Develop Reference

How do I check if the next item in a string is the alphabetical successor of the one before? + Inverse

I'm trying to compress a string in a way that any sequence of letters in strict alphabetical order is swapped with the first letter plus the length of the sequence.
For example, the string "abcdefxylmno", would become: "a6xyl4"
Single letters that aren't in order with the one before or after just stay the way they are.
How do I check that two letters are successors (a,b) and not simply in alphabetical order (a,c)? And how do I keep iterating on the string until I find a letter that doesn't meet this requirement?
I'm also trying to do this in a way that makes it easier to write an inverse function (that given the result string gives me back the original one).
EDIT :
I've managed to get the function working, thanks to your suggestion of using the alphabet string as comparison; now I'm very much stuck on the inverse function: given "a6xyl4" expand it back into "abcdefxylmno".
After quite some time I managed to split the string every time there's a number and I made a function that expands a 2 char string, but it fails to work when I use it on a longer string:
from string import ascii_lowercase as abc
def subString(start,n):
L=[]
ind = abc.index(start)
newAbc = abc[ind:]
for i in range(len(newAbc)):
while i < n:
L.append(newAbc[i])
i+=1
res = ''.join(L)
return res
def unpack(S):
for i in range(len(S)-1):
if S[i] in abc and S[i+1] not in abc:
lett = str(S[i])
num = int(S[i+1])
return subString(lett,num)
def separate(S):
lst = []
for i in S:
lst.append(i)
for el in lst:
if el.isnumeric():
ind = lst.index(el)
lst.insert(ind+1,"-")
a = ''.join(lst)
L = a.split("-")
if S[-1].isnumeric():
L.remove(L[-1])
return L
else:
return L
def inverse(S):
L = separate(S)
for i in L:
return unpack(i)
Each of these functions work singularly, but inverse(S) doesn't output anything. What's the mistake?

You can use the ord() function which returns an integer representing the Unicode character. Sequential letters in alphabetical order differ by 1. Thus said you can implement a simple funtion:
def is_successor(a,b):
# check for marginal cases if we dont ensure
# input restriction somewhere else
if ord(a) not in range(ord('a'), ord('z')) and ord(a) not in range(ord('A'),ord('Z')):
return False
if ord(b) not in range(ord('a'), ord('z')) and ord(b) not in range(ord('A'),ord('Z')):
return False
# returns true if they are sequential
return ((ord(b) - ord(a)) == 1)
You can use chr(int) method for your reversing stage as it returns a string representing a character whose Unicode code point is an integer given as argument.

This builds on the idea that acceptable subsequences will be substrings of the ABC:
from string import ascii_lowercase as abc # 'abcdefg...'
text = 'abcdefxylmno'
stack = []
cache = ''
# collect subsequences
for char in text:
if cache + char in abc:
cache += char
else:
stack.append(cache)
cache = char
# if present, append the last sequence
if cache:
stack.append(cache)
# stack is now ['abcdef', 'xy', 'lmno']
# Build the final string 'a6x2l4'
result = ''.join(f'{s[0]}{len(s)}' if len(s) > 1 else s for s in stack)

Python algorithm in list

In a list of N strings, implement an algorithm that outputs the largest n if the entire string is the same as the preceding n strings. (i.e., print out how many characters in front of all given strings match).
My code:
def solution(a):
import numpy as np
for index in range(0,a):
if np.equal(a[index], a[index-1]) == True:
i += 1
return solution
else:
break
return 0
# Test code
print(solution(['abcd', 'abce', 'abchg', 'abcfwqw', 'abcdfg'])) # 3
print(solution(['abcd', 'gbce', 'abchg', 'abcfwqw', 'abcdfg'])) # 0

Some comments on your code:
There is no need to use numpy if it is only used for string comparison
i is undefined when i += 1 is about to be executed, so that will not run. There is no actual use of i in your code.
index-1 is an invalid value for a list index in the first iteration of the loop
solution is your function, so return solution will return a function object. You need to return a number.
The if condition is only comparing complete words, so there is no attempt to only compare a prefix.
A possible way to do this, is to be optimistic and assume that the first word is a prefix of all other words. Then as you detect a word where this is not the case, reduce the size of the prefix until it is again a valid prefix of that word. Continue like that until all words have been processed. If at any moment you find the prefix is reduced to an empty string, you can actually exit and return 0, as it cannot get any less than that.
Here is how you could code it:
def solution(words):
prefix = words[0] # if there was only one word, this would be the prefix
for word in words:
while not word.startswith(prefix):
prefix = prefix[:-1] # reduce the size of the prefix
if not prefix: # is there any sense in continuing?
return 0 # ...: no.
return len(prefix)

