Functions and methods · phplyr

phplyr’s functions are namespaced functions in Phplyr. A function marked with a chain method also runs as that method on phplyr($table). Expression methods are called on the vector helpers that col() and other functions return.

Pipes and method chains

Function	Chain method	Purpose
`phplyr()`		Start a method chain on a table.
`pipe()`		Apply steps to a value in order.

Single-table functions

Rows

Function	Chain method	Purpose
`filter()`	`->filter()`	Keep rows where every condition is true.
`filter_out()`	`->filterOut()`	Drop rows where every condition is true (dplyr 1.2).
`arrange()`	`->arrange()`	Sort rows by columns or expressions.
`desc()`		Descending sort key for `arrange()`, and for rank helpers.
`distinct()`	`->distinct()`	Unique rows over the given columns (all columns when none are given).
`slice()`	`->slice()`	Rows by 1-based position within each group, like dplyr’s `slice()`.
`slice_head()`	`->sliceHead()`	First rows of each group.
`slice_tail()`	`->sliceTail()`	Last rows of each group.
`slice_min()`	`->sliceMin()`	Rows with the smallest values of `$orderBy` in each group, ordered by it.
`slice_max()`	`->sliceMax()`	Rows with the largest values of `$orderBy` in each group, ordered by it.
`slice_sample()`	`->sliceSample()`	Random rows from each group.
`head()`	`->head()`	First `$n` rows of the whole table, ignoring groups (like utils::head).
`top_n()`	`->topN()`	The n rows with the largest wt in each group, ties included, in their original order (superseded by `slice_max()`).
`top_frac()`	`->topFrac()`	`top_n()` with a proportion of each group’s rows (superseded by `slice_max(prop:)`).
`sample_n()`	`->sampleN()`	n random rows from each group (superseded by `slice_sample()`).
`sample_frac()`	`->sampleFrac()`	A proportion of random rows from each group (superseded by `slice_sample(prop:)`).
`add_row()`	`->addRow()`	Add rows of values.

Columns

Function	Chain method	Purpose
`select()`	`->select()`	Keep or drop columns.
`rename()`	`->rename()`	Rename columns, new name as the argument name.
`rename_with()`	`->renameWith()`	Rename selected columns (all by default) with a function of the old name.
`relocate()`	`->relocate()`	Move columns.
`mutate()`	`->mutate()`	Add, replace, or drop (with null) columns, named by argument.
`transmute()`	`->transmute()`	Like `mutate()`, keeping only the grouping columns and the new columns.
`pull()`	`->pull()`	Extract one column as a PHP list.

Summaries

Function	Chain method	Purpose
`summarise()`	`->summarise()`	One row per group.
`summarize()`	`->summarize()`	Alias of `summarise()`.
`reframe()`	`->reframe()`	Like `summarise()`, but each summary may return any number of values (a list).
`count()`	`->count()`	Count rows per combination of columns, within any existing groups.
`tally()`	`->tally()`	Count rows in each group (`summarise(n = n())`).
`add_count()`	`->addCount()`	Like `count()`, but adds the count as a column instead of summarising.
`add_tally()`	`->addTally()`	Like `tally()`, but adds the count as a column.
`glimpse()`	`->glimpse()`	Print a compact column-by-column summary and return the table unchanged, so it can sit in the middle of a pipeline.

Joins, binding, and set operations

Joins

Function	Chain method	Purpose
`inner_join()`	`->innerJoin()`	Rows of x with a match in y, with y’s columns.
`left_join()`	`->leftJoin()`	All rows of x, with y’s columns where they match (NA otherwise).
`right_join()`	`->rightJoin()`	All rows of y, with x’s columns where they match.
`full_join()`	`->fullJoin()`	All rows of both tables.
`semi_join()`	`->semiJoin()`	Rows of x that have a match in y (x’s columns only).
`anti_join()`	`->antiJoin()`	Rows of x that have no match in y.
`cross_join()`	`->crossJoin()`	Every combination of rows of x and y.
`nest_join()`	`->nestJoin()`	All rows of x, with a list-column (name) holding each row’s matching rows of y as a table.

join_by()

Function	Chain method	Purpose
`join_by()`		A join specification, as dplyr’s `join_by()`.
`closest()`		A rolling join.
`within()`		x’s interval [xLower, xUpper] lies within y’s [yLower, yUpper].
`overlaps()`		x’s interval [xLower, xUpper] overlaps y’s [yLower, yUpper].

