How to find a word with letters in specific places within a dataframe

I am trying to find words with letters in specific positions within my dataframe. My dataframe is a list of all 5 letter words in English in all lower-case and no special characters (i.e. only alpha characters).

df = list of 5 letter words

word = column of words

Code:

firstLetter = input('First Letter = ')
secondLetter = input('Second Letter = ')
thirdLetter = input('Third Letter = ')
fourthLetter = input('Fourth Letter = ')
fifthLetter = input('Fifth Letter = ')
total = str(firstLetter) str(secondLetter) str(thirdLetter) str(fourthLetter) str(fifthLetter)
df[df['word'].str.contains(total)]['word']

This will find all words that contain user inputed letters in the order inputed. While useful, it's not exactly what I'd like to do. How would I search for words that only contain letters in specific positions and print that list. For example:

First letter = t
Second Letter = r
Third Letter = 
Fourth Letter = i
Fifth letter = n

Out: Train

I'm quite new to both python and jupyter and I would like to thank you for your help in advance.

CodePudding user response：

The most reasonable to me seems to use a fullmatch regex:

# replace the character you inputed for unknown with "."
# Assuming space here
l1 = firstLetter.replace(' ', '.')
l2 = secondLetter.replace(' ', '.')
# same for all letters...

m = df['word'].fullmatch(f'{l1}{l2}{l3}{l4}{l5}')


df.loc[m]

Ideally you could even input the regex directly:

regex = input('enter the pattern with "." for unknown letters: ')
# example tr.in

m = df['word'].fullmatch(regex)

df.loc[m]

CodePudding user response：

Here is necessary return Trues if no value is typing in input (empty string), so mask for test values by positions is:

firstLetter = input('First Letter = ')
secondLetter = input('Second Letter = ')
thirdLetter = input('Third Letter = ')
fourthLetter = input('Fourth Letter = ')
fifthLetter = input('Fifth Letter = ')

m1 =  df['word'].str[0].eq(firstLetter) | (not bool(firstLetter))
m2 =  df['word'].str[1].eq(secondLetter) | (not bool(secondLetter))
m3 =  df['word'].str[2].eq(thirdLetter) | (not bool(thirdLetter))
m4 =  df['word'].str[3].eq(fourthLetter) | (not bool(fourthLetter))
m5 =  df['word'].str[4].eq(fifthLetter) | (not bool(fifthLetter))

s = df.loc[m1 & m2 & m3 & m4 & m5, 'word']

Or is possible create more general solution from above:

firstLetter = input('First Letter = ')
secondLetter = input('Second Letter = ')
thirdLetter = input('Third Letter = ')
fourthLetter = input('Fourth Letter = ')
fifthLetter = input('Fifth Letter = ')

tup = (firstLetter, secondLetter, thirdLetter, fourthLetter, fifthLetter)
m = [df['word'].str[i].eq(v) | (not bool(v)) for i, v in enumerate(tup)]

s = df.loc[np.logical_and.reduce(m), 'word']

Test:

print (df)
    word
0  train
1  yrasn

firstLetter = input('First Letter = ')
secondLetter = input('Second Letter = ')
thirdLetter = input('Third Letter = ')
fourthLetter = input('Fourth Letter = ')
fifthLetter = input('Fifth Letter = ')

First Letter = t

Second Letter = r

Third Letter = 

Fourth Letter = i

Fifth Letter = n

tup = (firstLetter, secondLetter, thirdLetter, fourthLetter, fifthLetter)
print (tup)
('t', 'r', '', 'i', 'n')

m = [df['word'].str[i].eq(v) | (not bool(v)) for i, v in enumerate(tup)]

s = df.loc[np.logical_and.reduce(m), 'word']
print (s)
0    train
Name: word, dtype: object

CodePudding user response：

L1=['magic','sweet','nails','squiz']
df=pd.DataFrame({'words':L1})

##Create a column for each of the letter . 

df['first']=df['words'].str[0]
df['second']=df['words'].str[1]
df['third']=df['words'].str[2]
df['fourth']=df['words'].str[3]
df['fifth']=df['words'].str[4]

display(df)

words   first   second  third   fourth  fifth
0   magic   m   a   g   i   c
1   sweet   s   w   e   e   t
2   nails   n   a   i   l   s
3   squiz   s   q   u   i   z

## Now you can filter this dataframe using query 
# example all words with 's' in first place : 

df.query('first=="s"')

## you can add multiple filters 

df.query('first=="s" and second=="t"')

CodePudding user response：

Try this, its easier to parse the data with list.

import re

wlist = ['crate', 'fight', 'aroma']

char = ['*','*','*','*','*']
char[0] = input('1 : ')
char[1] = input('2 : ')
char[2] = input('3 : ')
char[3] = input('4 : ')
char[4] = input('5 : ')
regex = ''.join([i if i !='' else r'\w' for i in char])
print([w for w in wlist if re.search(regex ,w)])