chirico) and unit="B", now case-insensitive.
42
43 - Improved right() so it returns the rightmost columns of a matrix or data
44 frame.
45
46 - Renamed first two arguments of startsWith() to 'x' and 'prefix'.
47
48 - Removed elem() - was deprecated in gregmisc/gdata 2.0.3 (2005-02-25).
49
50 - Excel file support is now deprecated, to be removed in the near future. At the
51 user level, the functions installXLSXsupport(), read.xls(), sheetCount(),
52 sheetNames(), xls2csv(), xls2sep(), xls2tab(), xls2tsv(), and xlsFormats() now
53 raise a warning about the functionality being deprecated. Excel file support
54 was first introduced in gregmisc/gdata 1.11.0 back in 2004, but today we have
55 packages such as openxlsx, readxl, XLConnect, and xlsx offering dedicated
56 Excel file support with more features. When Excel file support will be removed
57 from the gdata package, it will result in lighter package maintenance and no
58 more Perl warnings on the user side.
59
60 - Package maintainer is Arni Magnusson.
61
62
63
64
65 Changes in 2.18.0.1 (2022-05-10)
66 --------------------------------
67
68 Other changes:
69
70 - Syntax improvements to pass CRAN checks, packaged by Brian Ripley.
71
72
73
74
75 Changes in 2.18.0 (2017-06-05)
76 ------------------------------
77
78 New functions:
79
80 - Add update.list() function to replace named elements a list.
81
82 - Add mv() function to rename an object.
83
84 - Add first(), last(), first<-() and last<-() functions to extract
85 or replace first or last vector/list elements.
86
87 New features:
88
89 - Add 'byrow' argument to lowerTriangle() and upperTriangle() functions.
90
91 Other changes:
92
93 - humanReadable() now properly handles a single value for the argument
94 'justify'.
95
96 - Improve logging and error reporting for remove.vars().
97
98 - read.xls() now handles latin-1 files properly on MS Windows.
99
100 - write.fwf() now properly handles matrix arguments.
101
102
103
104
105 Changes in 2.17.0 (2015-07-02)
106 ------------------------------
107
108 New features:
109
110 - Add new argument 'byrow' to upperTriangle(), lowerTriangle(),
111 upperTriangle<-(), and lowerTriangle<-() to specify by row rather than
112 by column order. This makes it simpler to copy values between the
113 lower and upper triangular areas, e.g. to construct a symmetric
114 matrix.
115
116 Other changes:
117
118 - Add inline comments to tests to alert reviewers of expected diffs on
119 systems lacking the libraries for read.xls() to support XLSX
120 formatted files.
121
122
123
124
125 Changes in 2.16.1 (2015-04-28)
126 -----------------------------
127
128 Bug fixes:
129
130 - mapLevels() no longer generates warnings about conversion of lists to vectors.
131
132 Other changes:
133
134 - Requirement for Perl version 5.10.0 or later is specified in the
135 package DESCRITION.
136
137 - first() and last() are now simply wrappers for calls to 'head(x, n=1)' and
138 'tail(x, n=1)', respectively.
139
140
141
142
143 Changes in 2.16.0 (2015-04-25)
144 ------------------------------
145
146 New features:
147
148 - New functions first() and last() to return the first or last element
149 of an object.
150
151 - New functions left() and right() to return the leftmost or
152 rightmost n (default to 6) columns of a matrix or data frame.
153
154 - New 'scientific' argument to write.fwf(). Set 'scientific=FALSE' to
155 prevent numeric columns from being displayed using scientific
156 notification.
157
158 - The 'standard' argument to humanReadable() now accepts three values,
159 'SI' for base 1000 ('MB'), 'IEC' for base 1024 ('MiB'), and 'Unix'
160 for base 1024 and single-character units ('M')
161
162 - object.size() now returns objects with S3 class 'object_sizes' (note the
163 final 's') to avoid conflicts with methods in utils for class 'object_size'
164 provided by package 'utils' which can only handle a scalar size.
