scraper-rs: WIP: fetchear urls

scraper-rs/Cargo.lock

@@ -671,6 +671,15 @@ version = "2.9.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "8f518f335dce6725a761382244631d86cf0ccb2863413590b31338feb467f9c3"
 
+[[package]]
+name = "itertools"
+version = "0.12.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "25db6b064527c5d482d0423354fcd07a89a2dfe07b67892e62411946db7f07b0"
+dependencies = [
+ "either",
+]
+
 [[package]]
 name = "itoa"
 version = "1.0.10"
@@ -1156,6 +1165,7 @@ dependencies = [
  "anyhow",
  "async-channel",
  "clap",
+ "itertools",
  "nanoid",
  "r2d2",
  "r2d2_sqlite",

scraper-rs/Cargo.toml

@@ -10,6 +10,7 @@ again = "0.1.2"
 anyhow = "1.0.79"
 async-channel = "2.1.1"
 clap = { version = "4.4.15", features = ["derive"] }
+itertools = "0.12.0"
 nanoid = "0.4.0"
 r2d2 = "0.8.10"
 r2d2_sqlite = "0.23.0"

scraper-rs/src/main.rs

@@ -1,6 +1,6 @@
 use again::RetryPolicy;
 use async_channel::Receiver;
-use clap::Parser;
+use clap::{Parser, ValueEnum};
 use nanoid::nanoid;
 use r2d2::Pool;
 use r2d2_sqlite::SqliteConnectionManager;
@@ -14,10 +14,19 @@ use std::{
 };
 use thiserror::Error;
 
+#[derive(ValueEnum, Clone)]
+enum Supermercado {
+    Dia,
+    Jumbo,
+    Carrefour,
+    Coto,
+}
+
 #[derive(Parser)] // requires `derive` feature
 enum Args {
     FetchList(FetchListArgs),
     ParseFile(ParseFileArgs),
+    GetUrlList(GetUrlListArgs),
 }
 #[derive(clap::Args)]
 struct FetchListArgs {
@@ -27,6 +36,11 @@ struct FetchListArgs {
 struct ParseFileArgs {
     file_path: String,
 }
+#[derive(clap::Args)]
+struct GetUrlListArgs {
+    #[arg(value_enum)]
+    supermercado: Supermercado,
+}
 
 #[tokio::main]
 async fn main() -> anyhow::Result<()> {
@@ -35,6 +49,7 @@ async fn main() -> anyhow::Result<()> {
     match Args::parse() {
         Args::FetchList(a) => fetch_list_cli(a.list_path).await,
         Args::ParseFile(a) => parse_file_cli(a.file_path).await,
+        Args::GetUrlList(a) => get_url_list_cli(a.supermercado).await,
     }
 }
 
@@ -139,9 +154,19 @@ enum FetchError {
     Tl(#[from] tl::ParseError),
 }
 
-async fn do_request(client: &reqwest::Client, url: &str) -> reqwest::Result<reqwest::Response> {
+pub async fn do_request(client: &reqwest::Client, url: &str) -> anyhow::Result<reqwest::Response> {
     let request = client.get(url).build()?;
-    client.execute(request).await
+    let response = client.execute(request).await?;
+    if !response.status().is_success() {
+        bail!(FetchError::HttpStatus(response.status()));
+    }
+    Ok(response)
+}
+
+pub fn get_retry_policy() -> again::RetryPolicy {
+    RetryPolicy::exponential(Duration::from_millis(300))
+        .with_max_retries(10)
+        .with_jitter(true)
 }
 
 #[tracing::instrument(skip(client))]
@@ -149,18 +174,12 @@ async fn fetch_and_parse(
     client: &reqwest::Client,
     url: String,
 ) -> Result<PrecioPoint, anyhow::Error> {
-    let policy = RetryPolicy::exponential(Duration::from_millis(300))
+    let body = get_retry_policy()
-        .with_max_retries(10)
-        .with_jitter(true);
-
-    let response = policy
         .retry(|| do_request(client, &url))
+        .await?
+        .text()
         .await
         .map_err(FetchError::Http)?;
-    if !response.status().is_success() {
-        bail!(FetchError::HttpStatus(response.status()));
-    }
-    let body = response.text().await.map_err(FetchError::Http)?;
 
     let maybe_point = { scrap_url(client, url, &body).await };
 
@@ -202,6 +221,20 @@ async fn parse_file_cli(file_path: String) -> anyhow::Result<()> {
     Ok(())
 }
 
