Skip to content

Latest commit

 

History

History
229 lines (190 loc) · 8.72 KB

NEWS.md

File metadata and controls

229 lines (190 loc) · 8.72 KB

Version 2.4.0.3.9000 (2021-12-07)

  • latexTable() now handles tibbles (#60)
  • latexTable() now automatically escapes % signs in colNames; unescaped % signs prevent LaTeX from compiling the .tex file produced by latexTablePDF().
  • latexTable() recognizes NaN as missing data when replacing missing data with NA_text.
  • latexTable() gives an informative error message when colNames and colNameExpand are used inappropriately.
  • latexTable() now throws an error when any element in colNames contains a newline character.
  • rescale() now checks to ensure that x contains some non-missing data  
     
     

Version 2.4.0.3 (2021-11-20)

  • Significant improvements to output of reliability().  
     
     

Version 2.4.0.2 (2021-07-14)

  • Add suppress_warnings().
  • Fix small problem with makePercentage() documentation.  
     
     

Version 2.4.0.1 (2021-03-08)

  • Fix lsos() and add tests for it (#61).
  • Add makePercentage().  
     
     

Version 2.4.0 (2021-02-21)

New functions

  • allNA (#48), minNA() and maxNA().
  • cutYears() (#50),
  • printAll() to print all rows of a tibble.
  • sourcing() (#49)
  • spaceToCamel().

Other changes

  • Missing-value functions (lNA() and kin) now throw an error when x is NULL. For example, if a user runs lNA(foo$x) but x is not an element of foo, the function will throw an error message (#45). It previously returned 0.

  • Change default of firstPageEmpty in latexTablePDF() from TRUE to container; that is, firstPageEmpty will by default be TRUE if container is TRUE, and FALSE if it is FALSE.

  • Improve ability of latexTable() to handle lists that contain functions, e.g., lt_rSquaredRow.

  • In latexTable(), check that mat has class matrix to avoid an obscure error message when it doesn't (#34).

  • PDF_crop() now throws an error when new cropped PDF isn't created (#42).

  • Change default lsos() output so that object size is reported in MB (#47).

  • Add more tests of latexTable().

  • Fix test of PDF_crop() to handle cases in which Sys.getenv("R_SESSION_TMPDIR") is an empty string.

  • Deprecated mergeFac() in favor of dplyr::coalesce() (#46).  
     

  • Added, to the "R Markdown" vignette, instructions for using latexTable() output in a Sweave / Rnw document.

  • Small improvements to latexTable() documentation.

  • Roxygenized documentation for reliability() and NAMESPACE. All documentation is now roxygenized (#7).
     
     
     

Version 2.3.0 (2020-07-16)

  • Added strip0() (#41).
  • Added snakeToCamel(), medianNA(), rangeNA() (#32).
  • PDF_crop() can now take paths that have spaces.
  • Reduced test_regTable.R reliance on multiwayvcov package (#36).
  • Fix bug with latexTable() when mat is a tibble.
  • Minor update to qw() documentation (#37).  
     
     

Version 2.2.2 (2020-03-22)

Updated latexTable() and regTable() for compatibility with R 4.0.0.  
 
 

Version 2.2.1 (2020-03-20)

  • Added vignette on using latexTable() with R Markdown documents (#20).
  • Updated latexTable() and latexTable() PDF documentation on required LaTeX packages. Added check for the float package to latexTablePDF().
  • Moved knitr and rmarkdown from "Suggests" to "Imports" to prevent package installation errors when build_vignettes = TRUE  
     
     

Version 2.2.0 (2020-02-24)

regTable() changes

  • Removed class restrictions on objects passed to regTable(). regTable() now works with almost every kind of regression object.
  • Added [.regTable method for intelligent subsetting of regTable objects (#27).
  • Added global option Bullock.print.regTable.dp on package load so that users can change default number of digits that are displayed when they print a regTable object.
  • Added str.regTable().
  • In print.regTable(), use "crayon" package, when available, to render column-pair headings in bold (#23).