165
166 - New 'units' argument to humanReadable()--and hence to
167 print.object_sizes() and format.object_sizes()--that permits
168 specifying the unit to use for all values. Use 'bytes' to display
169 all values with the unit 'bytes', use 'auto' (or leave it missing) to
170 automatically select the best unit, and use a unit from the selected
171 standard to use that unit (i.e. 'MiB').
172
173 - The default arguments to humanReadable() have changed. The defaults
174 are now 'width=NULL' and 'digits=1', so that the default behavior is
175 now to show digit after the decimal for all values.
176
177 Bug fixes:
178
179 - reorder.factor() was ignoring the argument 'X' unless 'FUN' was
180 supplied, making it incompatible with the behavior of
181 stats:::reorder.default(). This has been corrected, so that calling
182 reorder on a factor with arguments 'X' and/or 'FUN' should now
183 return the same results whether gdata is loaded or not. (Reported
184 by Sam Hunter.)
185
186 - write.fwf() now properly supports matrix objects, including matrix
187 objects without column names. (Reported by Carl Witthoft.)
188
189 Other changes:
190
191 - Replaced deprecated PERL function POSIX::isdigit in xls2csv.pl
192 (which is used by read.xls() ) with an equivalent regular
193 expression. (Reported by both Charles Plessy, Gerrit-jan Schutten,
194 and Paul Johnson. Charles also provided a patch to correct the
195 issue.)
196
197 - aggregate.table(), which has been defunct gdata 2.13.3 (2014-04-04)
198 has now been completely removed.
199
200
201
202
203 Changes in 2.14.0 (2014-08-27)
204 ------------------------------
205
206 Bug Fixes:
207
208 - read.xls() can now properly process XLSX files with up to 16385 columns (the
209 maximum generated by Microsoft Excel).
210
211 - read.xls() now properly handles XLS/XLSX files that use 1904-01-01
212 as the reference value for dates instead of 1900-01-01 (the
213 default for MS-Excel files created on the Mac).
214
215 Other changes:
216
217 - Updated perl libraries and code underlying read.xls() to the latest
218 version, including switching from Spreadsheet::XLSX to
219 Spreadsheet::ParseXLSX.
220
221
222
223
224 Changes in 2.13.3 (2014-04-04)
225 ------------------------------
226
227 Bug Fixes:
228
229 - Unit tests were incorrectly checking for equality of optional POSIXlt
230 components. (Bug reported by Brian Ripley).
231
232 Other Changes:
233
234 - 'aggregate.table' is now defunct. See '?gdata-defunct' for details.
235
236 - Unit tests and vignettes now follow R standard practice.
237
238 - Minor changes to clean up R CMD check warnings.
239
240
241
242
243 Changes in 2.13.2 (2013-06-28)
244 ------------------------------
245
246 Enhancements:
247
248 - Simplify ll() by converting a passed list to an environment,
249 avoiding the need for special casing and the use of attach/detach.
250
251 - Working of deprecation warning message in aggregate.table clarified.
252
253
254
255
256 Changes in 2.13.1 (2013-03-24)
257 ------------------------------
258
259 Enhancements:
260
261 - Replaced calls to depreciated function ".path.package" with the new
262 public function "path.package".
263
264
265
266
267 Changes in 2.13.0 (2012-09-20)
268 -----------------------------
269
270 New features:
271
272 - New 'duplicated2' function which returns TRUE for *all* elements
273 that are duplicated, including the first, contributed by Liviu
274 Andronic. This differs from 'duplicated', which only returns the
275 second and following (second-to last and previous when
276 'fromLast=TRUE') duplicate elements.
277
278 - New 'ans' functon to return the value of the last evaluated
279 top-level function (a convenience function for accessing
280 .Last.value), contributed by Liviu Andonic.
