1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
|
Version 0.39
o Improved detection of sufficiency, in function recreate()
o Fixed printing bug in objects of class "admisc_fobject", when expressions
are evaluated using the "split.by" argument for a single variable
o Function recode() now correctly preserves the labels for the declared
missing values, if existing in the input data
o If the input is a declared object, function recode() now allows providing
a variable label for the resulting declared object, using the argument
"label" (see also the argument "label" in function declared() from the
package declared)
o Function recode() now discards the labels for declared missing values,
if no such declared values are found in the input data, for instance
because of drop_na()
o Bug fix in function recode(), the number of labels is now checked to be
equal to the number of recodings
o New class method for matrices (coerced to data frames) in function using()
Version 0.38
o Bug fix in function recode() treating NA values using the argument cut
o Improved function using() treating split.by argument and result printing
o Function expand() now returns the snames as an attribute, if provided,
and function translate() now detects it in the expression's attributes
o Function writePrimeimp() renamed to writePIs()
o Function negate() renamed to invert()
o Former function invert() renamed to sopos()
o Function finvert() renamed to frev()
Version 0.37
o Better output for function using() with a vector of expressions
o Function export() is now generic, allowing for class extensions
o New S3 class extension for function update() to update a file
(for the moment, specific to package DDIwR updating a DDI Codebook)
Version 0.36
o Function recreate() now captures a tilde for global objects
o Improved functions recode() and getInfo() for objects of class "declared"
o New function betweenQuotes()
o Function insideBrackets() is now deprecated, replaced by betweenBrackets()
o Bug fixes in possibleNumeric(), where diacritics are detected as
multi-byte characters, or when x has length 1
o Bug fixes in strsplit() and replaceText(), to avoid infinite loops
calling each other
Version 0.35
o Fixed recode() for the more recent treatment that c() is NULL
o Improved function change() with respect to QCA truth tables
o Function recreate() now recognizes "-.>" as a sufficiency operator
o Small code improvements
Version 0.34
o New function overwrite()
o New function change()
o Improved version of inside(), where now the argument "data" can be
anything (including a list component)
Version 0.33
o Minor changes to the internal function getInfo()
o Employed hexadecimal representation for replacing special characters
o More integration with the companion package QCA
Version 0.32
o New functions setColnames(), setRownames() and setDimnames()
o Bug fix in using(), when the split variable has missing range values
Version 0.31
o New function inside(), as an alternative to the base function within()
o New function scan.clipboard()
o New argument "protect" in function replaceText()
o Function using() is now generic, with exactly the same default
functionality as the base function with()
Version 0.30
o Improved treatment of multi-byte space characters in functions
possibleNumeric() and asNumeric()
o Function using() now accepts all types of variables for the "split.by"
argument, that can be coerced to factors
Version 0.29
o Functions asNumeric() and recode() are now generic, with class methods for
factors and objects of class "declared"
o New arguments "na_values" (for declared objects) in function recode()
o Improved function getName() for more than one variable
o New argument "object" in function getName()
Version 0.28
o Bug fix in asNumeric() preserving classes for some types of objects
o New arguments "levels" and "na_values" in function asNumeric()
Version 0.27
o New argument "maxdec" in function numdec()
o Correct way of checking the package QCA version for simplify()
o More robust way to calculate expressions even when a condition is
numeric, but of character mode
Version 0.26
o New function numdec() to count the number of decimals in a possibly
numeric value
o Improved treatment of the "split.by" argument in function using()
o Rewritten print method for resulting objects from function using(), now
of a more general class "admisc_fobject"
o Printing numerical vectors of class "admisc_fobject" are now automatically
rounded to maximum three decimals
Version 0.25
o Bug fix in function using(), function names were sometimes misinterpreted
as column names in the data
o Functions obj.rda() and list.rda() renamed to objRDA() and listRDA()
o Dropped functions obj.rdata() and list.rdata()
Version 0.24
o Fixed issue with too large whole numbers to be coerced to integers
(thanks to Sarah Goslee for the report)
o Function wholeNumeric() now returns FALSE for characters, instead of NA
Version 0.23
o Bug fix in mvSOP(), for situations when some conditions are not present
in the data
o Bug fix in compute(), avoiding situations when mvSOP() fails
o Improved possibleNumeric() and wholeNumeric() for logical vectors
