Creates a new ScrapeChain struct with the given chain, input_schema, and output_parser, you set up a scrapeChain with an input_schema and an output_parser, then you can ask it with whatever text you want.

example

Example:

create a chat to extract data:

chat = Chat.add_prompt_templates(%Chat{}, [ %{

role: "user",
prompt: %PromptTemplate{
  template: "Schema: """
  <%= input_schema %>
"""
Text: """
  <%= input_text %>
""    Extract the data from Text according to Schema and return it in <%= output_format %> format.
Format any datetime fields using ISO8601 standard.
"
}

} ])

create a ChainLink with the chat and parser function

chain_link = %ChainLink{ name: "schema_extractor", input: chat, output_parser: &schema_parser/2 } chain = %Chain{links: [chain_link]} input_schema = "{ name: String, age: Number, birthdate: Date }" schema_chain = LangChain.ScrapeChain.new(chain, input_schema)

default passthrough parser. 'result' will be a string so it is up to you to transform it into a native elixir structure or whatever you want.

Executes the scrapechain and returns the parsed result, can be called against the schema you defined when you made the chain, or you can override that schema: result = LangChain.ScrapeChain.scrape(schema_chain, "John Doe is 30 years old") # result will be %{ name: "John Doe", age: 30, birthdate: "1987-01-01"} # override the default schema input_variables = %{

input_text: "John Doe is 30 years old.",
input_schema: "{ firstName: String, lastName: String, age: Number, birthdate: Date }"

} alt_result = LangChain.ScrapeChain.scrape(schema_chain, input_variables) # alt_result will be %{ firstName: "John", lastName: "Doe", age: 30, birthdate: "1987-01-01"}