281
282 Bug Fixes:
283
284 - On windows, warning messages printed to stdout by perl were being
285 included in the return value from 'system', resulting in errors in
286 'sheetCount' and 'sheetNames'. Corrected.
287
288 - The 'MedUnits' column names 'SIUnits' and 'ConventionalUnits' were
289 reversed and misspelled.
290
291
292
293
294 Changes in 2.12.0 (2012-09-12)
295 ------------------------------
296
297 Other Changes:
298
299 - 'stats::aggregate' was made into a generic on 27-Jan-2010, so that
300 attempting to call 'aggregate' on a 'table' object will now
301 incorrectly call 'aggregate.table'. Since 'aggregate.table' can be
302 replaced by a call to tapply using two index vectors, e.g.
303 aggregate.table(x, by1=a, by2=b, mean)
304 can be replaced by
305 tapply(x, INDEX=list(a, b), FUN=mean),
306 the 'aggregate.table' function will now display a warning that it
307 is depreciated and recommending the equivalent call to tapply. It
308 will be removed entirely in a future version of gdata.
309
310
311
312
313 Changes in 2.11.1 (2012-08-22)
314 ------------------------------
315
316 Enhancements:
317
318 - read.xls() now supports fileEncoding argument to allow non-ascii
319 encoded data to be handled. See the manual page for an example.
320
321 Bug Fixes:
322
323 - The perl script utilized by read.xls() was incorrectly appending a
324 space character at the end of each line, causing problems with
325 character and NA entries in the final column.
326
327
328
329
330 Changes in 2.11.0 (2012-06-18)
331 ------------------------------
332
333 New Features:
334
335 - read.xls() and supporting functions now allow blank lines to be
336 preserved, rather than skipped, by supplying the argument
337 "blank.lines.skip=FALSE". The underlying perl function has been
338 extended to suppor this via an optional "-s" argument which, when
339 present, *preserves* blank lines during the conversion. (The
340 default behavior remains unchanged.)
341
342 Other Changes:
343
344 - Add SystemRequirements field specifying that perl is necessary for
345 gdata to function fully.
346
347
348
349
350 Changes in 2.10.6 (2012-06-12)
351 ------------------------------
352
353 Bug fixes:
354
355 - gdata::nobs.default() needs to handle logical vectors in addition to
356 numeric vectors.
357
358
359
360
361 Changes in 2.10.5 (2012-06-08)
362 ------------------------------
363
364 Bug fixes:
365
366 - Mark example for installXLSsupport() as dontrun so R CMD check won't
367 fail on systems where PERL is not fully functional.
368
369 - Correct name of installXLSsupport() in tests/test.read.xls.R.
370
371 Other Changes:
372
373 - Add dependency on R 2.13.0, since that is when stats::nobs appeared.
374
375
376
377
378 Changes in 2.10.2 (2012-06-06)
379 ------------------------------
380
381 Bug fixes:
382
383 - Fix issues in nobs.default identified in testing with the gmodels package.
384
385
386
387
388 Changes in 2.10.1 (2012-06-06)
389 ------------------------------
390
391 Bug fixes:
392
393 - Undo removal of 'nobs' and 'nobs.lm'. Instead define aliases for
394 'nobs' and 'nobs.lm' to support backward compatibility for packages
395 depending on gdata.
396
397
398
399
400 Changes in 2.10.0 (2012-06-05)
401 ------------------------------
402
403 New features:
404
405 - New ls.funs() function to list all objects of class function in the
406 specified environment.
407
408 - New startsWith() function to determine if a string "starts with" the
409 specified characters.
410
411 Enhancements:
412
413 - Add 'na.strings' argument to read.xls() to convert Excel's '#DIV/0!' to NA.
414
415 Bug fixes:
416
417 - Correct various R CMD check warnings
418
419 Other changes:
420
421 - Base S3 method for nobs() and nobs.lm() method removed since these
422 are now provided in the stats package.