Binding

Function	Chain method	Purpose
`bind_rows()`	`->bindRows()`	Stack tables under the piped one, matching columns by name.
`bind_cols()`	`->bindCols()`	Put tables side by side with the piped one.

Set operations and row updates

Function	Chain method	Purpose
`union()`	`->union()`	Distinct rows of both tables.
`union_all()`	`->unionAll()`	All rows of both tables.
`intersect()`	`->intersect()`	Distinct rows of x that also appear in y.
`setdiff()`	`->setdiff()`	Distinct rows of x that don’t appear in y.
`symdiff()`	`->symdiff()`	Rows in exactly one of the tables.
`setequal()`	`->setequal()`	True when both tables hold the same distinct rows, in any order.
`rows_insert()`	`->rowsInsert()`	Append rows of y whose keys are new.
`rows_append()`	`->rowsAppend()`	Append all rows of y.
`rows_update()`	`->rowsUpdate()`	Overwrite x’s values with y’s for matching keys.
`rows_patch()`	`->rowsPatch()`	Like `rows_update()`, but only fills x’s NA values.
`rows_upsert()`	`->rowsUpsert()`	Update matching keys and insert new ones.
`rows_delete()`	`->rowsDelete()`	Delete rows of x whose keys appear in y.

Grouping

Grouping and ungrouping

Function	Chain method	Purpose
`group_by()`	`->groupBy()`	Group by columns (names or selectors).
`ungroup()`	`->ungroup()`	Remove grouping, or only the named grouping columns.
`with_groups()`	`->withGroups()`	Run a table function on the table grouped by `_groups`, then restore the original grouping (superseded in dplyr by `_by`).

Group utilities

Function	Chain method	Purpose
`group_vars()`	`->groupVars()`	The grouping column names.
`n_groups()`	`->nGroups()`	The number of groups (1 for an ungrouped table).
`group_size()`	`->groupSize()`	Rows in each group, in group order.
`group_indices()`	`->groupIndices()`	Each row’s 1-based group number.
`group_keys()`	`->groupKeys()`	One row per group with the key columns.
`group_rows()`	`->groupRows()`	Each group’s 1-based row numbers, in group order.
`group_data()`	`->groupData()`	`group_keys()` with a .rows list-column of each group’s 1-based row numbers.
`group_trim()`	`->groupTrim()`	Drop unused factor levels from the grouping columns, and with them the empty groups that drop.
`group_split()`	`->groupSplit()`	Split into a list of ungrouped tables, one per group.
`group_map()`	`->groupMap()`	Call `fn (Table $group, array $key)` for each group and return the list of results.
`group_modify()`	`->groupModify()`	Replace each group with the table `fn (Table $group, array $key)` returns, then put the key columns back in front.
`group_walk()`	`->groupWalk()`	Call `fn (Table $group, array $key)` for each group.
`group_nest()`	`->groupNest()`	One row per group.
`nest_by()`	`->nestBy()`	Nest by the given columns into a rowwise table.
`cur_group_id()`		The current group’s 1-based number, inside `mutate()`, `summarise()`, or `reframe()`.
`cur_group_rows()`		The current group’s 1-based row numbers in the table.
`cur_group()`		The current group’s key, as an array of grouping column => value (dplyr returns a one-row tibble).

Row-wise operations

Function	Chain method	Purpose
`rowwise()`	`->rowwise()`	Treat each row as its own group, for row-wise summaries.

Vector helpers

Summaries

Function	Chain method	Purpose
`n()`		Number of rows in the group.
`n_distinct()`		Number of distinct values (or combinations).
`sum()`		Sum of x.
`mean()`		Arithmetic mean of x.
`median()`		Median of x.
`sd()`		Sample standard deviation (n - 1).
`var_()`		Sample variance (n - 1).
`variance()`		Alias of `var_()`.
`quantile()`		Type 7 quantile, R’s default.
`iqr()`		Interquartile range (type 7).
`min()`		Smallest value of x.
`max()`		Largest value of x.
`first()`		First value.
`last()`		Last value.
`nth()`		The `$n`-th value, 1-based.
`any()`		True when any value is true (R’s `any()`.
`all()`		True when all values are true (R’s `all()`.

