browse

Helpers for browse, head, and tail magics

Parsing Stata code for browse magic

in_range

def in_range(
    stata_in_code, count
):

Return in-statement range

test_eq(in_range("", 100), (None, None))
test_eq(in_range("in 10", 100), (0, 10))
test_eq(in_range("in 1/10", 100), (0, 10))
test_eq(in_range("in F/10", 100), (0, 10))
test_eq(in_range("in 3/10", 100), (2, 10))
test_fail(in_range, args=("in 10", 5))
test_fail(in_range, args=("in 10/1", 100))

Negative indices are converted to positive, following Stata syntax:

test_eq(in_range("in -10/-1", 41), (31, 41))

Head/tail/browse magic helpers

source

parse_browse_magic

def parse_browse_magic(
    code
):

code = "le if year==1920 in 1/10, noformat"
test_eq(parse_browse_magic(code), 
        (None, ['le'], 'if year==1920', 'in 1/10', '', 'noformat')
       )

run_single("gen date = dofy(year)")
test_eq(parse_browse_magic("10 le* if date==mdy(1, 1, 1910), noformat"),
        (10, ['le'], 'if date==mdy(1, 1, 1910)', '', '', 'noformat')
       )
run_single("drop date")

source

get_df

def get_df(
    obs_range, var, stata_if_code, missingval, valuelabel, sformat
):

Head/tail magic helpers

source

headtail_df_params

def headtail_df_params(
    code, count, missing_config, tail:bool=False
):

test_eq(headtail_df_params("", 50, "."),
        ((range(0, 5), None, '', ".", True, True)))

test_eq(headtail_df_params("10 y le*, noformat", 50, "."),
        ((range(0, 10), ['year', 'le'], '', ".", True, False)))

test_eq(headtail_df_params("10 y le*, noformat", 50, ".", tail=True),
        ((range(40, 50), ['year', 'le'], '', ".", True, False)))

source

headtail_get_df

def headtail_get_df(
    obs_range, var, stata_if_code, missingval, valuelabel, sformat
):

headtail_get_df(*headtail_df_params('if year<1903', obs_count(), "."))

	year	le
1	1900	47.3
2	1901	.
3	1902	51.5

test_eq(headtail_get_df(*headtail_df_params('if year<1903', obs_count(), ".")),
        headtail_get_df(*headtail_df_params('if year<1903', obs_count(), ".", tail=True)),)

headtail_get_df(*headtail_df_params('if year>1910', obs_count(), "."))

	year	le
12	1911	52.6
13	1912	53.5
14	1913	52.5
15	1914	54.2
16	1915	54.5

Browse magic helpers

source

browse_df_params

def browse_df_params(
    code, count, missing_config
):

browse_df_params("", 10, 'pandas')

(None, None, '', nan, True, True)

browse_df_params("y le* if year<1910, noformat", 50, '.')

(None, ['year', 'le'], 'if year<1910', '.', True, False)

browse_df_params('in 1/5', 41, '.')

(range(0, 5), None, '', '.', True, True)

get_df(*browse_df_params('in 1/5', obs_count(), '.'))

	year	le
1	1900	47.3
2	1901	.
3	1902	51.5
4	1903	50.5
5	1904	47.6

source

set_ipydatagrid_height

def set_ipydatagrid_height():

source

display_df_as_ipydatagrid

def display_df_as_ipydatagrid(
    df, auto_height:bool=True
):

#set_ipydatagrid_height()
run_single('gen long_name_1234567 = ""')
run_single('gen long_str = "10000000000200000000003000000000040000000000500000000006000000000070000000000800000000009000000000010000000000"')

(41 missing values generated)

df = get_df(*browse_df_params(', noformat', obs_count(), '.'))
display_df_as_ipydatagrid(df)

#set_ipydatagrid_height()
run_single('sysuse auto, clear')

(1978 automobile data)

df = get_df(*browse_df_params('', obs_count(), '.'))
display_df_as_ipydatagrid(df, False)

Test output when ‘if’ condition implies no observations:

df = get_df(*browse_df_params('if 0', obs_count(), '.'))
display_df_as_ipydatagrid(df, False)

No observations.

Test output when no observations at all:

run_single('clear')
df = get_df(*browse_df_params('', obs_count(), '.'))
display_df_as_ipydatagrid(df, False)

No observations.

df = get_df(*browse_df_params('if 1', obs_count(), '.'))
display_df_as_ipydatagrid(df, False)

No observations.