423
424
425
426
427 Changes in 2.9.0 (2011-09-30)
428 -----------------------------
429
430 New features:
431
432 - Add centerText() function to center text strings for a specified
433 width.
434
435 - Add case() function, a vectorized variant of the base::switch()
436 function, which is useful for converting numeric codes into factors.
437
438 Enhancements:
439
440 - Minor improvements to xls2csv() man page.
441
442
443
444
445 Changes in 2.8.1 (2011-04-15)
446 -----------------------------
447
448 Enhancements:
449
450 - nPairs() gains a summary method that shows how many times each variable
451 is known, while the other variable of a pair is not
452
453 Bug fixes:
454
455 - Fix errors on windows when R or Perl install path includes spaces by properly
456 quoting the path.
457
458
459
460
461 Changes in 2.8.1 (2010-11-12)
462 -----------------------------
463
464 Enhancements:
465
466 - Minor improvement to Args(), read.xls() man page.
467
468 Bug fixes:
469
470 - Modify write.fwf() to capture and pass on additional arguments for
471 write.table(). This resolves a bug reported by Jan Wijffels.
472
473 - Modify xls2sep.R to avoid use of file.access() which is unreliable
474 on Windows network shares.
475
476
477
478
479 Changes in 2.8.0 (2010-04-03)
480 -----------------------------
481
482 Enhancements:
483
484 - When loaded, gdata (via an .onAttach() function) now checks:
485 1) if perl is available
486 2) whether the perl libraries for XLS support are available
487 3) whether the perl libraries for XLSX support are available
488
489 If perl is not available, an appropriate warning message is displayed.
490
491 If necessary perl libraries are not available, a warning message is
492 displayed, as is a message suggesting the user run the (new)
493 installXLSXsupport() function to attempt to install the necessary
494 perl libraries.
495
496 - The function installXLSXsupport() has been provided to install the
497 binary perl modules that read.xls needs to support Excel 2007+
498 'XLSX' files.
499
500
501
502
503 Changes in 2.7.3 (2010-04-02)
504 -----------------------------
505
506 Enhancements:
507
508 - New xlsFormats() command to determine which Excel formats are
509 supported (XLS, XLSX).
510
511 Bug Fixes:
512
513 - No longer attempt to install perl modules Compress::Raw::Zlib and
514 Spreadsheet::XLSX at build/compile time. This should resolve recent
515 build issues, particularly on Windows.
516
517 - All perl code can now operate (but generate warnings) when perl modules
518 Compress::Raw::Zlib and Spreadsheet::XLSX when are not installed.
519
520 - Also update Greg's email address.
521
522
523
524
525 Changes in 2.7.1 (2010-02-19)
526 -----------------------------
527
528 Enhancements:
529
530 - on Windows attempts to locate ActiveState perl if perl= not specified and
531 Rtools perl would have otherwise been used in read.xls and other perl
532 dependent functions.
533
534
535
536
537 Changes in 2.7.0 (2010-01-25)
538 -----------------------------
539
540 Bug Fixes:
541
542 - Fix building of Perl libraries on Win32
543
544
545
546
547 Changes in 2.7.0 (2010-01-25)
548 -----------------------------
549
550 Enhancements:
551
552 - read.xls() now supports Excel 2007 'xlsx' files.
553
554 - read.xls() now allows specification of worksheet by name
555
556 - read.xls() now supports ftp URLs.
557
558 - Improved ll() so user can limit output to specified classes
559
560 New Functions:
561
562 - sheetCount() and sheetNames() to determine the number and names of
563 worksheets in an Excel file, respectively.
564
565 Bug Fixes:
566
567 - Fix formatting warning in frameApply().
568
569 - Resolve crash of "ll(.GlobalEnv)"
570
571
572
573
574 Changes in 2.6.1 (2009-07-15)
575 -----------------------------
576
577 Bug Fixes
578
579 - Modify unit tests to avoid issues related to time zones.