Window functions

Function	Chain method	Purpose
`lag()`		Previous values.
`lead()`		Next values.
`row_number()`		Row numbers.
`min_rank()`		Ranks with ties sharing the lowest rank (1, 2, 2, 4).
`dense_rank()`		Ranks with no gaps after ties (1, 2, 2, 3).
`percent_rank()`		min_rank rescaled to [0, 1].
`cume_dist()`		Proportion of values less than or equal to each value.
`ntile()`		Split into `$n` buckets of near-equal size.
`consecutive_id()`		A new id each time the value (or combination) changes.
`cumsum()`		Running sum.
`cumprod()`		Running product.
`cummin()`		Running minimum.
`cummax()`		Running maximum.
`cummean()`		Running mean.
`cumall()`		Running `all()`.
`cumany()`		Running `any()`.
`order_by()`		Evaluate a window function with the rows in another order, then put the results back.

Conditionals and NA

Function	Chain method	Purpose
`if_else()`		Vectorised if/else.
`case_when()`		The value of the first case whose condition is true, else `$default`.
`case_match()`		Match values like a switch (deprecated in dplyr 1.2 in favor of `recode_values()`).
`replace_when()`		Like `case_when()`, but rows matching no case keep x’s value.
`recode()`		Replace values by name (superseded in dplyr by `recode_values()`).
`recode_values()`		Map values to new values (dplyr 1.2).
`replace_values()`		Like `recode_values()`, but unmatched values keep their original value.
`coalesce()`		First non-NA value across the arguments, row by row.
`na_if()`		NA where `$x` equals `$y`.
`between()`		`$left` <= `$x` <= `$right`.
`near()`		Equality with tolerance, for floating point.
`when_any()`		Elementwise `any()` across conditions (x \| y \| z).
`when_all()`		Elementwise `all()` across conditions (x & y & z).

Strings and dates

Strings

Function	Chain method	Purpose
`str_detect()`		Does the pattern occur?.
`str_starts()`		Does the string start with the pattern?.
`str_ends()`		Does the string end with the pattern?.
`str_count()`		Number of matches.
`str_extract()`		The first match, or NA.
`str_extract_all()`		Every match, as a list per row (a list-column).
`str_replace()`		Replace the first match.
`str_replace_all()`		Replace every match.
`str_split()`		Split into pieces, as a list per row (a list-column).
`str_split_i()`		The i-th piece of a split (1-based.
`str_pad()`		Pad to a width.
`str_trunc()`		Truncate to a width, marking the cut with an ellipsis.
`str_trim()`		Trim whitespace.
`str_squish()`		Trim and collapse internal whitespace.
`str_sub()`		Substring by 1-based positions, as stringr’s `str_sub()`.
`str_length()`		Number of characters.
`str_to_title()`		Title case.
`str_to_upper()`		Uppercase, as stringr’s `str_to_upper()`.
`str_to_lower()`		Lowercase, as stringr’s `str_to_lower()`.
`str_upper()`		Uppercase.
`str_lower()`		Lowercase.
`paste()`		Paste values together, separated by sep (default a space).
`paste0()`		Paste values together.
`fixed()`		A literal pattern.
`regex()`		A regular expression with options.

Dates and times

Function	Chain method	Purpose
`ymd()`		Parse year-month-day strings to dates, with any separator.
`dmy()`		Parse day-month-year strings to dates, with any separator.
`mdy()`		Parse month-day-year strings to dates, with any separator.
`ymd_hms()`		Parse year-month-day hour:minute:second strings to datetimes, in UTC or tz.
`dmy_hms()`		Parse day-month-year hour:minute:second strings to datetimes, in UTC or tz.
`mdy_hms()`		Parse month-day-year hour:minute:second strings to datetimes, in UTC or tz.
`parse_date_time()`		Parse with lubridate orders.
`year()`		The year of a date or datetime.
`month()`		The month, 1-12.
`day()`		The day of the month.
`wday()`		Day of the week, 1-7.
`yday()`		The day of the year, 1-366.
`quarter()`		The quarter, 1-4.
`hour()`		The hour of a datetime, 0-23.
`minute()`		The minute of a datetime, 0-59.
`second()`		The second of a datetime, 0-59.
`floor_date()`		Round down to a unit.
`ceiling_date()`		Round up to a unit, as `floor_date()`.
`round_date()`		Round to the nearest unit, as `floor_date()`.
`seconds()`		A period of n seconds, for date arithmetic.
`minutes()`		A period of n minutes.
`hours()`		A period of n hours.
`days()`		A period of n days.
`weeks()`		A period of n weeks.
`months()`		A period of n months.
`years()`		A period of n years.
`difftime()`		The difference x - y in units ‘secs’, ‘mins’, ‘hours’, ‘days’, or ‘weeks’, as a double.
`with_tz()`		The same instant in another time zone.
`force_tz()`		The same clock time in another time zone.
`today()`		The current date, in the system time zone or tz.
`now()`		The current datetime, in the system time zone or tz.

