blockr.dplyr

Data wrangling, one block per verb: filter, select, mutate, summarize, join, pivot and friends. If you would reach for dplyr in R, you reach for these blocks in blockr.

Arrange Rows

new_arrange_block() · transform

Sort rows by one or more columns, ascending or descending. (dplyr: arrange)

Argument	Description
columns	Array of {column, direction} objects where direction is "asc" or "desc". Rows are sorted by the columns in order.

Bind Columns

new_bind_cols_block() · transform

Place tables side-by-side horizontally. Both must have the same number of rows. (dplyr: bind_cols)

Bind Rows

new_bind_rows_block() · transform

Stack tables vertically by matching column names. (dplyr: bind_rows)

Argument	Description
id_name	Optional string -- name of an .id column identifying which input each row came from

Filter Rows

new_filter_block() · transform

Keep or remove rows by values, comparisons, or expressions. (dplyr: filter)

Argument	Description
conditions	Array of condition objects. Types: "values" (column, values array of strings, mode "include"/"exclude"), "numeric" (column, op like ">"/">="/"<"/"<=", value as number), "expr" (expr as R expression string)
operator	How to combine conditions: "&" for AND (all match), "|" for OR (any match)
preserve_order	Boolean -- when true, keep the original row order rather than reordering

Join

new_join_block() · transform

Combine two tables by matching values in shared columns. Left, inner, right, full, anti, semi joins. (dplyr: *_join)

Argument	Description
type	Join type string: left_join, inner_join, right_join, full_join, anti_join, or semi_join
keys	Array of {xCol, op, yCol} objects defining the column pairs to match
exprs	Array of R expression strings (optional, advanced join predicates)
suffix_x	Suffix added to overlapping column names from the left table (default '.x')
suffix_y	Suffix added to overlapping column names from the right table (default '.y')

Mutate Columns

new_mutate_block() · transform

Add new columns or transform existing ones using R expressions. (dplyr: mutate)

Argument	Description
mutations	Array of objects, each with 'name' (new column name) and 'expr' (R expression string)
by	Optional array of grouping column names (group the operation, e.g. mean per group)

Pivot Longer

new_pivot_longer_block() · transform

Reshape wide data to long — gather columns into rows. (tidyr: pivot_longer)

Argument	Description
cols	Array of column names to pivot to long form
names_to	String -- name of the new column holding the pivoted names
values_to	String -- name of the new column holding the pivoted values
values_drop_na	Boolean -- drop rows with NA values after pivoting
names_prefix	String -- regex prefix stripped from the pivoted names

Pivot Wider

new_pivot_wider_block() · transform

Reshape long data to wide — spread rows into columns. (tidyr: pivot_wider)

Argument	Description
names_from	Array of column(s) whose values become the new column names
values_from	Array of column(s) whose values fill the new columns
id_cols	Array (optional) of columns that uniquely identify each output row
values_fill	Value (or null) used to fill missing cells
names_sep	String separator joining multiple names_from / prefix parts
names_prefix	String prepended to every new column name
values_fn	String (or null) aggregation function (e.g. 'mean', 'sum', 'first') for duplicate id+name combinations

Rename

new_rename_block() · transform

Change column names without modifying data. (dplyr: rename)

Argument	Description
renames	Object mapping new names to old names: {new_name: 'old_name', ...}. Keys are the desired new column names, values are the existing column names.

Select Columns

new_select_block() · transform

Choose, reorder, or exclude columns. Optionally remove duplicate rows. (dplyr: select, distinct)

Argument	Description
columns	Array of column name strings to keep (or to remove, when exclude=true)
exclude	Boolean -- when true, drop the listed columns instead of keeping them
distinct	Boolean -- when true, deduplicate rows after selecting

Separate Column

new_separate_block() · transform

Split one column into multiple columns using a delimiter. (tidyr: separate)

Argument	Description
col	Source column to split
into	Array of new column names to split into
sep	Separator string (or regex) to split on
remove	Boolean -- remove the source column after splitting
convert	Boolean -- auto-convert the new columns to numeric/logical types

Slice Rows

new_slice_block() · transform

Pick rows by position, random sample, or min/max of a column. (dplyr: slice)

Argument	Description
type	One of 'head' (first n rows), 'tail' (last n), 'min'/'max' (rows with smallest/largest order_by values), 'sample' (random sample)
n	Integer row count
prop	Number 0-1, proportion of rows (alternative to n -- not both)
order_by	Column to rank by, for min/max
with_ties	Boolean -- whether tied rows at the cutoff are all kept (min/max)
weight_by	Column of weights, for sampling
replace	Boolean -- sample with replacement
by	Array of grouping columns (slice within each group)

Summarize

new_summarize_block() · transform

Calculate totals, averages, counts, and other statistics, optionally by group. (dplyr: summarize)

Argument	Description
summaries	Array of summary objects. Types: 'simple' (name, func like 'mean'/'sd'/'sum'/'n'/'n_distinct', col) or 'expr' (name, expr as R expression string)
by	Array of grouping column names -- this is how you group (do NOT write group_by() in exprs)

Unite Columns

new_unite_block() · transform

Merge multiple columns into one by pasting values together. (tidyr: unite)

Argument	Description
col	New column name for the united result
cols	Array of columns to unite
sep	Separator string placed between values
remove	Boolean -- remove the input columns after uniting
na_rm	Boolean -- drop NA values before uniting