Skip to content

pandas/tz-localize-without-dst

Rule tz-localize-without-dst
Pack pandas
Severity warning
Fix none, edit by hand
Library versions every version

tz_localize raises on a time that repeats or is skipped at a daylight saving change unless you say what those times become. Pass both ambiguous= and nonexistent=.

local = df["time"].dt.tz_localize("Europe/Copenhagen")
local = df["time"].dt.tz_localize("Europe/Copenhagen", ambiguous="infer", nonexistent="shift_forward")

The rule reports df["d"].dt.tz_localize("Europe/Copenhagen") without ambiguous= or without nonexistent=. A zone with daylight saving has an hour each year that happens twice and an hour that never happens. tz_localize raises on a time in either hour, unless ambiguous= and nonexistent= say what to do. On half-hourly data for 2024 in Europe/Copenhagen the call raised.

Zone Reported
a zone with daylight saving, such as "Europe/Copenhagen" yes
"UTC", "GMT", a zone under Etc/, or an offset such as "+01:00" no, because they did not raise
a zone held in a name no
a zone without daylight saving, such as "Asia/Tokyo" yes, by mistake

xpd has no automatic fix, so make the change by hand.

The table lists the values that the two arguments take.

Argument Value Result
ambiguous= "infer" reads the order of the rows, and needs both copies of the repeated hour
ambiguous= "NaT" makes the time missing
ambiguous= a boolean array says which rows are summer time
nonexistent= "shift_forward", "shift_backward", "NaT" or a Timedelta moves the time out of the missing hour, or makes it missing

xpd prints this message with the issue. A name that starts with $ stands for code from your file.

tz_localize raises for a time that repeats when the clocks go back unless ambiguous= is set, and for a time that is skipped when they go forward unless nonexistent= is set. Pass both to say what those times become.

Browse the other rules of the pandas pack.