Tokenize and Extract Patterns | Dataford Interview Questions

Given a string of unstructured text, write a function to tokenize and extract specific patterns without using heavy external libraries for Intelligent Medical Objects clinical text processing. The function receives text and a dictionary of named regular-expression patterns. Return word-like tokens and all matching substrings for each pattern, preserving their original order. Use only Python's standard library and return an empty list when a pattern has no matches.

Input: text is a string and patterns maps unique names to regular-expression strings. Output: return {"tokens": [...], "matches": {name: [...]}}.

Constraints

0 <= len(text) <= 100000
0 <= len(patterns) <= 20
Pattern names are unique strings
Each pattern is a valid Python regular-expression string
Tokens and matches preserve their original text and order

Function Signature

def tokenize_and_extract(text, patterns):

Your question is Tokenize and Extract Patterns. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem

Input: text is a string and patterns maps unique names to regular-expression strings. Output: return {"tokens": [...], "matches": {name: [...]}}.

Constraints

0 <= len(text) <= 100000
0 <= len(patterns) <= 20
Pattern names are unique strings
Each pattern is a valid Python regular-expression string
Tokens and matches preserve their original text and order

Function Signature

def tokenize_and_extract(text, patterns):

Your question is Tokenize and Extract Patterns. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.