fix: MLE5Large GCS URL transform (#3)

* fix: MLE5Large GCS URL transform

* ci: PR compliance, tests

* ci: refactor compliance.yml

.github/workflows/compliance.yml

@@ -1,4 +1,4 @@
-name: "Compliance"
+name: "PR Compliance"
 
 on:
   pull_request_target:
@@ -11,5 +11,42 @@ permissions:
   pull-requests: write
 
 jobs:
+  prepare-matrix:
+    runs-on: ubuntu-latest
+    outputs:
+      matrix: ${{ steps.set-matrix.outputs.matrix }}
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v3
+      - name: Set matrix for test
+        id: set-matrix
+        run: echo "matrix=$(find . -name '*.test.ts' | jq -R -s -c 'split("\n") | map(select(length > 0))')" >> $GITHUB_OUTPUT
+
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        files: ${{fromJson(needs.prepare-matrix.outputs.matrix)}}
+    needs:
+      - prepare-matrix
+    steps:
+      - name: "☁️ checkout repository"
+        uses: actions/checkout@v3
+        with:
+          fetch-depth: 0
+
+      - name: "🔧 setup Bun"
+        uses: oven-sh/setup-bun@v1
+
+      - name: "📦 install dependencies"
+        run: bun install
+
+      - name: "Run Vitest"
+        run: bun x vitest ${{ matrix.files }}
+
   compliance:
-    uses: open-sauced/hot/.github/workflows/compliance.yml@main
+    uses: open-sauced/hot/.github/workflows/compliance.yml@main
+    needs: 
+      - test
+
+

src/fastembed.ts

@@ -18,7 +18,7 @@ export enum EmbeddingModel {
   AllMiniLML6V2 = "fast-all-MiniLM-L6-v2",
   BGEBaseEN = "fast-bge-base-en",
   BGESmallEN = "fast-bge-small-en",
-  MLE5Large = "intfloat-multilingual-e5-large",
+  MLE5Large = "fast-multilingual-e5-large",
 }
 
 interface InitOptions {
@@ -149,6 +149,13 @@ export class FlagEmbedding extends Embedding {
       return outputFilePath;
     }
 
+    // The MLE5Large model URL doesn't follow the same naming convention as the other models
+    // So, we tranform "fast-multilingual-e5-large" -> "intfloat-multilingual-e5-large" in the download URL
+    // The model directory name in the GCS storage is "fast-multilingual-e5-large", like the others
+    if (model === EmbeddingModel.MLE5Large) {
+      model = "intfloat" + model.substring(model.indexOf("-"));
+      console.log("It is:", model);
+    }
     const url = `https://storage.googleapis.com/qdrant-fastembed/${model}.tar.gz`;
     const fileStream = fs.createWriteStream(outputFilePath);