o New argument "bincat" for equality check functions
(thanks to Brice Richard for the suggestion)
Version 0.22
o Improved function tryCatchWEM(), now also returning the actual output value
(thanks to John Fox for the suggestion)
o New argument "regexp" to all brackets functions, extending functionality
for any general purpose (thanks to Brice Richard for the suggestion)
o New function using(), allowing to evaluate an expression in every subset
of a split file
o New function hclr(), to produce colors from the HCL spectrum
o New function coerceMode(), to coerce objects to numeric or integer, if at
all possible
Version 0.21
o Bug fix in function negate(), expressions were not properly concatenated
(thanks to Alessandra Costa for the report)
Version 0.20
o New argument "each" in functions possibleNumeric() and wholeNumeric()
Version 0.19
o New function asSOP(), to coerce a POS expression to a standard SOP format
o New function mvSOP(), to coerce an expression from crisp set notation to
multi-value notation
Version 0.18
o Fixed bug affecting the function negate() when the SOP expression contains
a single condition in one of the conjuncts / products
(thanks to Michael Baumgartner for the report)
o Fixed bug in asNumeric() preventing certain character objects of class
"haven_labelled" to be converted as numeric
o Improved function stopError(), printing error messages containing newline
characters
Version 0.17
o More robust support for multi-byte locales when detecting tilde and
dash operators
o Fixed bug in possibleNumeric() for objects of class declared
Version 0.16
o New functions agtb(), altb() and aneqb() to test (in)equality of floats
o New utility function getName() to return the name of the object being
used in a function call
o Fixed bug when recoding objects of class "declared"
o Fixed bug detecting multibyte strings
Version 0.15
o possibleNumeric() and asNumeric() are now more robust in situations
with invalid multibyte strings
o Argument "cuts" renamed to "cut" in function recode()
o Fixed bug in function recode() that prevented creating ordered
factors
Version 0.12
o Solved bug in function translate() when called from plumber or
callr (thanks to Trevor Strobel for the report)
o Solved bug in dealing with expressions containing brackets with
single letter conditions and no star signs to indicate conjunctions
o New functions list.rda() and names.rda()
o Small improvement of the recode() function
Version 0.11
o New function finvert(), to invert a factor's values (and optional
its levels)
o New function frelevel(), an improved version of the base relevel()
o New function permutations()
o Improved version of combnk(), to cover input vectors of any type
o Improved error trapping for functions negate() and simplify, when
dealing with multivalue expressions
Version 0.10
o Minor, internal functionality changes
Version 0.9
o Solved bug in translate() recognizing column names for datasets with more
than 27 columns (thanks to Sophia Birchinger for the report)
o New function export(), moved here from package QCA
Version 0.8
o Extended functionality to other types of vectors, such as having the
class "haven_labelled"
o Novel way of recognizing SOP expressions, even without quotes
o New utility function recreate() to facilitate substitution
Version 0.7
o Minor modification in function simplify(), to avoid the check error
from the CRAN servers for the OS X platform
Version 0.6
o Major modification (and *not* backwards compatible!) with respect to
denoting negations. Using upper and lower case letters for presence and
absence is no longer supported, a tilde being the only and the default
method to signal a negation (thanks to Charles Ragin for making the point)
o Removed deprecated argument "use.tilde" from all related functions
o All functions treating a DNF/SOP expression now obey this major
(and not backwards compatible) change denoting a negation. Upper
and lower case conditions are no longer supported
o All printing classes are now prefixed with "admisc", to avoid possible
namespace collisions with (previous) versions of package QCA
o New function invert() to convert a SOP expression to a POS expression
(thanks to Charles Ragin for the suggestion)
o New function expand() to perform a full or a partial Quine expansion
to a SOP expression
Version 0.5
o Functions compute(), factorize(), intersection(), negate()
and simplify(), moved here from package QCA
o New function replaceText()
o Minor changes to internal functions getInfo() and getLevels()
o Improved function translate() using replaceText(), now better
suited in dealing with set names of variable number of
characters, including space
o As a result, argument "snames" from function venn() can deal with
spaces in set names (thanks to Andre Gohr for the suggestion)
Version 0.4
o Fixed small printing bug in possibleNumeric()
o Function translate() is now more robust against non-printable
characters
Version 0.3
o Function combinations() renamed to combnk()
o Improved function possibleNumeric() to deal with objects of
class "haven_labelled"
Version 0.2
o Function combinations() renamed to combnk()
Version 0.1
o Start of the package
|
