.dockerignore

@@ -12,6 +12,4 @@ Dockerfile
 *.warc.zst
 .git
 scraper/debug/
-*/target/
-scraper-rs/target/
-*.db*
+*/target/

.github/workflows/container.yml

@@ -11,7 +11,7 @@ env:
 jobs:
   check:
     name: chequear typescript del sitio
-    runs-on: ubicloud-standard-4
+    runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
       - uses: pnpm/action-setup@v3
@@ -31,7 +31,7 @@ jobs:
   build-and-push-sitio:
     name: Compilar contenedor del sitio
     needs: check
-    runs-on: buildjet-4vcpu-ubuntu-2204
+    runs-on: ubuntu-latest
     permissions:
       contents: read
       packages: write
@@ -59,13 +59,13 @@ jobs:
           push: true
           tags: ${{ steps.meta.outputs.tags }}
           labels: ${{ steps.meta.outputs.labels }}
-          cache-from: type=registry,ref=${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:latest
-          cache-to: type=inline
-          platforms: linux/amd64
+          cache-from: type=registry,ref=${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}/sitio:buildcache
+          cache-to: type=registry,ref=${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}/sitio:buildcache,mode=max
+          platforms: linux/amd64,linux/arm64
 
-  build-and-push-scraper-amd64:
-    name: "[amd64] oci:scraper"
-    runs-on: ubicloud-standard-16
+  build-and-push-scraper:
+    name: Compilar contenedor del scraper
+    runs-on: ubuntu-latest
     permissions:
       contents: read
       packages: write
@@ -91,38 +91,6 @@ jobs:
           push: true
           tags: ${{ steps.meta.outputs.tags }}
           labels: ${{ steps.meta.outputs.labels }}
-          cache-from: type=registry,ref=${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:latest
-          cache-to: type=inline
-          platforms: linux/amd64
-
-  build-and-push-scraper-arm64:
-    name: "[arm64] oci:scraper"
-    runs-on: ubicloud-standard-16-arm
-    permissions:
-      contents: read
-      packages: write
-    steps:
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-      - name: Log in to the Container registry
-        uses: docker/login-action@v3
-        with:
-          registry: ${{ env.REGISTRY }}
-          username: ${{ github.actor }}
-          password: ${{ secrets.GITHUB_TOKEN }}
-      - name: Extract metadata (tags, labels) for Docker
-        id: meta
-        uses: docker/metadata-action@v5
-        with:
-          images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}/scraper
-      - name: Build and push Docker image
-        uses: docker/build-push-action@v5
-        with:
-          context: "{{defaultContext}}:scraper-rs/"
-          file: Dockerfile
-          push: true
-          tags: ${{ steps.meta.outputs.tags }}
-          labels: ${{ steps.meta.outputs.labels }}
-          cache-from: type=registry,ref=${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:latest
-          cache-to: type=inline
-          platforms: linux/arm64
+          cache-from: type=registry,ref=${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}/scraper:buildcache
+          cache-to: type=registry,ref=${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}/scraper:buildcache,mode=max
+          platforms: linux/amd64,linux/arm64

Dockerfile

@@ -1,11 +1,12 @@
-FROM docker.io/node:20 AS base
+FROM cgr.dev/chainguard/wolfi-base AS base
 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 ./
+COPY pnpm-lock.yaml pnpm-workspace.yaml .
 RUN cd sitio && pnpm install
 COPY . .
 COPY db-datos/drizzle .
@@ -15,7 +16,7 @@ RUN cd sitio && \
 
 FROM base
 ENV NODE_ENV=production
-RUN apt-get update && apt-get install -y jq sqlite3 && apt-get clean
+RUN apk add --no-cache nodejs npm jq sqlite
 
 # Sitio
 COPY --from=build /usr/src/app/sitio/package.json package.real.json
@@ -27,4 +28,4 @@ COPY --from=build /usr/src/app/db-datos/drizzle drizzle
 ENV DB_PATH=/db/db.db
 EXPOSE 3000
 
-CMD ["node", "."]
+CMD ["node", "."]

scraper-rs/src/sites/coto.rs

@@ -82,7 +82,7 @@ pub async fn get_urls() -> anyhow::Result<Vec<String>> {
     let client = build_client();
     let initial = Url::parse("https://www.cotodigital3.com.ar/sitios/cdigi/browse?Nf=product.endDate%7CGTEQ+1.7032032E12%7C%7Cproduct.startDate%7CLTEQ+1.7032032E12&Nr=AND%28product.sDisp_200%3A1004%2Cproduct.language%3Aespa%C3%B1ol%2COR%28product.siteId%3ACotoDigital%29%29")?;
 
-    let page_size = 50;
+    let page_size = 100;
     let handles: Vec<Vec<String>> = stream::iter(0..29000 / page_size)
         .map(|i| {
             let mut u = initial.clone();