580
581
582
583
584 Changes in 2.6.0 (2009-07-15)
585 -----------------------------
586
587 Bug Fixes
588
589 - Correct minor typos & issues in man pages for write.fwf(),
590 resample() (Greg Warnes)
591
592 - Correct calculation of object sizes in env() and ll() (Gregor Gorjanc)
593
594 New Features
595
596 - Add support for using tab for field separator during translation
597 from xls format in read.xls (Greg Warnes)
598
599 - Enhanced function object.size that returns the size of multiple
600 objects. There is also a handy print method that can print size of
601 an object in "human readable" format when
602 options(humanReadable=TRUE)
603 or
604 print(object.size(x), humanReadable=TRUE).
605 (Gregor Gorjanc)
606
607 - New function wideByFactor that reshapes given dataset by a given
608 factor - it creates a "multivariate" data.frame. (Gregor Gorjanc)
609
610 - New function nPairs that gives the number of variable pairs in a
611 data.frame or a matrix. (Gregor Gorjanc)
612
613 - New functions getYear, getMonth, getDay, getHour, getMin, and getSec
614 for extracting the date/time parts from objects of a date/time
615 class. (Gregor Gorjanc)
616
617 - New function bindData that binds two data frames into a multivariate
618 data frame in a different way than merge. (Gregor Gorjanc)
619
620 Other Changes
621
622 - Correct Greg's email address
623
624
625
626
627 Changes in 2.5.0 (2008-12-31)
628 -----------------------------
629
630 - New function .runRUnitTestsGdata that enables run of all RUnit tests during
631 the R CMD check as well as directly from within R.
632
633 - Enhanced function object.size that returns the size of multiple objects.
634 There is also a handy print method that can print size of an object in
635 "human readable" format when options(humanReadable=TRUE) or
636 print(x, humanReadable=TRUE).
637
638 - New function bindData that binds two data frames into a multivariate data
639 frame in a different way than merge.
640
641 - New function wideByFactor that reshapes given dataset by a given factor -
642 it creates a "multivariate" data.frame.
643
644 - New functions getYear, getMonth, getDay, getHour, getMin, and getSec for
645 extracting the date/time parts from objects of a date/time class.
646
647 - New function nPairs that gives the number of variable pairs in a data.frame
648 or a matrix.
649
650 - New function trimSum that sums trimmed values.
651
652 - New function cbindX that can bind objects with different number of rows.
653
654 - write.fwf gains the width argument. The value for unknown can increase or
655 decrease the width of the columns. Additional tests and documentation fixes.
656
657
658
659
660 Changes in 2.4.2 (2008-05-11)
661 -----------------------------
662
663 - Enhancements and bug fixes for read.xls() and xls2csv():
664
665 - More informative log messages when verbose=TRUE
666
667 - File paths containing spaces or other non-traditional characters
668 are now properly handled
669
670 - Better error messages, particularly when perl fails to generate
671 an output .csv file.
672
673 - The 'shortcut' character "~" (meaning user's home directory) is
674 now properly handled in file paths.
675
676 - XLS files created by OpenOffice are now properly handled. Thanks to
677 Robert Burns for pointing out the patch
678 (http://rt.cpan.org/Public/Bug/Display.html?id=7206)
679
680
681
682
683 Changes in 2.4.1 (2008-03-24)
684 -----------------------------
685
686 - Update perl libraries needed by xls2csv() and read.xls()
687 to latest available versions on CRAN.
688
689 - Add read.xls() to exported function list
690
691 - Correct iris.xls example file. It didn't contain the complete
692 & properly formatted iris data set. Fixed.
693
694 - Fix typo in win32 example for read.xls()
695
696
697
698
699 Changes in 2.4.0 (2008-01-30)
700 -----------------------------
701
702 - The keep() function now includes an 'all' argument to specify how
703 objects with names starting with '.' are handled.
