- Bug fix: make it an error if tab names are started with a numeric value (#124).
- Reduce gif file size, include additional frame (#121).
- The new
custom_rows
argument is the third-to-last argument tocreate_a11ytable()
(it's been inserted beforesources
andtables
) so that the argument order reflects the order of the pre-table rows that appear above tables in the spreadsheet output; this will be a problem for users of {a11ytables} before v0.3 who might have previously specified arguments by place rather than by name (I assume most will have specified arguments by name, so I expect this won't be much of a problem).
- Added the
custom_rows
argument to thecreate_a11ytables()
function to supply arbitrary sentences to separate rows above a table in contents, cover and notes sheets (#74). - Allowed custom rows to be hyperlinks if the user provides them in Markdown-style (#74).
- Added the datasets
demo_df
('data.frame' class) anddemo_a11ytable
('a11ytables' class) to help demo the newcustom_rows
argument (and do a better job of showcasing other features of the package), supersedingmtcars_df
andmtcars_df2
.
- Adjusted Markdown-link detection in cases where the text portion contained parentheses (#119).
- Ensured the summary method printed table dimensions of cover list items (#79).
- Updated {roxygen2} function documentation given the introduction of
custom_rows
. - Updated vignettes to include detail on how to use
custom_rows
. - Updated accessibility checklist vignette to include 'sensible column width' item (#117).
- Simplified the output of the summary method (#79).
- Updated tests to use
demo_*
datasets. - Updated RStudio Addin given
custom_rows
and simplified to data.frame example only. - Added internal
.vector_to_sentence()
function to help construct comma-separated lists from a vector.
- Bugfix: corrected spelling error in installation instructions in README (#111).
- Accessibility fix: display number of tables with text rather than a numeral (#109).
- Allowed the cover information passed to
create_a11ytable()
to be supplied as a list rather than a data.frame, which means you can have an arbitrary number of rows under each section (#102). - Allowed the user to supply links in Markdown format (e.g.
[GOV.UK](https://www.gov.uk)
) when passing the cover information to thetables
argument ofcreate_a11ytables()
(#47). - Allowed the user to supply links in Markdown format (e.g.
[GOV.UK](https://gov.uk)
) when passing the data source reference to thesource
argument ofcreate_a11ytables()
(#47). - Added an
mtcars_df2
demo data set that contains a list input for the cover page, which itself contains Markdown-formatted hyperlinks.
- Ensured
stringsAsFactors
is set explicitly toFALSE
in thedata.frame
call withincreate_a11ytable()
, given that this default behaviour changed in R version 4 (#85).
- Updated the 'a11ytables' vignette given the new features.
- Updated function documentation given the new features.
- Renamed the
new_a11ytable()
function tocreate_a11ytable()
(#27). - Renamed the
create_a11y_wb()
function togenerate_workbook()
and changed the main argument from 'content' to 'a11ytable' (#27). - Removed the 'table_names' argument to
create_a11ytable()
and instead autogenerate them from the user-provided 'tab_title' (#61). - Introduced the
blank_cells
argument tocreate_a11ytable()
so the user can provide a table-by-table reason for why cells might be empty (#62). - Adjusted the in-built
mtcars_df
data set to better fit the changes to the functions.
- Provided a warning, rather than error, if you don't provide any 'source' argument to
create_a11ytable()
.
- Added a vignette to provide an overview of the code underlying the package so that it's easier for developers to contribute (#58).
- Added a vignette to explain the terminology of a spreadsheet, as used in the package (#58).
- Referred to the updated gptables Python package and the best practice spreadsheet guidance on the Analysis Function website (#73).
- Added hex logo (#81).
- Updated README and other vignettes to reflect changes to the function API.
- Corrected various typos.
- Increased test coverage (#63).
- Updated user-input sanitising: it now removes punctuation in 'tab_title', inserts underscores in place of spaces and adds full-stops to blank_cells and source data (#78).
- Consolidated all a11ytable-class-related functioning into
as_a11ytable()
so thatcreate_a11ytable()
only has to build a data.frame from supplied arguments and applyas_a11ytable()
to it (#80). - Added warning for user if the 'blank_cells' argument is provided to
create_a11ytable()
but there are no blank cells in the table, and vice versa. - Updated content of RStudio Addins to reflect function and argument changes.
