rows

In [1]:

%run nb_helpers.py
from datar.all import *

nb_header(
    rows_insert, 
    rows_update, 
    rows_patch, 
    rows_upsert, 
    rows_delete, 
    book='rows'
)

%run nb_helpers.py from datar.all import * nb_header( rows_insert, rows_update, rows_patch, rows_upsert, rows_delete, book='rows' )

Try this notebook on binder.

★ rows_insert

¶

Insert rows from y into x¶

See original API
https://dplyr.tidyverse.org/reference/rows.html

Args:¶

  x: A data frame
  y: A data frame
  by: An unnamed character vector giving the key columns.
    The key columns must exist in both x and y.
    Keys typically uniquely identify each row, but this is only
    enforced for the key values of y
    By default, we use the first column in y, since the first column is
    a reasonable place to put an identifier variable.

  conflict: How to handle conflicts
    - "error": Throw an error

    - "ignore": Ignore conflicts

  copy: If x and y are not from the same data source, and copy is TRUE,
    then y will be copied into the same src as x.
    This allows you to join tables across srcs, but it is a potentially
    expensive operation so you must opt into it.

  in_place: Should x be modified in place?
    This may not be supported, depending on the backend implementation.

Returns:¶

  A data frame with all existing rows and potentially new rows

★ rows_update

¶

Update rows in x with values from y¶

See original API
https://dplyr.tidyverse.org/reference/rows.html

Args:¶

  x: A data frame
  y: A data frame
  by: An unnamed character vector giving the key columns.
    The key columns must exist in both x and y.
    Keys typically uniquely identify each row, but this is only
    enforced for the key values of y
    By default, we use the first column in y, since the first column is
    a reasonable place to put an identifier variable.

  unmatched: how should keys in y that are unmatched by the keys
    in x be handled?
    One of -
    "error", the default, will error if there are any keys in y that
    are unmatched by the keys in x.
    "ignore" will ignore rows in y with keys that are unmatched
    by the keys in x.

  copy: If x and y are not from the same data source, and copy is TRUE,
    then y will be copied into the same src as x.
    This allows you to join tables across srcs, but it is a potentially
    expensive operation so you must opt into it.

  in_place: Should x be modified in place?
    This may not be supported, depending on the backend implementation.

Returns:¶

  A data frame with all existing rows and potentially new rows

★ rows_patch

¶

Patch rows in x with values from y¶

See original API
https://dplyr.tidyverse.org/reference/rows.html

Args:¶

  x: A data frame
  y: A data frame
  by: An unnamed character vector giving the key columns.
    The key columns must exist in both x and y.
    Keys typically uniquely identify each row, but this is only
    enforced for the key values of y
    By default, we use the first column in y, since the first column is
    a reasonable place to put an identifier variable.

  unmatched: how should keys in y that are unmatched by the keys
    in x be handled?
    One of -
    "error", the default, will error if there are any keys in y that
    are unmatched by the keys in x.
    "ignore" will ignore rows in y with keys that are unmatched
    by the keys in x.

  copy: If x and y are not from the same data source, and copy is TRUE,
    then y will be copied into the same src as x.
    This allows you to join tables across srcs, but it is a potentially
    expensive operation so you must opt into it.

  in_place: Should x be modified in place?
    This may not be supported, depending on the backend implementation.

Returns:¶

  A data frame with NA values overwritten and the number of rows preserved

★ rows_upsert

¶

Upsert rows in x with values from y¶

See original API
https://dplyr.tidyverse.org/reference/rows.html

Args:¶

  x: A data frame
  y: A data frame
  by: An unnamed character vector giving the key columns.
    The key columns must exist in both x and y.
    Keys typically uniquely identify each row, but this is only
    enforced for the key values of y
    By default, we use the first column in y, since the first column is
    a reasonable place to put an identifier variable.

  copy: If x and y are not from the same data source, and copy is TRUE,
    then y will be copied into the same src as x.
    This allows you to join tables across srcs, but it is a potentially
    expensive operation so you must opt into it.

  in_place: Should x be modified in place?
    This may not be supported, depending on the backend implementation.

Returns:¶

  A data frame with inserted or updated depending on whether or not
  the key value in y already exists in x. Key values in y must be unique.