Other enhancements

  • Added PDF_crop().
  • latexTablePDF() is now able to write PDF files even when the latexTable objects that it is processing were created with latexTable(..., callCommand = FALSE).
  • Added envir argument to lsos() and .ls.objects() (#6).
  • Improved vignette formatting (#19).

Bug fix

  • Fixed bug in latexTable footers when SE_table is FALSE (#24).  
     
     

Version 2.1.0 (2020-01-21)

regTable() changes

  • Added support for glm objects to regTable().
  • Allowed regTable() to work with regressions of different classes (i.e., "lm" and "ivreg") when clustering SEs.
  • Improved print.regTable() output (#11, #18).
  • Improved checking of classes of objList elements.

latexTable() changes

  • In latexTable(), changed spaceBetweenColNameRows argument from TRUE to -.025in. Users can now arbitrarily adjust the vertical space between column-name rows by specifying any LaTeX length as the spaceBetweenColNameRows argument.
  • Improved latexTable() output when formatNumbers is FALSE.
  • Added c.latexTable() method (#12).
  • Better latexTable() warning message when colNames[[1]] has the wrong length.

Other enhancements

  • Reduced lNAv() output from two lines to 1 (#16).
  • If one goes to help.start() and then clicks on the "Bullock" entry, "Bullock-package" is now the first entry in the help index.
  • Added examples to documentation of stack utilities (#5) and missing-value utilities.
  • Small improvements to documentation formatting.

Bug fixes

  • Fixed failure to export print.regTable(). Printing of "regTable" objects
    to screen will now look better. Also updated vignette with new regTable output (#21).
  • Fixed latexTable() calls for which decimalPlaces is 0 (#8).
  • Fixed bug in latexTable() calls that arose when SE_table was FALSE and an element of the colNames list had a length less than ncol(mat).
  • Fixed float placement of landscaped tables made by latexTable().  
     
     

Version 2.0.0 (2020-01-08)

  • Added regTable(), latexTable(), and latexTablePDF().
  • Modified behavior of lNA() so that it no longer prints any output by default. The new lNAv() behaves like the previous version of lNA().
  • Adopted camelCase for function names and arguments throughout the package. modal_value() is now modalValue(), and so on.
  • Rewrote moveToDF() (formerly move.to.df()) from scratch.
  • Rewrote the stack utilities: pop(), push(), shift(), and unshift().
  • Removed factorToDummyMatrix(). See caret::class2ind or makedummies::makedummies for a replacement.
  • Removed ancient functions: latable(), noNAmatrix(), split_fac().
  • Converted to roxygen for documentation of most functions.  
     
     

Pre-release versions.

  • Version 1.19.1. Updated contact email address in DESCRIPTION.
  • Version 1.19. Minor update to modal_value, which seems to make it faster.
  • Version 1.18. Minor update to rescale() so that it works with tibble columns.
  • Version 1.17. Added factorToDummyMatrix().
  • Version 1.16. Minor update to qw().
  • Version 1.15. Substantial revisions to move.to.df().
  • Version 1.14. Added %IN%, a binary match operator that can return NA.
  • Version 1.13. Added qw(), a Perl-like string-quoting function.
  • Version 1.12. Changed default argument to rescale().
  • Version 1.11. Improved lsos() documentation.
  • Version 1.10. Added lsos() to the package.
  • Version 1.09. Adjusted merge_fac() so that it can work with variables in environments other than the global environment.
  • Version 1.08. Added stack utilities (push(), pop(), shift(), and unshift()) that are based on Jeffrey A. Ryan's code at [http://www.lemnica.com/esotericR/Introducing-Closures/].
  • Version 1.07. Fixed bug in latable() handling of ivreg() output. (The AER package contains an ivreg() function for instrumental-variables regression.)
  • Version 1.06. Added rows.to.remove argument to latable() so that latable() can return "incomplete" regression tables.
  • Version 1.05. Added NAMESPACE to make the Windows binary installable in R 2.14.
  • Version 1.04. Added reliability() to compute Cronbach's alpha for batteries of survey items or other measurements.
  • Version 1.03. Added modal_value() and fixed a related bug in move.to.df().
  • Version 1.02. Added merge_fac().