Recent Releases of expss
expss - 0.11.1
0.11.1 (07.01.2022)
- add 'weight_by' - function for "brute force" frequency weighting
- 'readspss' now exposes 'usemissings' argument from foreign::read.spss
- 'apply_labels' now can accept list with variable and value labels
- 'calccro*'s are renamed to more clear 'cross_'s
- 'compute', 'do_if', 'where' and 'calc' are deprecated in favor of the 'maditr'
package. 'maditr' is thin layer above fastest data manipulation package 'data.tabe'
- R
Published by gdemin over 4 years ago
expss - 0.10.7
0.10.7 (15.11.2020)
- fix compatibility with SPSS in 'w_median' function (issue #72)
- fix bug with incorrect labelling in some cases in 'recode' (issue #73)
- 'default_dataset' functionality is deprecated
- 'apply_labels' now respects data.table (and assign labels by reference for data.table)
- 'not', 'or', 'and', 'when' now respects criteria functions
- R
Published by gdemin over 5 years ago
expss - 0.10.2
0.10.2 (25.03.2020)
- fixes for R4.0
- new vignette with examples
- add functions 'hide' and 'unhide' for selective hiding and showing categories in 'net' and 'subtotal'
- now logical don't have special meaning in the 'recode'. If you use this functionality than use 'when' instead. recode(y, x>5 ~ 1) need to be replaced with recode(y, when(x>5) ~ 1).
- add 'xlwritefile' for quick export tables to Excel file
- Add 'tabcaption' - alias for 'setcaption'
- R
Published by gdemin about 6 years ago
expss - 0.9.1
0.9.1 (26.08.2019)
- 'subtotal' and 'net' produce consistent ordering of subototal items (thanks Roberto Gilsaura for extensive testing)
- 'subtotal' and 'net' allow overlapping categories
- fix vignette for new 'reformulate'
- fix recode for list recodings. Now c(1, 5 %thru% 99) ~ 999 works properly.
- R
Published by gdemin almost 7 years ago
expss - 0.9.0
0.9.0 (01.07.2019)
- add 'subtotal' and 'net' functions. See '?net'.
- 'recode' now can recode value labels to keep them consistent with values (see argument 'with_labels')
- add alias 'rec' to recode with argument 'with_labels = TRUE' by default
- now you can assign new labels in 'recode' ("TOP" = 4:5 ~ 5)
- improve 'recode' performance
- simplify 'recode' behavior. Now RHS and LHS part in the recoding formula should be single column object. However, you can still recode entire data.frame or list.
- improve preformance of 'sumrow', 'meanrow', 'median_row' and etc.
- change labelled 'csv' format. Now labels are stored in the same file as data. Labels are located in the commented lines at the begining of the file. Ispite of changing format, 'readlabelledcsv' should detect and read the old format automatically.
- add new functions: 'readlabelledxlsx' and 'writelabelledxlsx'. Now you can save and restore labelled data to Excel file. Labels are stored on the separate sheet.
- add new functions: 'readlabelledfst' and 'writelabelledfst' for storing data in the 'fst' format (https://www.fstpackage.org/). With this format you can read and write a huge amount of labelled data very quickly.
- Small bugfixes.
- R
Published by gdemin almost 7 years ago
expss - v0.8.11
0.8.11 (16.04.2019)
- UPDATE IS REQUIRED. fixes for R>=3.6 and data.table >=1.12.2.
- now htmlTable respects 'align' argument (Issue #20)
- fix 'addlabelledclass' for 'haven' 2.0
- improve SPSS syntax in 'writelabelledspss'
- add new significance test - 'significancecellchisq'/'tablastsigcellchisq'. Idea by Roberto Gilsaura.
- now 'significance_cases' has new argument 'correct' for Yates' correction. By default it is TRUE
- add 'qe' function for making list of expressions
- ..$ selector now supports arbitrary expression
- fix bug in 'cromeansd_n with incorrect weighted valid n in some cases
- Numerous small bugfixes
- R
Published by gdemin about 7 years ago
expss - v0.8.8
0.8.8 (11.11.2018)
- add '*.xlsx' export for tables via 'openxlsx' package. See ?xl_write
- add 'setcaption' function and support for it in 'htmlTable', 'as.datatablewidget' and *.xlsx export
- add function 'expssenablevaluelabelssupport_extreme'
- small bugfixes and performance improvement
- 'fromtext' renamed to 'textto_columns'
- 'by_groups' is deprecated now. Use 'take' from 'maditr' instead.
- 'dtfrm'/'as.dtfrm' are finally removed. Use 'sheet'/'as.sheet' instead
- It seems nobody uses 'lst'/'.lst' functions so they are removed.