The description is somewhat convoluted but it does seem that you're looking for the length of the longest common prefix.
You can get the length of the common prefix between two strings using the next() function. It can find the first index where characters differ which will correspond to the length of the common prefix:
def maxCommon(S):
cp = S[0] if S else "" # first string is common prefix (cp)
for s in S[1:]: # go through other strings (s)
cs = next((i for i,(a,b) in enumerate(zip(s,cp)) if a!=b),len(cp))
cp = cp[:cs] # truncate to new common size (cs)
return len(cp) # return length of common prefix
output:
print(maxCommon(['abcd', 'abce', 'abchg', 'abcfwqw', 'abcdfg'])) # 3
print(maxCommon(['abcd', 'gbce', 'abchg', 'abcfwqw', 'abcdfg'])) # 0

Scatter palindrome - How to parse through a dictionary to figure out combinations of substrings that form a palindrome

I was given this HC problem at an interview. I was able to come up with what I'll call a brute force method.
Here is the problem statement:
Find all the scatter palindromes in a given string, "aabb". The
substrings can be scattered but rearranged to form a palindrome.
example: a, aa, aab, aabb, a, abb, b, bb, bba and b are the substrings
that satisfy this criteria.
My logic:
divide the str into substrings
counter = 0
if the len(substr) is even:
and substr == reverse(substr)
increment counter
else:
store all the number of occurrences of each str of substr in a dict
process dict somehow to figure out if it can be arranged into a palindrome
###### this is where I ran out of ideas ######
My code:
class Solution:
def countSubstrings(self, s: str) -> int:
n = len(s)
c=0
for i in range(0,n-1): #i=0
print('i:',i)
for j in range(i+1,n+1): #j=1
print('j',j)
temp = s[i:j]
if len(temp) == 1:
c+=1
# if the len of substring is even,
# check if the reverse of the string is same as the string
elif(len(temp)%2 == 0):
if (temp == temp[::-1]):
c+=1
print("c",c)
else:
# create a dict to check how many times
# each value has occurred
d = {}
for l in range(len(temp)):
if temp[l] in d:
d[temp[l]] = d[temp[l]]+1
else:
d[temp[l]] = 1
print(d)
return c
op = Solution()
op.countSubstrings('aabb')
By now, it must be obvious I'm a beginner. I'm sure there are better, more complicated ways to solve this. Some of my code is adapted from visleck's logic here and I wasn't able to follow the second half of it. If someone can explain it, that would be great as well.

As a partial answer, the test for a string being a scattered palindrome is simple: if the number of letters which occur an odd number of times is at most 1, it is a scattered palindrome. Otherwise it isn't.
It can be implemented like this:
from collections import Counter
def scattered_palindrome(s):
counts = Counter(s)
return len([count for count in counts.values() if count % 2 == 1]) <= 1
For example,
>>> scattered_palindrome('abb')
True
>>> scattered_palindrome('abbb')
False
Note that at no stage is it necessary to compare a string with its reverse. Also, note that I used a Counter object to keep track of the letter counts. This is a streamlined way of creating a dictionary-like collection of letter counts.

'List index out of range' according to leetcode. But works in IDLE

I'm working on a leetcode problem which asks me to find the longest common prefix, the same letters, among some strings. The language is python3. The problem is as followed. (https://leetcode.com/problems/longest-common-prefix/)
Write a function to find the longest common prefix string amongst an
array of strings.
If there is no common prefix, return an empty string "".
Example 1:
Input: ["flower","flow","flight"]
Output: "fl"
Example 2:
Input: ["dog","racecar","car"]
Output: ""
Explanation: There is no common prefix among the input strings.
I tried to get the longest string as it must contain all the common letters. To do so I try to get a list of length of the strings and get the longest length. Then I can use the longest length to get the longest string. But a 'list index out of range' bug occurs when I submit my code to leetcode.
ls.sort()
lmax = ls[len(ls)-1]
I don't see there is a list index range problem here as I have used this method before. Also, I copied the code to IDLE and it didn't report a bug.
The code below is just the first part of my code. The rest of them is just to check for the same letters.
I would be really grateful if anyone could help me with this problem. Thank you.
class Solution:
def longestCommonPrefix(self, x):
list1 = []
for element in x: #convert strings into lists to search for common letters
list1.append(list(element))
i = 0 #for searching longest string
m = 0 #for indicating letters in the longest string
n = 0 #for indicating other strings
p = 0 #for indicating letters in other strings
ls = [] #search for the longest string
for strs in list1:
ls.append(len(strs))
ls.sort()
lmax = ls[len(ls)-1] ###This is the step that the bug occurs.##
while i <= len(list1)-1: # To get the longest string.
if len(list1[i]) == lmax:
break
else:
i += 1
longestStr = list1[i]

I don't think the bug is where you think it is. The line of code lmax = ls[len(ls)-1] will only fail if the list ls is empty. Side note: to get the last element in a list, use a negative index, which starts from the end and counts backward: lmax = ls[-1]
So I would double check that ls contains strings. Perhaps when switching IDEs, the param x isn't being passed in correctly.

