dockerfile: oops

.dockerignore

@@ -1,6 +1,7 @@
 data/warcs/
 data/carrefour/
 */*.db*
+sqlite.db
 downloader/
 node_modules/
 */node_modules/

Dockerfile

@@ -4,6 +4,10 @@ WORKDIR /usr/src/app
 FROM base as build
 RUN apk add --no-cache nodejs npm
 RUN npm install --global pnpm
+COPY db-datos/package.json db-datos/package.json
+COPY sitio/package.json sitio/package.json
+COPY pnpm-lock.yaml pnpm-workspace.yaml .
+RUN cd sitio && pnpm install
 COPY . .
 COPY db-datos/drizzle .
 RUN cd sitio && \

db-datos/db.js

@@ -3,6 +3,13 @@ import Database from "better-sqlite3";
 import { drizzle } from "drizzle-orm/better-sqlite3";
 import { DB_PATH } from "./drizzle.config.js";
 import * as schema from "./schema.js";
+import { migrateDb } from "./migrate.js";
 
-export const sqlite = new Database(DB_PATH);
-export const db = drizzle(sqlite, { schema });
+/** @type {null | import("drizzle-orm/better-sqlite3").BetterSQLite3Database<schema>} */
+let db = null;
+export function getDb() {
+  const sqlite = new Database(DB_PATH);
+  db = drizzle(sqlite, { schema });
+  migrateDb(db);
+  return db;
+}

db-datos/migrate.js

@@ -1,21 +1,13 @@
 // @ts-check
-import Database from "better-sqlite3";
-import { join, dirname } from "node:path";
-import { drizzle } from "drizzle-orm/better-sqlite3";
 import { migrate } from "drizzle-orm/better-sqlite3/migrator";
 import * as schema from "./schema.js";
-import { DB_PATH } from "./drizzle.config.js";
+import { sql } from "drizzle-orm";
 
-const url = new URL(import.meta.url);
-export function migrateDb() {
-  const sqlite = new Database(DB_PATH);
-  const db = drizzle(sqlite, { schema });
-
-  migrate(db, { migrationsFolder: join(dirname(url.pathname), "drizzle") });
-  sqlite.exec(`
-pragma journal_mode = WAL;
-PRAGMA synchronous = NORMAL;
-`);
-
-  sqlite.close();
+/**
+ * @param {import("drizzle-orm/better-sqlite3").BetterSQLite3Database<schema>} db
+ */
+export function migrateDb(db) {
+  migrate(db, { migrationsFolder: "node_modules/db-datos/drizzle" });
+  db.run(sql`pragma journal_mode = WAL;`);
+  db.run(sql`PRAGMA synchronous = NORMAL;`);
 }

db-datos/urlHelpers.ts

@@ -1,25 +0,0 @@
-import { sql } from "drizzle-orm";
-import { db } from "./db.js";
-import { productoUrls } from "./schema.js";
-
-export function saveUrls(urls: string[]) {
-  db.transaction((tx) => {
-    const now = new Date();
-    const insertUrlTra = tx
-      .insert(productoUrls)
-      .values({
-        url: sql.placeholder("url"),
-        firstSeen: now,
-        lastSeen: now,
-      })
-      .onConflictDoUpdate({
-        target: productoUrls.url,
-        set: { lastSeen: now },
-      })
-      .prepare();
-
-    for (const href of urls) {
-      insertUrlTra.run({ url: href });
-    }
-  });
-}

scraper-rs/Cargo.lock

@@ -61,6 +61,21 @@ version = "0.2.16"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "0942ffc6dcaadf03badf6e6a2d0228460359d5e34b57ccdc720b7382dfbd5ec5"
 
+[[package]]
+name = "android-tzdata"
+version = "0.1.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e999941b234f3131b00bc13c22d06e8c5ff726d1b6318ac7eb276997bbb4fef0"
+
+[[package]]
+name = "android_system_properties"
+version = "0.1.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "819e7219dbd41043ac279b19830f2efc897156490d7fd6ea916720117ee66311"
+dependencies = [
+ "libc",
+]
+
 [[package]]
 name = "anstream"
 version = "0.6.5"
@@ -227,6 +242,20 @@ version = "1.0.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
 