- R
Published by gdemin over 7 years ago
expss - v0.8.7
0.8.7 (12.06.2018)
- fix long-standing bug with broken html rendering in knitr (issue #6)
- add '..p' and '..f' - perl regular expression/fixed pattern selectors. Also 'mrsetp', 'mrsetf', 'mdsetp', 'mdsetf' are provided.
- add ..data shortcut for entire data.frame in 'use_labels'
- fix awful bug with incorrect nesting multiple response on multiple response ('nest' function)
- more exports from data.table
- fix 'add_columns' for data.table
- fix 'do_if' for data.table
- fix 'w_cor' - now diag elements always equal to one
- add 'splitby' - alias for 'splitseparate'
- R
Published by gdemin almost 8 years ago
expss - v0.8.6
0.8.6 (24.01.2018)
- fix tests for systems with no long doubles support
- new arguments and small improvements in 'writelabelledcsv'/'writelabelledsps'
- minor fixes
- change 'vlookup'/'vlookup_df' behavior - now it also matches on NA
- 'datatable' is deprecated. Use as.datatable_widget instead.
- move DT and htmltools dependencies from Imports to Suggests
- new function 'add_columns' which is inspired by MATCH FILES (Add variables...) from SPSS Statistics
- R
Published by gdemin over 8 years ago
expss - v0.8.2
0.8.2 (26.09.2017)
- fix bug with single column "mrset"
- fix bug with significance tests on tables with complex headers
- add uboxing possibilities for ..
- add 'from_text' for creation data.frame from text lines
- improve performance of 'cro_*'
- remove deprecated function ".set"
- R
Published by gdemin over 8 years ago
expss - v0.8.1
0.8.1 (10.08.2017)
- add significance testing (see ?significance)
- add support for table output for Jupyter notebooks
- add 'uselabels' function for experimental support variable labels in base R functions and third party packages (see ?uselabels)
- increased table functions performance
- increased 'as.dichotomy'/'dummy' performance
- increased 'vlookup'/'vlookup_df' performance
- numerous bugfixes
- now 'keep'/'except'/'by_groups' operate with NSE. For standard evaluation just surround you variable with round brackets
- add print options 'commented' for printing tables to console with '#'
- add functions 'splitseparate'/'splitoff'
- add 'prependnames'/'prependlabels' - issue #2
- add 'indirect'/'indirectlist' - aliases for 'vars'/'varslist'
- add 'misval' - alias for 'naif'
- add 'unsafe' argument for _fun/fundf.
- add '..' object for parameter substitution inside 'compute' and etc. Usage: '..$param_name'. For details: ?..
- '.set'/'set' inside 'compute' and etc. are deprecated. Use '%into%' instead.
- expansion of variables in backticks inside 'vars'/'vars_list'/'%into%' is removed. Use explicit 'subst' instead
- infix/assignment versions removed: '%modify%', '%compute%', '%sortasc%', '%sortdesc%', '%keep%', '%except%', '%where%', '%by_groups%'
- R
Published by gdemin almost 9 years ago
expss - v0.7.1
0.7.1 (10.04.2017)
- make value labels support in the base R and other packages (via 'factor' for class 'labelled')
- radically improved performace for 'cro*' and 'fre*' due to 'data.table' under the hood
- multiple/nested banners/variables suppport in 'cro_*'
- add 'datatable' for tables representation in Shiny
- add 'htmlTables' for tables representation in RStudio viewer and for knitting to html
- add methods for
wherefor vectors/matrices/lists - less cryptic message about bad number of rows in
modify/modify_if - .N now is preferable to .n inside
modifyand etc. - Functions for default dataset now invisibly return modified default dataset
- Fix logical arguments in boolean operations for criteria functions
%in_row%/%in_col%are deprecated. Use%row_in%/%col_in%instead.- NULL argument for criterion is removed. Use
not_nainstead. - remove
defaultargument inifs. Use TRUE ~ default_value instead. - add usual vectors functions additionally to infix versions.
- add 'do_repeat' function
- add functions for weighted statistics - 'wmean', 'wmedian', 'w_cor' and etc.
- add 'nest' function for creation nested tables.
- remove 'category', 'category_df' - use 'as.category' instead
- remove 'dichotomy', 'dichotomy1' - use 'dummy', 'dummy1' instead
- remove 'dichotomydf', 'dichotomydf1' - use 'as.dichotomy' instead
- remove labelled matrix support
- bugfixes
- R
Published by gdemin about 9 years ago
expss - v0.5.5
0.5.5 (19.10.2016)
- numerous bugfixes
- change behavior of
if_val()- now it dowsn't copies old values by default. There are 'other' and 'copy' instead of dots. - add
%n_i%,%n_d%- names intersection and diffs - add
add_rowsfunction - add
keepandexceptfunctions for selecting/dropping columns in data.frames - add 'where' for filtering dataset
- add
sort_ascandsort_descfunctions for sorting data.frames
- R
Published by gdemin over 9 years ago