★ rows_delete

¶

Delete rows in x that match keys in y¶

See original API
https://dplyr.tidyverse.org/reference/rows.html

Args:¶

  x: A data frame
  y: A data frame
  by: An unnamed character vector giving the key columns.
    The key columns must exist in both x and y.
    Keys typically uniquely identify each row, but this is only
    enforced for the key values of y
    By default, we use the first column in y, since the first column is
    a reasonable place to put an identifier variable.

  unmatched: how should keys in y that are unmatched by the keys
    in x be handled?
    One of -
    "error", the default, will error if there are any keys in y that
    are unmatched by the keys in x.
    "ignore" will ignore rows in y with keys that are unmatched
    by the keys in x.

  copy: If x and y are not from the same data source, and copy is TRUE,
    then y will be copied into the same src as x.
    This allows you to join tables across srcs, but it is a potentially
    expensive operation so you must opt into it.

  in_place: Should x be modified in place?
    This may not be supported, depending on the backend implementation.

Returns:¶

  A data frame with rows deleted

In [2]:

data = tibble(a = seq(1, 3), b = c(letters[[0, 1]], NA), c = [.5, 1.5, 2.5])
data

data = tibble(a = seq(1, 3), b = c(letters[[0, 1]], NA), c = [.5, 1.5, 2.5]) data

Out[2]:

	a	b	c
	<int64>	<object>	<float64>
0	1	a	0.5
1	2	b	1.5
2	3	NaN	2.5

In [3]:

rows_insert(data, tibble(a = 4, b = "z"))

rows_insert(data, tibble(a = 4, b = "z"))

[2022-12-02 14:22:14][datar][   INFO] Matching, by='a'

Out[3]:

	a	b	c
	<int64>	<object>	<float64>
0	1	a	0.5
1	2	b	1.5
2	3	NaN	2.5
3	4	z	NaN

In [4]:

with try_catch():
    rows_insert(data, tibble(a = 3, b = "z"))

with try_catch(): rows_insert(data, tibble(a = 3, b = "z"))

[2022-12-02 14:22:14][datar][   INFO] Matching, by='a'

[ValueError] Attempting to insert duplicate rows.

In [5]:

rows_update(data, tibble(a = [2,3], b = "z"))

rows_update(data, tibble(a = [2,3], b = "z"))

[2022-12-02 14:22:15][datar][   INFO] Matching, by='a'

Out[5]:

	a	b	c
	<int64>	<object>	<float64>
0	1	a	0.5
1	2	z	1.5
2	3	z	2.5

In [6]:

rows_update(data, tibble(b = "z", a = [2,3]), by = "a")

rows_update(data, tibble(b = "z", a = [2,3]), by = "a")

Out[6]:

	a	b	c
	<int64>	<object>	<float64>
0	1	a	0.5
1	2	z	1.5
2	3	z	2.5

In [7]:

rows_patch(data, tibble(a = [2,3], b = "z"))

rows_patch(data, tibble(a = [2,3], b = "z"))

[2022-12-02 14:22:17][datar][   INFO] Matching, by='a'

Out[7]:

	a	b	c
	<int64>	<object>	<float64>
0	1	a	0.5
1	2	b	1.5
2	3	z	2.5

In [8]:

rows_upsert(data, tibble(a = seq(2, 4), b = "z"))

rows_upsert(data, tibble(a = seq(2, 4), b = "z"))

[2022-12-02 14:22:18][datar][   INFO] Matching, by='a'

Out[8]:

	a	b	c
	<int64>	<object>	<float64>
0	1	a	0.5
1	2	z	1.5
2	3	z	2.5
3	4	z	NaN

In [9]:

rows_delete(data, tibble(a = [2, 3]))

rows_delete(data, tibble(a = [2, 3]))

[2022-12-02 14:22:18][datar][   INFO] Matching, by='a'

Out[9]:

	a	b	c
	<int64>	<object>	<float64>
0	1	a	0.5

In [10]:

rows_delete(data, tibble(a = [2, 3], b = "b"))

rows_delete(data, tibble(a = [2, 3], b = "b"))

[2022-12-02 14:22:19][datar][   INFO] Matching, by='a'
[2022-12-02 14:22:19][datar][   INFO] Ignoring extra columns: ['b']

Out[10]:

	a	b	c
	<int64>	<object>	<float64>
0	1	a	0.5

In [11]:

with try_catch():
    rows_delete(data, tibble(a = [2,3], b = "b"), by = c("a", "b"))

with try_catch(): rows_delete(data, tibble(a = [2,3], b = "b"), by = c("a", "b"))

[ValueError] Attempting to delete missing rows.