Selecting columns

Selectors

Function	Chain method	Purpose
`starts_with()`		Columns whose names start with a prefix (case-insensitive by default).
`ends_with()`		Columns whose names end with a suffix (case-insensitive by default).
`contains()`		Columns whose names contain a string (case-insensitive by default).
`matches()`		Columns whose names match a regular expression.
`num_range()`		Columns named prefix + number for each number.
`everything()`		Every column.
`last_col()`		The last column, or the one `$offset` before it.
`all_of()`		Exactly these columns.
`any_of()`		These columns, skipping any that don’t exist.
`where()`		Columns for which a predicate holds.
`except()`		Everything except these columns.
`any_selected()`		Columns matched by any of the selections (union).
`all_selected()`		Columns matched by every selection (intersection, like a & b in tidyselect).
`group_cols()`		The grouping columns.

across()

Function	Chain method	Purpose
`across()`		Apply functions to columns inside `mutate()` or `summarise()`.
`if_any()`		Per-row test that a predicate holds for any of the columns, for `filter()`.
`if_all()`		Per-row test that a predicate holds for all of the columns.
`c_across()`		The values of the selected columns in the current row, for `rowwise()` tables.
`pick()`	`->pick()`	The selected columns of the current group as a table, for use in `summarise()` closures or with `vec()`.
`cur_column()`		The name of the column `across()` is working on, inside one of its functions.

Reshaping with tidyr

Pivoting

Function	Chain method	Purpose
`pivot_longer()`	`->pivotLonger()`	Lengthen.
`pivot_wider()`	`->pivotWider()`	Widen.
`build_longer_spec()`	`->buildLongerSpec()`	The spec `pivot_longer()` would use, as a table with .name, .value and the namesTo columns.
`pivot_longer_spec()`	`->pivotLongerSpec()`	Lengthen with a spec from `build_longer_spec()` (a table or a list of rows).
`build_wider_spec()`	`->buildWiderSpec()`	The spec `pivot_wider()` would use, as a table with .name, .value and the namesFrom columns.
`pivot_wider_spec()`	`->pivotWiderSpec()`	Widen with a spec from `build_wider_spec()`.
`gather()`	`->gather()`	Gather columns into key-value pairs (superseded by `pivot_longer()`).
`spread()`	`->spread()`	Spread a key-value pair across columns (superseded by `pivot_wider()`).

Missing combinations and values

Function	Chain method	Purpose
`complete()`	`->complete()`	Add missing combinations, filling other columns with NA (or with fill).
`expand()`	`->expand()`	Every combination of the given columns, per group.
`nesting()`		Only the combinations that occur in the data, for `complete()` and `expand()`.
`crossing()`		Every combination of unique values (sorted), as a new table.
`expand_grid()`		Every combination of the values as given, first column varying slowest.
`full_seq()`		Every value from the smallest to the largest of x in steps of period.
`fill()`	`->fill()`	Fill NA with the previous (or next) value, within groups.
`drop_na()`	`->dropNa()`	Drop rows with NA in any of the columns (any column when none are given).
`replace_na()`	`->replaceNa()`	Replace NA per column.
`uncount()`	`->uncount()`	Repeat each row by a weight column.

Splitting and combining

Function	Chain method	Purpose
`separate_wider_delim()`	`->separateWiderDelim()`	Split columns on a fixed delimiter into new columns.
`separate_wider_position()`	`->separateWiderPosition()`	Split columns by character widths.
`separate_wider_regex()`	`->separateWiderRegex()`	Split columns with regular expressions.
`separate_longer_delim()`	`->separateLongerDelim()`	Split columns on a fixed delimiter into rows.
`separate_longer_position()`	`->separateLongerPosition()`	Split columns into rows of `$width` characters.
`separate()`	`->separate()`	Split a string column into several (superseded in tidyr, still common).
`separate_rows()`	`->separateRows()`	Split columns on a regular expression into rows (superseded by `separate_longer_delim()`).
`extract()`	`->extract()`	Pull regular-expression groups into new columns (superseded by `separate_wider_regex()`).
`unite()`	`->unite()`	Paste columns together into one.

