[Python] Count number of characters in each row and drop if all are below a certain number

I have a dataframe with many columns, all of which contain text data mixed with NaNs.

I want to count the number of characters in each cell in each column and then drop any rows where all the columns have less than 5 characters (if any cells have more than 5 characters, then the row is not dropped).

I was considering making a new column with str.len for each column and then filter out rows using that, but it seems very cumbersome.

Example:

>>> df
column_1 column_2 column_3
0 werhi dsfhjk dh ---> not filtered because some columns have more than 5 characters
1 sgds fuo g ---> filtered
2 wqyuio dsklh fhkjfj
3 fhi d fgho ---> filtered
4 sadfhkj sdjfkhs yyisdk

>>> df_filtered
column_1 column_2 column_3
0 werhio dsfhjk dh
2 wqyuio dsfjklh fhkjfj
4 sadfhkj sdjfkhs yyisdk

Continue reading...