chore: parse csv and talk id for evaluation files

This commit is contained in:
2025-10-29 12:03:00 -03:00
parent 197642727c
commit ca406b4601

View File

@@ -2,12 +2,35 @@ use std::fmt::Debug;
use chrono::Datelike;
use itertools::Itertools;
use polars::prelude::buffer::validate_utf8;
use polars::prelude::*;
use reqwest;
use walkdir;
use std::time::Duration;
use std::env;
use csv;
#[derive(Debug, serde::Deserialize)]
struct CsvHeader {
CATEGORIA: String,
PONTOS: Option<u8>,
}
#[derive(Debug, serde::Deserialize)]
struct CsvEvaluation {
APRESENTAÇÃO: u8,
CONFIRMAÇÃO_DE_EMAIL: u8,
CONFIRMAÇÃO_DE_TELEFONE: u8,
PROTOCOLO: u8,
USO_DO_PORTUGUÊS: u8,
PACIÊNCIA_E_EDUCAÇÃO: u8,
DISPONIBILIDADE: u8,
CONHECIMENTO_TÉCNICO: u8,
DIDATISMO: u8,
ID_TALK: String
}
fn main() {
match dotenv::dotenv().ok() {
Some(_) => println!("Environment variables loaded from .env file"),
@@ -105,14 +128,15 @@ fn main() {
})
.take(1)
.map(|file_name_csv| {
println!("{:?}", file_name_csv.path());
let file_contents = std::fs::read_to_string(file_name_csv.path()).expect("Failed to read CSV file");
println!("{}", format!("{prompt_data_sanitization} \n{file_contents}"));
let ollama_api_request = client.post(format!("http://{OLLAMA_SANITIZED_IP}:{OLLAMA_PORT}/api/generate"))
.body(
serde_json::json!({
"model": OLLAMA_AI_MODEL_DATA_SANITIZATION,
"prompt": format!("{prompt_data_sanitization} \n{file_contents}"),
"temperature": 0.0, // Get predictable and reproducible output
"stream": false,
}).to_string()
);
@@ -125,13 +149,65 @@ fn main() {
let ai_response = response_json["response"]
.as_str()
.expect("Failed to get AI response as string");
println!("AI Response: {}", ai_response);
let ai_response = ai_response.to_string();
let ai_response = if let Some(resp) = ai_response.strip_prefix(" ").unwrap_or(&ai_response).strip_prefix("```csv\n") { resp.to_string() } else { ai_response };
let ai_response = if let Some(resp) = ai_response.strip_suffix(" ").unwrap_or(&ai_response).strip_suffix("```") { resp.to_string() } else { ai_response };
return Ok((ai_response, file_name_csv));
},
Err(error) => {println!("Error {error}");}
Err(error) => {println!("Error {error}"); return Err(error);}
};
})
.for_each(|value| println!("{:?}", value));
.filter_map_ok(|(ai_repsonse, file_path_csv)| {
let mut reader = csv::ReaderBuilder::new()
.has_headers(true)
.delimiter(b';')
.from_reader(ai_repsonse.as_bytes());
let mut deserialized_iter = reader.deserialize::<CsvHeader>();
let correctly_parsed = deserialized_iter.all(|value| {
value.is_ok() && value.unwrap().PONTOS.is_some()
});
if !correctly_parsed { return None; }
// Parse id talk from file_path
// filename example is: CC - Erraoander Quintana - 515578 - 20251020515578.csv
// id talk is the last information, so in the example is: 20251020515578
let regex_filename = regex::Regex::new(r"(CC - )((\w+\s*)+) - (\d+) - (\d+).csv").unwrap();
let filename = file_path_csv
.file_name()
.into_string()
.expect("Failed to convert file name as Rust &str");
let found_regex_groups_in_filename = regex_filename.captures(
filename.as_str()
).expect("Failed to do regex capture");
let talk_id = found_regex_groups_in_filename.get(5).expect("Failed to get the id from regex maches").clone();
println!("{:?}", talk_id);
// a.for_each(|val| { println!("{:?}", val)});
// println!("{:?}", a);
// Do validation on CSV parsed
// let records_parsed = reader.records();
// if records_parsed.try_len().expect("Failed to obtain lenght") != 9 { return None; }
// reader.records().for_each(|record| println!("{:?}", record));
return Some(());
})
.for_each(|_value| {});
// println!("{:?}", files_inside_folder_on_date);
return Some(());