eatGADS 1.2.0

new features

• dropDuplicateIDs() allows dropping duplicate IDs based on number of missings on selected variables (#67)
• changeValLabels() and changeMissings() allow changing value labels and missing tags for multiple variables at once (#33)
• recodeGADS() allows recoding multiple variables at once (#107)
• merge.GADSdat() allows automatically assigning NAs created through merging a missing code and a value label (#1)
• equalData() and equalMeta() now split the functionality of equalGADS()
• equalData() and equalGADS() now support multiple id variables (id)

changes

• inspectMetaDifferences() now provides consistent output for variable label differences

bug fixes

• extractData2() and extractData() no longer throw an error if multiple values of the same variable are labeled NA (#96)
• extractData2() and extractData() no longer produce an error if there are multiple duplicate value labels in a variable
• extractData2() and extractData() no longer produce a warning if multiple duplicate value labels occur which are tagged and transformed to NA anyway (#98)
• extractData2() and extractData() now provide consistent output for values which have NA as value label (#100)
• inspectMetaDifferences() now correctly reports differences for meta data with differing row names (#102)
• hot fix for existingMeta argument in recodeGADS() and applyChangeMeta(), which was ignored when values were recoded into each other (#104)
• fix unclear documentation in removeValLabels() (#111)
• fix changeMissings() bug which dropped a single, existing value label and threw an error (#121)
• checkMissingValLabels() now correctly recognizes large integers (#124)
• Checks for case insensitive duplicate variable names are now run by checkVarNames() and not outside of it
• equalData() and equalGADS() now work with GADSdat objects of length one (#127)

eatGADS 1.1.1

new features

• inspectDifferences() and inspectMetaDifferences() now allow comparisons of variables within the same GADSdat object (#62)
• new function import_tibble() allows importing tibbles as GADSdat objects (#88)

bug fixes

• applyLookup() now works for multiple variables that contain NAs (#68)
• applyChangeMeta() and recodeGADS() now correctly perform recodings (and throw errors) if multiple meta data conflicts occur (#57)
• removeEmptyValLabels() for removing unused missing tags and value labels (#4)
• extractData() and extractData2() now correctly apply value labels, even when value and value label conflicts exist
• write_spss2() now also handles variables containing only NA values when format is also NA (#72)
• changeMissings() now correctly changes missing labels for values of variables with partially non-existent values and/or valLabels (#73)
• autoRecode() now correctly overwrites the existing variable if var_suffix = "" (#84)
• The output of inspectMetaDifferences() is now correctly named even if differences in variable labels and SPSS format occur (#81)
• extractData2() and extractData() now correctly transform variables with duplicate value labels (#77)

documentation

• import_spss vignette updated to use extractData2() instead of extractData()

internal

• refactored extractData() to use extractData2() internally (#82)

eatGADS 1.1.0

new features

• equalGADS() now allows pre-sorting the data by an identifier variable
• updateMeta(), applyChangeMeta(), changeVarNames(), cloneVariable(), createVariable(), composeVar(), and dummies2char() now have optional checks of new variables names via the checkVarNames argument
• reuseMeta() now can be use on multiple variables at once
• inspectMetaDifferences() now can be applied to data bases as well
• recodeNA2missing() for recoding NAs to a specific missing code
• recode2NA() now allows the recoding of multiple values at once and returns a warning, if the recoded values have existing value labels in the recoded variables
• updateMeta() is now compatible with extractData() and extractData2()
• S3 method of extractData2() now available for trend_GADSdat objects
• recodeGADS() and applyChangeMeta() allow recoding values without recoding value labels (via existingMeta = "ignore")
• insertVariable() has been renamed relocateVariable() for clarity. Variables can now be inserted at the very beginning of a GADSdat
• cloneVariable() and autoRecode() now allow automatic appending of variable label suffixes via the label_suffix argument
• new function emptyTheseVariables() allows setting multiple variables to NA

bug fixes

• import_spss() now removes duplicate meta data rows, which caused an error in, e.g., recodeGADS()
• export_tibble() and write_spss() now throw an error if a conversion of four or more discrete missing tags into a missing range has undesired side effects
• bug fix in checkMissingsByValues(), now correctly reports missing tags outside of the specified value range
• bug fix in cloneVariable(), now new variables names which are also SQLite keywords no longer throw a error (fixed conflicts caused by checkVarName())

eatGADS 1.0.0

new features

• multiChar2fac() now allows converting to upper or lower cases via the convertCases argument
• checkMissingsByValues() for checking missing tags for specific, labeled values (or a value range)
• import_spss() with more informative error message
• checkMissingValLabels() now allows checks for specific variable classes
• checkMissingValLabels() now offers data.frame output format
• extractData2() for better usability and support of ordered factors
• fixEncoding() now supports imports of windows-1250 imported files via import_spss() with UTF-8 encoding
• import_spss() now transforms DATEATIME and ADATE variables to character, assigning an appropriate format
• cloneVariable() for duplicating an existing variable under a new name
• createVariable() for creating a new, empty variable
• insertVariable() for changing the position of a variable in the data set
• autoRecode() for automatically recoding (numerical) variables

bug fixes

• bug fix in multiChar2fac() for cases with existing value labels (with values labeled greater than zero)
• composeVar() now correctly prioritizes if both variables have missing values
• changeMissings() now does no longer add arbitrary value labels for newly tagged values
• bug fix in checkUniqueness2() for cases with differing numbers of cases per imputed data set
• bug fix in changeValLabels() for a mixture of already existing and new values
• bug fixes in import_spss() when setting labeledStirngs = "transform"
• bug fix in import_spss() for variables with missing tags but no value labels (these missing tags were dropped before)

