Pandas check if a row contains a string rather than exact match [duplicate]

Question

I have the following function to check if a row within a DataFrame contains a string. This approach does work however it will only match if the provided string is exactly the same as what is in the DataFrame and I need it to match if it contains a string.

e.g. searching for 'fox' in 'a quick brown fox' will yield no return

def search_excel_files(file_list, search_term):
    #list of row indexes that contain the search term
    rows = {}
    for file in file_list:
        df = pd.read_excel("files/" + file)
        for row in df.iterrows():
            if search_term in row[1].values:
                #get row index
                row_index = row[0]
                #add row index to dictionary
                rows = df.iloc[row_index].to_dict()
    return rows

How can I check if the row contains the provided string in this instance?

pandas.pydata.org/docs/reference/api/… is of no use for you? — 9769953
– 9769953, Commented Jul 5, 2022 at 13:51
@9769953 if row[1].str.contains(search_term, regex=False).any(): returns the error Can only use .str accessor with string values! how can I access the value as a string? — Ivory
– Ivory, Commented Jul 5, 2022 at 14:15
Operate on the dataframe column in total: df['column_name'].str.contains(....). Your code iterates over the rows, thus operates on single values (row[1]). — 9769953
– 9769953, Commented Jul 5, 2022 at 14:47
@9769953 Ok that makes sense however I am still getting the same error of Can only use .str accessor with string values! when looping through the 'column_name' using for col in df.columns: any idea why that could be? — Ivory
– Ivory, Commented Jul 5, 2022 at 15:06
Because you are now doing this for all your columns (with the for loop)! If any of those columns is not of a string type (has a None, or is of integer type or floating point), then it obviously fails. That should be obvious from the error message. — 9769953
– 9769953, Commented Jul 5, 2022 at 17:17

PascalVKooten · Accepted Answer · 2022-07-05 13:52:10Z

1

It's better to think in columns than in rows when using pandas

df.your_col.str.contains("fox")

which will return an array of booleans, one bool for each row

Below you will get a dataframe where each row has fox in your_col column.

df[df.your_col.str.contains("fox")]

answered Jul 5, 2022 at 13:52

PascalVKooten

21.6k18 gold badges115 silver badges169 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

Michael Butscher Over a year ago

The "regex" argument should be set to "False" if regex search isn't wanted.

Ivory Over a year ago

@PascalVKooten in your example "your_col" how can I iterate through the columns? I am currently using IterRows however I do not see an equivalent for IterCols?

PascalVKooten Over a year ago

[x for x in df.columns]

Collectives™ on Stack Overflow

Pandas check if a row contains a string rather than exact match [duplicate]

1 Answer 1

3 Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Linked

Related