Detect and remove matched words in a char stream

You are processing a stream of characters (arriving one-by-one). You are also given a list of target words.

Each time a new character arrives, it is appended to the end of the current stream buffer. After appending, if the buffer contains any target word as a contiguous substring, you must:

1. Return/report a matched word (assume if multiple words match, you may return any one of them, but you must define and follow a consistent rule such as “return the first match found” or “return the longest match”).
2. Delete that occurrence of the matched word from the buffer (remove exactly those characters), then continue processing future characters.

Clarifications to make explicit in your solution

• Can matches overlap? If multiple matches exist simultaneously, which one do you remove?
• Do you remove only one word per incoming character, or repeatedly remove until no word remains?

Input/Output (one reasonable interface)

• Input: words: List[str] , and a sequence of calls query(c: char) .
• Output per call: the matched word (or empty string / null if no match). The internal buffer should reflect deletions.

Constraints (typical interview-scale)

• Number of words: up to ~10^4
• Total length of all words: up to ~10^5
• Stream length can be large (online processing required).