+[[package]]
+name = "chrono"
+version = "0.4.32"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "41daef31d7a747c5c847246f36de49ced6f7403b4cdabc807a97b5cc184cda7a"
+dependencies = [
+ "android-tzdata",
+ "iana-time-zone",
+ "js-sys",
+ "num-traits",
+ "wasm-bindgen",
+ "windows-targets 0.52.0",
+]
+
 [[package]]
 name = "clap"
 version = "4.4.15"
@@ -298,6 +327,17 @@ dependencies = [
  "cfg-if",
 ]
 
+[[package]]
+name = "cron"
+version = "0.12.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1ff76b51e4c068c52bfd2866e1567bee7c567ae8f24ada09fd4307019e25eab7"
+dependencies = [
+ "chrono",
+ "nom",
+ "once_cell",
+]
+
 [[package]]
 name = "deadpool"
 version = "0.10.0"
@@ -636,6 +676,29 @@ dependencies = [
  "tokio-rustls",
 ]
 
+[[package]]
+name = "iana-time-zone"
+version = "0.1.59"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b6a67363e2aa4443928ce15e57ebae94fd8949958fd1223c4cfc0cd473ad7539"
+dependencies = [
+ "android_system_properties",
+ "core-foundation-sys",
+ "iana-time-zone-haiku",
+ "js-sys",
+ "wasm-bindgen",
+ "windows-core",
+]
+
+[[package]]
+name = "iana-time-zone-haiku"
+version = "0.1.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f31827a206f56af32e590ba56d5d2d085f558508192593743f16b2306495269f"
+dependencies = [
+ "cc",
+]
+
 [[package]]
 name = "idna"
 version = "0.5.0"
@@ -745,6 +808,12 @@ version = "0.3.17"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "6877bb514081ee2a7ff5ef9de3281f14a4dd4bceac4c09388074a6b5df8a139a"
 
+[[package]]
+name = "minimal-lexical"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "68354c5c6bd36d73ff3feceb05efa59b6acb7626617f4962be322a825e61f79a"
+
 [[package]]
 name = "miniz_oxide"
 version = "0.7.1"
@@ -774,6 +843,16 @@ dependencies = [
  "rand 0.8.5",
 ]
 
+[[package]]
+name = "nom"
+version = "7.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d273983c5a657a70a3e8f2a01329822f3b8c8172b73826411a55751e404a0a4a"
+dependencies = [
+ "memchr",
+ "minimal-lexical",
+]
+
 [[package]]
 name = "nu-ansi-term"
 version = "0.46.0"
@@ -784,6 +863,15 @@ dependencies = [
  "winapi",
 ]
 
+[[package]]
+name = "num-traits"
+version = "0.2.17"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "39e3200413f237f41ab11ad6d161bc7239c84dcb631773ccd7de3dfe4b5c267c"
+dependencies = [
+ "autocfg",
+]
+
 [[package]]
 name = "num_cpus"
 version = "1.16.0"
@@ -1135,7 +1223,9 @@ version = "0.1.0"
 dependencies = [
  "again",
  "anyhow",
+ "chrono",
  "clap",
+ "cron",
  "deadpool",
  "deadpool-sqlite",
  "futures",
@@ -1688,6 +1778,15 @@ version = "0.4.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"
 
+[[package]]
+name = "windows-core"
+version = "0.52.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "33ab640c8d7e35bf8ba19b884ba838ceb4fba93a4e8c65a9059d08afcfc683d9"
+dependencies = [
+ "windows-targets 0.52.0",
+]
+
 [[package]]
 name = "windows-sys"
 version = "0.48.0"

scraper-rs/Cargo.toml

@@ -8,7 +8,9 @@ edition = "2021"
 [dependencies]
 again = "0.1.2"
 anyhow = "1.0.79"
+chrono = "0.4.32"
 clap = { version = "4.4.15", features = ["derive"] }
+cron = "0.12.0"
 deadpool = "0.10.0"
 deadpool-sqlite = "0.7.0"
 futures = "0.3.30"

scraper-rs/src/main.rs

@@ -1,5 +1,6 @@
 use again::RetryPolicy;
 use clap::{Parser, ValueEnum};
+use cron::Schedule;
 use deadpool_sqlite::Pool;
 use futures::{future, stream, StreamExt};
 use nanoid::nanoid;
@@ -9,10 +10,10 @@ use std::{
     env::{self},
     fs,
     path::PathBuf,
-    time::Duration,
+    str::FromStr,
+    time::{Duration, SystemTime, UNIX_EPOCH},
 };
 use thiserror::Error;
-use tokio::time;
 
