CSV Dataset Validation (Multi-part)

Problem: Multi-part CSV Dataset Validation You are given a multi-line CSV string csvText as input. You must parse it line by line and implement the following three parts in order. To simplify the CSV format: Rows are separated by \n. Columns are separated by commas ,. You do not need to handle quoted fields or escaped commas; a plain split is sufficient. Part 1: Empty-field validation Parse each row and each field (cell) and validate: No field is an empty string (length 0). Return false if any empty field exists; otherwise return true. Part 2: Banned-word validation Given a set bannedWords (assume case-sensitive matching unless specified otherwise). Assuming Part 1 passes, further validate: No field is exactly equal to any word in bannedWords. Return false if a banned word is found. Part 3: Stop-word counting on selected columns Given: stopWords: a set of stop words targetCols: a list of 0-based column indices to include Assuming Part 2 passes, for each row and each targetCols field: Tokenize the field by splitting on non-alphanumeric characters (e.g., delimiter regex [^A-Za-z0-9]+). Count how many tokens are contained in stopWords (count every occurrence). Return the total count as an integer. Constraints 1 <= rows <= 1e5 1 <= cols <= 100 total csvText length <= 1e7 You should process the input in a streaming / line-by-line manner to avoid building a huge 2D structure. Example csvText: hello,world foo, Part 1 returns false (the second row second field is empty). Example Input hello,world foo,bar Output Part1=true (no empty) Part2 depends on bannedWords Part3 depends on stopWords/targetCols