Long standing issues

snakecase 0.11.0 (25.05.19)

Changes in 0.10.0 and 0.10.0.9000

New functionality:

• abbreviations: Abbreviations are now ...
  • ... matched case-insensitive inside of groups of connected lower/upper case sequences.
  • ... consistently formatted according to the supplied case.
  • ... formatted exactly as supplied for the cases title, mixed, lower and upper camel.
  • ... protected from the parsing. This means that
    • special characters in abbreviations don't need to be taken care of anymore. The formatting of digits or special characters like hyphens, colons etc. will be formatted as specified.
• parsing_option:
  • new parsing option 3 parses "SOmeNIceSTUFf".
  • parsing options starting with a minus (-1, -2, -3) don't interprete non-alphanumeric characters as word boundaries. E.g. "This.text" will stay "This.text" in upper camel case.
• cases
  • title: New case which should be especially useful for automatic generation of proper labels within graphics or business reports. Builds up on sentence case which is wrapped within tools::toTitleCase().
  • random: New case, which will randomly convert letters into upper or lower case.
• numerals: new alignment option "tight" which allows to suppress all underscores between numerals and letters.
• sep_out: sep_out gets proper vectorization. In case of length(sep_out) > 1 differing substrings are connected by the (possibly) differing elements of sep_out. Therefore, sep_out gets accordingly shortened or the last element of sep_out gets recycled to match the number of needed separators for each element of string.
• UTF8 Conversion: Input is now always converted to UTF-8 and returned as UTF8. Also the transliterations argument is now aware of non-UTF8 encoded input.

Infrastructure:

• CII best practices badge: Achieved the criteria for the best practices badge. Current status is under https://bestpractices.coreinfrastructure.org/de/projects/2475
• R Version: Increase from 3.1 to 3.2 regarding the usage of tools::toTitleCase() inside to_title_case().
• Vignettes: The blog article "Introducing the snakecase package" was added as a vignette.
• Documentation: The readme, the examples and the function documentation were updated according to the new functionality.
• Resolve CRAN notes:
  • Tests: Skip to_any_case() tests (janitor-pkg-tests, transliterations and complex strings) when platform charset is not UTF-8 to resolve CRAN notification regarding character encoding.
  • Vignettes: Build the package with new version of knitr to resolve CRAN notification regarding vignette encoding.

More options

snakecase 0.9.2

• cases: added to_sentence_case() (same as snake, but first letter is uppercase and default sep_out is space).
• numerals: added numerals argument to all caseconverters including to_any_case() to format the alignment of digits (middle, left,right, asis). Therefore parsing_option nr 4 might be removed in later releases, as it is the same as parsing_option = 1 and numerals = "asis".
• transliterations: When named character elements are supplied as part of transliterations, anyting that mateches the name is replaced by the corresponding value.
• attributes are now preserved (not only names as before)

Beginner friendly

sep_in: Change default from NULL to "[^[:alnum:]]". This will make it easier for beginners and in general also faster to modify cases from strings like names(iris). Updated the regarding sections in the vignette, docu and readme.

Roundup

• parsing_options:
  • old parsing_options 3 and 4 are replaced now by new
  • parsing_option 3, which suppreses case conversion around alpha numerics
  • parsing_option 4, which introduces less formatting of numerals in the output,
    and leaves them very close to the way that they appeared in the input strings.
• abbreviations:
  • they work now more consistent with cases like lower- and upper camel case
• new converters:
  • to_swap_case is new. Within to_any_case this conversion can be called also via case = "flip".
• removed deprecated arguments
  • replace_special_characters, which is now called transliterations
  • preprocess, which is now called sep_in
  • postprocess, which is now called sep_out
• removed dependencies:
  • purrr and magrittr are not longer dependencies
  • stringr is the only dependency now (including stringi of course).

Interface simplifications

• reordering of the arguments of all to_xxx_case() functions
• renaming preprocess to sep_in, postprocess to sep_out, replace_special_characters to transliterations. Ols names are deprecated.
• removed protect argument

CRAN update

• to_xxx_case() shortcuts are now exact wrappers around to_any_case()
• protect is deprecated after changing implementation and setting a reosonable default
• added abbreviations argument
• case "none" is now a lot more general for formatting
• added abbreviation specific behaviour for mixed case
• new parsing_option 5, which suppresses conversion after ., @, etc
• renamig of:
  • to_small / to_big_camel_case have been renamed to to_lower / to_upper_camel_case. The old names are and will still be supported in to_any_case()
  • parsingoption to parsing_option
• introduced rule that parsing_option <= 0 suppresses parsing from now on
• lots of additional tests and smaller bugfixes
• several documentation updates including help, examples, readme and vignette

Preserve names attribute

• Changes since last CRAN submission include:
  • to_any_case() and all to_xxx_case() shortcuts now preserve the names attribute, which can be especially handy to integrate snakecase within programmatic ggplot wrappers to automatically convert labels of a plot.

snakecase on CRAN now

Functions

• to_any_case():
  • has a highly modular and logical internal design
  • arguments: string, case, preprocess, protect, replace_special_characters, postprocess, prefix, postfix, unique_sep and empty_fill
  • vectorised over string, pre-/postprocess, pre-/postfix and protect arguments
  • groups of digits are not separated in between. Digits, which are not direct next to each other, will be split via "_" in both camel case versions. Otherwise information would be lost and also the consistency rules in the readme wouldn't hold in this case.
• to_parsed_case_internal()
  • does the preprocessing and parsing and simplifies all other functions.
• shortcuts:
  • to_snake_case(), to_small_camel_case(), to_big_camel_case(), to_screaming_snake_case() and to_parsed_case(), which are equivalent to to_any_case(string, case).

Documentation

• githubpage contains a readme (including a logo and badges for cran status, codecoverage etc.), which describes the usage and also the design philosophy
• added a vignette
• added documentation and examples for all functions

Tests

• added systematic tests
• added integrated tests on linux via travis.ci and on windows via appveyor