Palindrome Checker

Where palindromes actually turn up

The best-known group of palindromes is the set of Russian phrases invented in the nineteenth century. "A rose fell on the lap of Azor" became the model of the genre: long, rhythmic, and unforgiving about letters and spaces. Nineteenth-century writers treated the genre as a failure, on the grounds that Russian simply could not produce one. They were half right: long Russian palindromes are nearly impossible without normalization, but with it they appear at once.

Palindromes live in engineering too. A palindromic DNA stretch is reverse-complementary to itself and serves as the point where replication restarts. In programming you run into palindromic patterns, numbers and strings, and in data analysis people look for them to understand the shape of a sequence. Vehicle registration numbers and phone numbers are a separate case: a symmetrical number looks tidier, and people pick it on purpose.

There are everyday uses that involve no science at all. A project name that reads the same from both ends is easier to remember. A riddle whose answer hides in a palindrome works on any audience. A barcode or an article number typed symmetrically is quicker to check by eye while someone is standing in a queue. The quickest way to try any of that is to use the example buttons under the form and watch them land in the input field.

• Long phrases like "step on no pets" or "no lemon, no melon"
• Names and handles you want to read from either end
• Palindromic DNA stretches in bioinformatics
• Symmetrical phone numbers and plate numbers
• Palindromes in computer science assignments and contests
• Puzzles and rebuses where the answer is hidden symmetrically

Why a palindrome does not always read the same

The main source of confusion is that normalization changes the very thing being checked. The string "Dovod dovod" contains a space, so a character-by-character comparison rejects it. Drop the space and you get "dovoddovod", and that is a palindrome. Formally you are looking at two different answers to one question, and which one is right depends on who is asking.

The second trap has to do with alphabets. Comparison is character-based, so "A" and "a" are different characters, and in Russian "yo" and "e" are different letters. A palindrome can fall apart over a single yo, and people rarely notice it by eye. If exact wording matters, keep the case rule on and run the check a second time with it off, so you know what the result actually rests on.

The third mistake is about language. An Arabic string read from right to left gives a different answer than a machine comparison going left to right. Arabic palindromes have traditionally been written and read in that direction, and the same text in Latin script looks different. This matters for signs, embroidery and inscriptions, where the reading direction is set by a person rather than by an algorithm.

• A space inside a phrase breaks symmetry under a strict comparison
• Case and the letter yo quietly wreck symmetry
• Reading direction for Arabic and Hebrew inscriptions
• An empty string is not a palindrome even though the algorithm will not reject it
• One symmetric word does not make the whole sentence symmetric
• A string of digits follows exactly the same rules

Walking through the algorithm on real examples

The algorithm keeps two pointers: one at the start of the string, one at the end. At each step the characters under them are compared, and once they differ there is no point in continuing, because that first mismatching pair already disproves the palindrome. If they match, both pointers move toward the centre. When they meet without finding a mismatch, the string is symmetric.

Walking through it this way shows where the answer comes from. The table below marks the exact position where the characters stopped agreeing. That position is the shortest stretch that makes the string fail, and everything before it does match. This is the level of detail you need when you have to explain a solution in a computer science class, not just print the verdict.

The check makes a single pass over the string, so long text does not slow the tool down.

Palindromes in numbers, in code and on labels

Numeric palindromes are easy to spot by eye: 121, 12321, 45654. There is arithmetic structure behind them too: the more digits a number has, the more of them qualify, and from five digits upwards there are dozens per hundred. That is why palindromes show up in brute-force exercises and in contest problems about divisibility, where the shape of the number gives you the idea.

In code, palindromes turn up as coincidences, and they are regularly spotted in logs, test fixtures and generated identifiers. When you run across one, look at the surroundings first: spaces, dashes and letter case explain most findings. It also helps to know how numbers behave in other bases, and for that there is a number base converter plus a longer piece on the binary code.

• Numbers: 121, 12321, 45654, recognizable without any algorithm
• Code and logs: unexpected symmetry usually comes from formatting
• Article numbers and barcodes: checkable by eye, without a program
• Names and handles: readable from both ends saves the explaining
• Random strings: if there is a match, run it again without normalization