representation.text.paragraphMulti-sentence free-form text (descriptions, reviews, recipes, articles). Distinguished from sentence by containing multiple sentences (period followed by space or newline and another capital letter).
$ finetype infer -i "This product works great. I would recommend it to anyone looking for quality."
→ representation.text.paragraphSELECT finetype('This product works great. I would recommend it to anyone looking for quality.');
-- → 'representation.text.paragraph'CAST({col} AS VARCHAR)-- Normalise and cast in one step
SELECT TRY_CAST(finetype_cast(my_column) AS VARCHAR) AS clean_value
FROM my_table
WHERE finetype(my_column) = 'representation.text.paragraph';sentence_count: CAST(REGEXP_COUNT({col}, '[.!?]\s') + 1 AS SMALLINT)
word_count: CAST(REGEXP_COUNT({col}, '\s+') + 1 AS SMALLINT){
"$id": "https://meridian.online/schemas/representation.text.paragraph",
"$schema": "https://json-schema.org/draft/2020-12/schema",
"description": "Multi-sentence free-form text (descriptions, reviews, recipes, articles). Distinguished from sentence by containing multiple sentences (period followed by space or newline and another capital letter).",
"examples": [
"This product works great. I would recommend it to anyone looking for quality.",
"First, preheat the oven to 350F. Then mix the dry ingredients together. Add the wet ingredients slowly.",
"The hotel was located in the city center. Rooms were clean and spacious. Staff was very friendly and helpful."
],
"maxLength": 65536,
"minLength": 50,
"title": "Paragraph",
"type": "string",
"x-finetype-broad-type": "VARCHAR",
"x-finetype-transform": "CAST({col} AS VARCHAR)"
}This product works great. I would recommend it to anyone looking for quality.First, preheat the oven to 350F. Then mix the dry ingredients together. Add the wet ingredients slowly.The hotel was located in the city center. Rooms were clean and spacious. Staff was very friendly and helpful.