704
705 - keep() now shows an informative warning message when a requested
706 object does not exist
707
708 - New vignette "Mapping Levels of a Factor" describing the use of
709 mapLevels().
710
711 - New vignette "Working with Unknown Values" describing the use of
712 isUnknown() and unknownToNA().
713
714 - Several enhancements to read.xls() (thanks to Gabor Grothendieck):
715
716 - New function xls2csv(), which handles converting an xls file
717 to a csv file and returns a connection to the
718 temporary csv file
719
720 - xls2csv() and read.xls() both allow a file or a url to be specified
721
722 - read.xls() has a new 'pattern' argument which, if supplied,
723 will ignore everything prior to the first line in th csv file
724 that matches the pattern. This is typically used if there
725 are a variable number of comment lines prior to the header
726 in which case one can specify one of the column
727 headings as the pattern. read.xls should
728 be compatible with the old read.xls.
729
730 - Minor fixes to drop.levels(), is.what().
731
732 - Implementation of unit tests for most functions.
733
734
735
736
737 Changes in 2.3.1 (2006-10-29)
738 -----------------------------
739
740 - Arguments as well as their position of reorder.factor have been changed
741 to conform with reorder.factor method in stats package, due to collision
742 bug. Argument 'make.ordered' is now 'order' and old argument 'order' is
743 now 'new.order'! Therefore, you have to implicitly specify new.order i.e.
744
745 reorder(trt, new.order=c("PLACEBO", "300 MG", "600 MG", "1200 MG"))
746
747 - trim() gains ... argument.
748
749 - Added "unknown" methods for matrices.
750
751 - Added c() method for factors based on mapLevels() functions.
752
753 - Added write.fwf, which writes file in *F*ixed *W*idth *F*ormat.
754
755
756
757
758 Changes in 2.3.0 (2006-09-19)
759 -----------------------------
760
761 - Added mapLevels(), which produces a map with information on levels and/or
762 internal integer codes. Contributed by Gregor Gorjanc.
763
764 - Extended dropLevels() to work on the factors contained in a data
765 frame, as well as individual factors.
766
767 - Add unknown(), which changes given unknown value to NA and vice
768 versa. Contributed by Gregor Gorjanc.
769
770 - Extended trim() to handle a variety of data types data.frames,
771 lists, factors, etc. Code changes contributed by Gregor Gorjanc.
772
773 - Added resample() command that acts like sample() except that it
774 _always_ samples from the arguments provided, even if only a single
775 argument is present. This differs from sample() which behaves
776 differently in this case.
777
778 - Updated my email address.
779
780
781
782
783 Changes in 2.1.2 (2012-10-29)
784 -----------------------------
785
786 - Fixed bug in interleave.R - option to covert 1-column matrices to
787 vector (based on Andrew Burgess's suggestion)
788
789 - Updated Greg and Jim's email adresses
790
791 - ll.R: Suppressed warning message in attach() call.
792
793 - frameApply.Rd, reorder.Rd: Remove explicit loading of
794 gtools in examples, so that failure to import functions from
795 gtools gets properly caught by running the examples.
796
797 - upperTriangle.R, man/upperTriangle.Rd: Add functions for
798 extracting and modifying the upper and lower trianglular components of
799 matrices.
800
801 - is.what.R: Replaced the "not.using" vector with a more robust
802 try(get(test)) to find out whether a particular is.* function
803 returns a logical of length one.
804
805 - DESCRIPTION: Added Suggests field
806
807 - Updated the example in frameApply
808
809
810
811
812 Changes in 2.0.8 (2005-06-02)
813 -----------------------------
814
815 - Added DESCRIPTION and removed DESCRIPTION.in
816
817 - Updated ll.Rd documentation
818
819 - Fixed bug in Args.R, is.what.R, ll.R