eatGADS 0.20.0

• extractData() adds now all variable labels as a label attribute to the resulting data.frame
• a numeric tolerance can now be specified within equalGADS()
• fixed warnings for getTrendGADS()
• inspectMetaDifferences() for inspecting meta differences between two GADSdats and a single variable
• checkUniqueness2() for faster checking if a variable is unique within an identifier variable
• fixEncoding() for fixing encoding issues caused during the import via import_spss() using ASCII or other encoding
• composeVar() for combining the information of two variables in a third variable
• subImputations() now works with differing varName and varName_imp
• fillImputations() for efficiently filling missing values in imputed variables via not imputed variables
• checkFormat() now works correctly for variables without value labels
• checkFormat() now works correctly for variables with purely NAs
• import_raw() now works correctly if variables are provided as integers
• reuseMeta() now works correctly if variables occur in multiple sheets in a data base or all_GADSdat

eatGADS 0.19.1

• fix permanent URL redirect in README

eatGADS 0.19.0

breaking changes

• getTrendGADS() now supports multiple measurement points but support for linking errors has been dropped
• extractData() now supports multiple measurement points but support for linking errors has been dropped
• trend example data bases implemented
• getTrendGADSOld() and extractDataOld() provide backward compatibility
• checkLEStructure() deprecated
• import_spss() now can automatically transform labeled or missing tagged character values via the changed labeledStrings argument

major changes

• assimilateValLabels() for assimilating value labels of multiple variables
• cbind.GADSdat() method for binding multiple GADSdat objects by column
• calculateScale() for calculating scales from item sets
• checkUniqueness() for checking uniqueness of a variable within an id variable

minor changes

• clarifications and additional explanations in vignettes (meta_data)
• bug fix updateMeta() (all added variables are now checked for illegal naming)

eatGADS 0.18.1

• equalGADS() now ignores irrelevant format differences (F8.0 vs F8)
• fix JSS link in vignette

eatGADS 0.18.0

• applyChangeMeta() and changeValLabels() now automatically assign "valid" to the missings columns for labeled values
• bug fix equalGADS() (meta data differences solely due to different meta data sorting are now ignored)
• equalGADS() can now exclude certain meta data columns from the comparison
• checkFormat() for checking and modifyingSPSS format type compatibility with actual data
• check4SPSS() for checking SPSS meta data conventions and requirements
• changeVarNames() and applyChangeMeta() now enforce SQLite3 naming conventions
• bug fixes write_spss2()
• bug fix applyChangeMeta(): now recognizes differences in NAs in old columns when checking the change table

eatGADS 0.17.0

• inspectDifferences() now checks whether variable is of the same type in both GADSdats
• subImputations() for substituting imputations with original, not imputed values
• applyChangeMeta() now returns a more informative error if the variable sets in the GADSdat and the changeTable differ
• applyChangeMeta() tries to convert value columns in changeTable to numeric before throwing an error
• dummies2char() for recoding a set of dummy variables to a set of left filled character variables
• fac2dummies() and fac2dummies_complex() for recoding a labeled factor variable to multiple labeled dummy variables
• equalGADS() for comparing the complete structure and content of two GADSdat objects
• inspectDifferences() for comparing specific variables
• reuseMeta() now allows only transferring value labels for values coded as missing
• checkEmptyValLabels() and checkMissingValLabels() provide cleaner output
• write_spss2() for writing .sav files via a text file and a SPSS syntax
• import_spss() and write_spss() now deal correctly with missing codes for character variables
• recodeGADS() allows recoding of unlabeled variables and values
• import_spss() now takes an encoding argument to work around faulty defaults

Internal

• better performance of applyChangeMeta() on value level
• bug fix import_spss() (all columns are forced to be double when imported instead of integer)
• bug fix fac2dummies_complex() (value labels of dummy variables are tagged as valid)
• bug fixes recodeGADS()
• bug fix removeValLabels() (when a single value label was removed)
• bug fixes collapseMultiMC_Text() (left_fill() and drop_empty())
• bug fix remove2NAchar() (with partially labeled variables)
• bug fix multiChar2fac() (with partially labeled variables)
• bug fix applyLookup_expandVar() (removed unnecessary messages)
• warnings removed for import_spss() (haven patch for labeled character variables)

eatGADS 0.16.0

• checkEmptyValLabels() and checkMissingValLabels() for comparing values in the data and value labels
• compareGADS() for comparing multiple variables between two GADSdats
• recodeGADS() allows recoding (multiple) values into existing values now, argument existingMeta added
• recodeGADS() allows recoding of NA now
• recode2NA() allows recoding of numeric and character values to NA
• recodeString2NA() has been deprecated
• bug fix collapseColumns() (now supports new_value as column name in look up table)
• checkValue() now returns occurrence count as vector instead of giving a message
• checkValue() now works on a subset of variables (via argument vars)
• checkValue() now supports checking for NA
• removeValLabels() now can remove only specific value - valLabel pairs if required
• changing value labels inside of recodeGADS() via newValueLabels deprecated (use changeValLabels()instead)
• applyChangeMeta() now works with data.frames (for compatibility with excel import and export)
• import_spss() now supports variables of type datetime
• import_stata() and write_stata() for exporting and importing Stata files

Internal

• check_GADSdat() now checks for duplicate value rows
• switch from travis and appVeyor to Github Action
• applyChangeMeta() checks format_new column now for illegal entries
• applyChangeMeta() throws error if result has corrupted meta data
• applyChangeMeta() orders meta data within a variable (by ascending value)
• better performance of applyChangeMeta() on value level
• bug fix for applyChangeMeta() if only missings_new is changed
• Setup revdepcheck

eatGADS 0.15.2

• Initial release on CRAN.