CODEWITHSUNDEEP
pythonregexpandas
sayth
sayth

Reputation: 7048

Pandas column replace $ and letter from rows

I have a column with currency values. All have a leading '$' and some have a trailing F. Trying to remove the charcters using replace with pandas.

My data is:

$4.00F
$21.00
$6.00
$6.00
$5.50
$13.00
$4.60F

This is my code using replace. However, it removes no characters at all.

df["Price"].replace("^/$$[F]", "", regex=True)

Where is my regex incorrect?

Upvotes: 1

Views: 90

Answers (3)

JvdV
JvdV

Reputation: 75900

In this instance you can simply avoid any regex and use .strip('$F'):

df['Price'] = df['Price'].str.strip('$F')

Upvotes: 1

Wiktor Stribiżew
Wiktor Stribiżew

Reputation: 627065

The ^/$$[F] pattern matches a string that starts with /, then makes sure the string ends right after / (twice) and then the regex engine searches for F (after the end of string). As you see, the pattern has no sense.

You can use

df["Price"] = df["Price"].str.replace(r"^\$|F$", "", regex=True)

This will remove a $ at the start and F at the end.

See the regex demo.

If you do not care where the $ and F are, you may use

df["Price"] = df["Price"].str.replace(r"[$F]+", "", regex=True)

See this regex demo where [$F]+ matches one or more F or $ chars.

Also, consider the classic pattern removing any chars other than digits and dots:

df["Price"] = df["Price"].str.replace(r"[^0-9.]+", "", regex=True)

See this regex demo.

Upvotes: 1

Corralien
Corralien

Reputation: 120469

If you want to remove any character (not just '$' and 'F') that is not a number at the start and the end of your string:

>>> df['Price'].str.replace(r'^[^\d+]|[^\d+]$', '', regex=True)
0     4.00
1    21.00
2     6.00
3     6.00
4     5.50
5    13.00
6     4.60
Name: Price, dtype: object

Upvotes: 0

Related Questions