List-columns

Function	Chain method	Purpose
`nest()`	`->nest()`	Nest columns into a list-column of tables.
`unnest()`	`->unnest()`	Unnest list-columns of tables back into rows.
`unnest_longer()`	`->unnestLonger()`	Turn each element of list-columns into rows.
`unnest_wider()`	`->unnestWider()`	Turn each element of list-columns into columns.
`hoist()`	`->hoist()`	Pull components out of a list-column into new columns, before it.
`chop()`	`->chop()`	Collapse rows into list-columns.
`unchop()`	`->unchop()`	Expand list-columns into rows, the reverse of `chop()`.
`pack()`	`->pack()`	Pack columns into a column of named arrays (tidyr’s data frame columns).
`unpack()`	`->unpack()`	Spread columns of named arrays into columns, the reverse of `pack()`.

Cleaning with janitor

Clean names

Function	Chain method	Purpose
`make_clean_names()`		janitor’s `make_clean_names()`.
`clean_names()`	`->cleanNames()`	Rename every column with `make_clean_names()`, as janitor’s `clean_names()`.

Remove empty rows and columns

Function	Chain method	Purpose
`remove_empty()`	`->removeEmpty()`	Drop rows and/or columns that are entirely NA, as janitor’s `remove_empty()`.

Duplicates

Function	Chain method	Purpose
`get_dupes()`	`->getDupes()`	Rows whose values in these columns (all columns when none are given) appear more than once, as janitor’s `get_dupes()`.

Frequency tables

Function	Chain method	Purpose
`tabyl()`	`->tabyl()`	janitor’s `tabyl()` for one or two columns.

Models

Function	Chain method	Purpose
`lm()`	`->lm()`	Linear regression.
`glm()`	`->glm()`	Generalised linear model.
`lmer()`	`->lmer()`	Linear mixed model with lme4-style random effects.
`glmer()`	`->glmer()`	Generalised linear mixed model.

Databases and expressions

Expressions

Function	Chain method	Purpose
`col()`		A column as a vector.
`lit()`		A literal value as a vector.
`vec()`		A row closure as a vector, for use where a Vec is expected.
`is_na()`		True where x is NA.
`abs()`		Absolute value.
`sqrt()`		Square root.
`exp()`		Exponential, e to the power x.
`log()`		Natural logarithm, or base `$base`.
`log10()`		Base-10 logarithm.
`round()`		R’s `round()`.
`floor()`		Largest integer not greater than x.
`ceiling()`		Smallest integer not less than x.
`as_int()`		Convert to integer.
`as_float()`		Convert to double.
`as_string()`		Convert to string.
`as_bool()`		Convert to boolean.

Database tables

Function	Chain method	Purpose
`show_query()`	`->showQuery()`	The SQL a lazy table would run (database backends only).
`collect()`	`->collect()`	Run a lazy table’s query and return an in-memory Phibble.

Chain methods

phplyr($table) starts a chain. Each function above with a chain method runs as that method. These methods belong to the chain itself:

Method	Purpose
`->get()`	Return the table.
`->pipe()`	Apply your own functions to the table and continue the chain with the result.

Expression methods

Called on a vector helper, such as col('price')->mul(1.1).

Method	Purpose
`->add()`	x + y.
`->sub()`	x - y.
`->mul()`	x * y.
`->div()`	x / y.
`->mod()`	x modulo y, R’s %%.
`->pow()`	x to the power y.
`->neg()`	Negation, -x.
`->eq()`	x == y.
`->ne()`	x != y.
`->lt()`	x < y.
`->le()`	x <= y.
`->gt()`	x > y.
`->ge()`	x >= y.
`->between()`	low <= x <= high, inclusive.
`->isIn()`	True where x is one of the values, R’s %in%.
`->isNa()`	True where x is NA.
`->and()`	Logical and, with R’s NA rules.
`->or()`	Logical or, with R’s NA rules.
`->not()`	Logical not.
`->desc()`	Sort in descending order, for `arrange()` and order_by.