 #[derive(ValueEnum, Clone, Debug)]
 enum Supermercado {
@@ -399,16 +400,24 @@ async fn auto_cli() -> anyhow::Result<()> {
     Ok(())
 }
 async fn cron_cli() -> anyhow::Result<()> {
-    let mut interval = time::interval(std::time::Duration::from_secs(60 * 60 * 24));
+    // https://crontab.guru
+    let schedule = Schedule::from_str("0 0 2 * * * *").unwrap();
+    // let schedule = Schedule::from_str("0 26 21 * * * *").unwrap();
 
     loop {
-        interval.tick().await;
-        tokio::spawn(auto_cli());
+        let t = schedule
+            .upcoming(chrono::Utc)
+            .next()
+            .unwrap()
+            .signed_duration_since(chrono::Utc::now())
+            .to_std()
+            .unwrap();
+        println!("Waiting for {:?}", t);
+        tokio::time::sleep(t).await;
+        auto_cli().await.unwrap();
     }
 }
 
-use std::time::{SystemTime, UNIX_EPOCH};
-
 mod sites;
 
 #[derive(Debug)]

sitio/src/lib/server/db.ts

@@ -1,4 +1,2 @@
-export { db } from "db-datos/db.js";
+export { getDb } from "db-datos/db.js";
 export * as schema from "db-datos/schema.js";
-import { migrateDb } from "db-datos/migrate.js";
-migrateDb();

sitio/src/routes/+layout.server.ts

@@ -1,9 +1,10 @@
 import { countDistinct } from "drizzle-orm";
 import type { PageServerLoad } from "./$types";
-import { db, schema } from "$lib/server/db";
+import { getDb, schema } from "$lib/server/db";
 const { precios } = schema;
 
 export const load: PageServerLoad = async () => {
+  const db = await getDb();
   const nProductosR = await db
     .select({
       count: countDistinct(precios.ean),

sitio/src/routes/+page.server.ts

@@ -1,11 +1,12 @@
 import type { PageData, PageServerLoad } from "./$types";
-import { db, schema } from "$lib/server/db";
+import { getDb, schema } from "$lib/server/db";
 const { precios } = schema;
 import { sql } from "drizzle-orm";
 
 let cache: Promise<{ key: Date; data: { precios: Precios } }> = doQuery();
 
 async function doQuery() {
+  const db = await getDb();
   const q = db
     .select({
       ean: precios.ean,

sitio/src/routes/ean/[ean]/+page.server.ts

@@ -1,10 +1,11 @@
 import { error } from "@sveltejs/kit";
 import { eq, max } from "drizzle-orm";
 import type { PageServerLoad } from "./$types";
-import { db, schema } from "$lib/server/db";
+import { getDb, schema } from "$lib/server/db";
 const { precios } = schema;
 
 export const load: PageServerLoad = async ({ params }) => {
+  const db = await getDb();
   const q = db
     .select()
     .from(precios)

sitio/src/routes/search/+page.server.ts

@@ -1,9 +1,10 @@
 import { error } from "@sveltejs/kit";
 import { sql } from "drizzle-orm";
 import type { PageServerLoad } from "./$types";
-import { db } from "$lib/server/db";
+import { getDb } from "$lib/server/db";
 
 export const load: PageServerLoad = async ({ url }) => {
+  const db = await getDb();
   const query = url.searchParams.get("q");
   let results: null | { ean: string; name: string; imageUrl: string }[] = null;
   if (query) {

tsconfig.json

@@ -17,5 +17,5 @@
     "forceConsistentCasingInFileNames": true
   },
   "include": ["**/*.ts", "**/*.js"],
-  "exclude": ["sitio/build"]
+  "exclude": ["./scraper-rs", "data"]
 }