+async fn get_url_list_cli(supermercado: Supermercado) -> anyhow::Result<()> {
+    let urls = match supermercado {
+        Supermercado::Dia => sites::dia::get_urls().await?,
+        Supermercado::Jumbo => sites::jumbo::get_urls().await?,
+        Supermercado::Carrefour => sites::carrefour::get_urls().await?,
+        _ => todo!(),
+    };
+    urls.iter().for_each(|s| {
+        println!("{}", s);
+    });
+
+    Ok(())
+}
+
 async fn scrap_url(
     client: &reqwest::Client,
     url: String,

scraper-rs/src/sites/carrefour.rs

@@ -66,3 +66,19 @@ pub fn parse(url: String, dom: &tl::VDom) -> Result<PrecioPoint, anyhow::Error>
         url,
     })
 }
+
+pub async fn get_urls() -> anyhow::Result<Vec<String>> {
+    let urls = vec![
+        "https://www.carrefour.com.ar/sitemap/product-0.xml",
+        "https://www.carrefour.com.ar/sitemap/product-1.xml",
+        "https://www.carrefour.com.ar/sitemap/product-2.xml",
+        "https://www.carrefour.com.ar/sitemap/product-3.xml",
+        "https://www.carrefour.com.ar/sitemap/product-4.xml",
+        "https://www.carrefour.com.ar/sitemap/product-5.xml",
+        "https://www.carrefour.com.ar/sitemap/product-6.xml",
+        "https://www.carrefour.com.ar/sitemap/product-7.xml",
+        "https://www.carrefour.com.ar/sitemap/product-8.xml",
+        "https://www.carrefour.com.ar/sitemap/product-9.xml",
+    ];
+    vtex::get_urls_from_sitemap(&urls).await
+}

scraper-rs/src/sites/dia.rs

@@ -4,6 +4,7 @@ use simple_error::bail;
 use crate::sites::common;
 use crate::PrecioPoint;
 
+use super::vtex;
 use super::vtex::find_product_ld;
 use super::vtex::AvailabilityLd;
 
@@ -39,3 +40,14 @@ pub fn parse(url: String, dom: &tl::VDom) -> Result<PrecioPoint, anyhow::Error>
         url,
     })
 }
+
+pub async fn get_urls() -> anyhow::Result<Vec<String>> {
+    let urls = vec![
+        "https://diaonline.supermercadosdia.com.ar/sitemap/product-1.xml",
+        "https://diaonline.supermercadosdia.com.ar/sitemap/product-2.xml",
+        "https://diaonline.supermercadosdia.com.ar/sitemap/product-3.xml",
+        "https://diaonline.supermercadosdia.com.ar/sitemap/product-4.xml",
+        "https://diaonline.supermercadosdia.com.ar/sitemap/product-5.xml",
+    ];
+    vtex::get_urls_from_sitemap(&urls).await
+}

scraper-rs/src/sites/jumbo.rs

@@ -90,3 +90,25 @@ pub async fn scrap(
         url,
     })
 }
+
+pub async fn get_urls() -> anyhow::Result<Vec<String>> {
+    // de https://www.jumbo.com.ar/sitemap.xml
+    let urls = vec![
+        "https://www.jumbo.com.ar/sitemap/product-1.xml",
+        "https://www.jumbo.com.ar/sitemap/product-10.xml",
+        "https://www.jumbo.com.ar/sitemap/product-11.xml",
+        "https://www.jumbo.com.ar/sitemap/product-12.xml",
+        "https://www.jumbo.com.ar/sitemap/product-13.xml",
+        "https://www.jumbo.com.ar/sitemap/product-14.xml",
+        "https://www.jumbo.com.ar/sitemap/product-15.xml",
+        "https://www.jumbo.com.ar/sitemap/product-2.xml",
+        "https://www.jumbo.com.ar/sitemap/product-3.xml",
+        "https://www.jumbo.com.ar/sitemap/product-4.xml",
+        "https://www.jumbo.com.ar/sitemap/product-5.xml",
+        "https://www.jumbo.com.ar/sitemap/product-6.xml",
+        "https://www.jumbo.com.ar/sitemap/product-7.xml",
+        "https://www.jumbo.com.ar/sitemap/product-8.xml",
+        "https://www.jumbo.com.ar/sitemap/product-9.xml",
+    ];
+    vtex::get_urls_from_sitemap(&urls).await
+}

scraper-rs/src/sites/vtex.rs

@@ -1,8 +1,11 @@
 use anyhow::{bail, Context};
+use itertools::Itertools;
 use serde::Deserialize;
 use simple_error::SimpleError;
 use tl::VDom;
 
+use crate::{build_client, do_request, get_retry_policy};
+
 use super::common;
 
 pub fn parse_script_json(dom: &VDom, varname: &str) -> Result<serde_json::Value, anyhow::Error> {
@@ -100,3 +103,30 @@ pub fn in_stock_from_meta(dom: &VDom) -> anyhow::Result<bool> {
         },
     )
 }
+
+pub fn parse_urls_from_sitemap(sitemap: &str) -> anyhow::Result<Vec<String>> {
+    let dom = tl::parse(sitemap, tl::ParserOptions::default())?;
+    Ok(dom
+        .query_selector("loc")
+        .unwrap()
+        .filter_map(|h| h.get(dom.parser()))
+        .filter_map(|n| n.as_tag())
+        .map(|t| t.inner_text(dom.parser()))
+        .map(|s| s.to_string())
+        .collect())
+}
+
+pub async fn get_urls_from_sitemap(sitemaps: &[&str]) -> anyhow::Result<Vec<String>> {
+    let mut total = vec![];
+    let client = &build_client();
+    for url in sitemaps {
+        let text = get_retry_policy()
+            .retry(|| do_request(client, url))
+            .await?
+            .text()
+            .await?;
+        let mut urls = parse_urls_from_sitemap(&text)?;
+        total.append(&mut urls);
+    }
+    Ok(total.into_iter().unique().collect())
+}