- Adjusted default column widths in notes and contents.
- Renamed files in R/ for consistency and clarity.
- Switched to
inherits()
for class detection inif()
statements.
- Added and improved warnings and errors (#34).
- Simplified the in-built data set so it matches the 'get started' vignette.
- Updated the gif used in the 'get started' vignette.
- Added explanations to the accessibility checklist vignette.
- Converted the 'quickstart' vignette to 'get started', used a simpler example, added gif.
- Removed the 'construct' vignette.
- Used 'reference' in _pkgdown.yaml to better separate functions on the site.
- Added an RStudio Addin to insert code for an {a11ytables} workflow (#54).
- Made source, notes and table insertion dynamic, so the row of insertion of each is dependent on the others (#33).
- Applied table styling based on dynamic table placement (#33).
- Prevented the tab title being pasted into the titles of tables sheets (#48).
- Made sure to wrap non-header cells in tables sheets (#41).
- Provided an auto-column-width adjustment in tables where cell content or headers is long (#41).
- Added warnings if issues are detected when making an a11ytables-class object (#34).
- Added a modified mtcars example data set,
mtcars_df
(#20). - Expunged
lfs_tables
(#20). - Added 'Crown Copyright' to authors as the copyright holder .
- Updated README and vignettes given these changes.
- Added new quickstart vignette and simplified README (#43).
- Handled plurals and periods in meta sheet elements, update detection of notes in column headers.
- HOTFIX: corrected
.detect_notes()
so it actually works. - Converted the
print
method to asummary
method, as suggested by @TimTaylor (#23). - Imported {pillar} for {tibble}-style printing, as suggested by @TimTaylor (#26).
- Added @TimTaylor as a contributor.
- Updated vignettes and README.
- Ensured numeric columns (even if they contain a string like '[c]' to indicate a suppressed value) are right-aligned (#32).
- Isolated out
.insert_*()
functions for table count and note presence from.insert_prelim_a11y()
, for clarity.
- BREAKING CHANGE: removed subtables because they're not being used for now and are confusing.
- Added the first draft vignettes on how to create an a11ytable and an accessibility guidance checklist (#22).
- HOTFIX: fixed a problem where the
tab_title
was being used to filter when it should have beensheet_type
.
- BREAKING CHANGE: the user-supplied table for the cover sheet should now be supplied as a tidy two-column data.frame with a row per subsection, with columns for the 'subsection title' and 'subsection body'.
- Updated the
lfs_tables
andlfs_subtables
example data sets given the above; for an example seelfs_tables[lfs_tables$sheet_type == "cover", "table"][[1]]
.
- BREAKING CHANGE: switched to using the
sheet_type
column of an a11ytable-class object to infer the sheet type as 'cover', 'contents', 'notes' or 'tables' (replacing 'meta'); it's no longer thetab_title
that serves this role (#18). - Ensured styles and
.add_*()
now make internal references tosheet_type
rather thantab_title
.
- BREAKING CHANGE: retired fully the
add_*()
function family for building a workbook 'by hand', given the introduction ofcreate_a11y_wb()
. - Introduced S3 class 'a11ytable', including exported functions to generate (
new_a11ytable()
,as_a11ytable()
) and test (is_a11ytable()
) objects of this class.
- Simplified all
add_*()
functions to a singlecreate_a11y_wb()
function (moving towards #15), thoughadd_*()
functions are all still exported. - Split supplied data into two, with and without sub-table examples:
lfs_tables
andlfs_subtables
, respectively. - Updated README to detail simplified approach.
- Added un-exported
.style_*()
utils functions to style worksheets (#1). - Incorporated style functions into
add_*table()*
functions. - Updated README.
- Added basic tests, defensive stops (#10, #11).
- Added GitHub Actions for RMD check, coverage and {pkgdown} (#7).
- Transferred ownership from matt-dray to co-analysis (#9).
- Added a
NEWS.md
,CODE_OF_CONDUCT.md
,README.Rmd
. - Added exported
add_*()
functions to create different tab types (cover, contents, notes, tables). - Added un-exported
.insert_*()
utils functions to insert within-sheet elements. - Added MIT license with
LICENSE.md
.