How to implement a brute force solution to "Finding first unique character in a string"

As described here:
https://leetcode.com/problems/first-unique-character-in-a-string/description/
I attempted one here but couldn't quite finish:
https://paste.pound-python.org/show/JuPLgdgqceMQYh5kk0Sf/
#Given a string, find the first non-repeating character in it and return it's index. If it doesn't exist, return -1.
#xamples:
#s = "leetcode"
#return 0.
#s = "loveleetcode",
#return 2.
#Note: You may assume the string contain only lowercase letters.
class Solution(object):
def firstUniqChar(self, s):
"""
:type s: str
:rtype: int
"""
for i in range(len(s)):
for j in range(i+1,len(s)):
if s[i] == s[j]:
break
#But now what. let's say i have complete loop of j where there's no match with i, how do I return i?
I'm ONLY interested in the brute force N^2 solution, nothing fancier. The idea in the above solution is to start a double loop, where inner loop searches for a match with the outer loop's char, and if there's match, break the inner loop and continue onto the next char on the outer loop.
But the question is, how do I handle when there's NO match, which is when I need to return the outer loop's index as the first unique one.
Can't quite figure out a graceful way to do it, and can handle edge case like a single char string.

Iterate over each char, and check if it appears in any of the following chars. We need to keep track of the characters we've already seen, to avoid falling into edge cases. Try this, it's an O(n^2) solution:
def firstUniqChar(s):
# store already seen chars
seen = []
for i, c in enumerate(s):
# return if char not previously seen and not in rest
if c not in seen and c not in s[i+1:]:
return i
# mark char as seen
seen.append(c)
# no unique chars were found
return -1
For completeness' sake, here's an O(n) solution:
def firstUniqChar(s):
# build frequency table
freq = {}
for i, c in enumerate(s):
if c not in freq:
# store [frequency, index]
freq[c] = [1, i]
else:
# update frequency
freq[c][0] += 1
# find leftmost char with frequency == 1
# it's more efficient to traverse the freq table
# instead of the (potentially big) input string
leftidx = float('+inf')
for f, i in freq.values():
if f == 1 and i < leftidx:
leftidx = i
# handle edge case: no unique chars were found
return leftidx if leftidx != float('+inf') else -1
For example:
firstUniqChar('cc')
=> -1
firstUniqChar('ccdd')
=> -1
firstUniqChar('leetcode')
=> 0
firstUniqChar('loveleetcode')
=> 2

Add an else to the for loop where you return.
for j ...:
...
else:
return i

I'd first like to note that your current algorithm for finding unique characters doesn't work correctly. That's because you can't assume the character at index i is unique just because none of the indexes j found the same character later in the string. The character at index i could be a repeat of an earlier character (which you'd have skipped when the previous j was equal to the current i).
You could fix the algorithm by letting j iterate over the whole range of indexes, and adding an extra check to ignore the matches when the indexes are the same to your if:
for i in range(len(s)):
for j in range(len(s)):
if i != j and s[i] == s[j]:
break
As Ignacio Vazquez-Abrams suggests in his answer, you can then add an else block to the inner for loop to make the code return when no match was found:
else: # this line should be indented to match the "for j" loop
return i
There are also a few ways you can solve this problem more simply if you use the builtin functions and types available in Python.
For instance, you can implement an O(n^2) solution equivalent to the one above using only one explicit loop, and using str.count to replace the inner one:
def firstUniqChar(s):
for i, c in enumerate(s):
if s.count(c) == 1:
return i
return None
I'm also using enumerate to get the character values and indexes together in one step, rather than iterating over a range and indexing later.
There's also a very easy way to make an O(n) solution using collections.Counter, which can do all the counting in one pass before you start checking the characters in order to try to find the first one that is unique:
from collections import Counter
def firstUniqChar(s):
count = Counter(s)
for i, c in enumerate(s):
if count[c] == 1:
return i
return None

I'm not sure your approach will work on an even palindrome, e.g. "redder" (note the second d). Try this instead:
s1 = "leetcode"
s2 = "loveleetcode"
s3 = "redder"
def unique_index(s):
ahead, behind = list(s), set()
for idx, char in enumerate(s):
ahead = ahead[1:]
if (char not in ahead) and (char not in behind):
return idx
behind.add(s[idx])
return -1
assert unique_index(s1) == 0
assert unique_index(s2) == 2
assert unique_index(s3) == -1
For each character, we look ahead and behind. Only characters disjoint from both groups will return an index. As iteration progresses, the list of what is observed ahead shortens, while what is seen behind extends. The default is -1 as stated in the actual leetcode challenge.
A second list is not required. #Óscar López's answer is the simplified answer.