chore: Better VectorDb and Embedder error messages (#620)

* chore: propogate embedder and vectordb errors during document mutations * add default value for errors on addDocuments
2024-11-05 06:20:10 +01:00 · 2024-01-18 11:40:48 -08:00 · 2024-01-18 11:40:48 -08:00 · 658e7fa390
commit 658e7fa390
parent 0df86699e7
13 changed files with 77 additions and 47 deletions
--- a/frontend/src/index.css
+++ b/frontend/src/index.css
@ -403,3 +403,7 @@ dialog::backdrop {
 .tooltip {
  @apply !bg-black !text-white !py-2 !px-3 !rounded-md;
 }
+
+.Toastify__toast-body {
+  white-space: pre-line;
+}
--- a/server/endpoints/workspaces.js
+++ b/server/endpoints/workspaces.js
@ -159,7 +159,7 @@ function workspaceEndpoints(app) {
        }

        await Document.removeDocuments(currWorkspace, deletes);
-        const { failed = [] } = await Document.addDocuments(
+        const { failedToEmbed = [], errors = [] } = await Document.addDocuments(
          currWorkspace,
          adds
        );
@ -167,8 +167,10 @@ function workspaceEndpoints(app) {
        response.status(200).json({
          workspace: updatedWorkspace,
          message:
-            failed.length > 0
-              ? `${failed.length} documents could not be embedded.`
+            failedToEmbed.length > 0
+              ? `${failedToEmbed.length} documents failed to add.\n\n${errors
+                  .map((msg) => `${msg}`)
+                  .join("\n\n")}`
              : null,
        });
      } catch (e) {
--- a/server/models/documents.js
+++ b/server/models/documents.js
@ -39,6 +39,7 @@ const Document = {
    if (additions.length === 0) return { failed: [], embedded: [] };
    const embedded = [];
    const failedToEmbed = [];
+    const errors = new Set();

    for (const path of additions) {
      const data = await fileData(path);
@ -53,14 +54,20 @@ const Document = {
        workspaceId: workspace.id,
        metadata: JSON.stringify(metadata),
      };
-      const vectorized = await VectorDb.addDocumentToNamespace(
+
+      const { vectorized, error } = await VectorDb.addDocumentToNamespace(
        workspace.slug,
        { ...data, docId },
        path
      );
+
      if (!vectorized) {
-        console.error("Failed to vectorize", path);
-        failedToEmbed.push(path);
+        console.error(
+          "Failed to vectorize",
+          metadata?.title || newDoc.filename
+        );
+        failedToEmbed.push(metadata?.title || newDoc.filename);
+        errors.add(error);
        continue;
      }

@ -77,7 +84,7 @@ const Document = {
      Embedder: process.env.EMBEDDING_ENGINE || "inherit",
      VectorDbSelection: process.env.VECTOR_DB || "pinecone",
    });
-    return { failed: failedToEmbed, embedded };
+    return { failedToEmbed, errors: Array.from(errors), embedded };
  },

  removeDocuments: async function (workspace, removals = []) {
--- a/server/utils/EmbeddingEngines/azureOpenAi/index.js
+++ b/server/utils/EmbeddingEngines/azureOpenAi/index.js
@ -46,7 +46,12 @@ class AzureOpenAiEmbedder {
              resolve({ data: res.data, error: null });
            })
            .catch((e) => {
-              resolve({ data: [], error: e?.error });
+              e.type =
+                e?.response?.data?.error?.code ||
+                e?.response?.status ||
+                "failed_to_embed";
+              e.message = e?.response?.data?.error?.message || e.message;
+              resolve({ data: [], error: e });
            });
        })
      );
@ -62,11 +67,14 @@ class AzureOpenAiEmbedder {
        .map((res) => res.error)
        .flat();
      if (errors.length > 0) {
+        let uniqueErrors = new Set();
+        errors.map((error) =>
+          uniqueErrors.add(`[${error.type}]: ${error.message}`)
+        );
+
        return {
          data: [],
-          error: `(${errors.length}) Embedding Errors! ${errors
-            .map((error) => `[${error.type}]: ${error.message}`)
-            .join(", ")}`,
+          error: Array.from(uniqueErrors).join(", "),
        };
      }
      return {
--- a/server/utils/EmbeddingEngines/localAi/index.js
+++ b/server/utils/EmbeddingEngines/localAi/index.js
@ -41,7 +41,12 @@ class LocalAiEmbedder {
              resolve({ data: res.data?.data, error: null });
            })
            .catch((e) => {
-              resolve({ data: [], error: e?.error });
+              e.type =
+                e?.response?.data?.error?.code ||
+                e?.response?.status ||
+                "failed_to_embed";
+              e.message = e?.response?.data?.error?.message || e.message;
+              resolve({ data: [], error: e });
            });
        })
      );
@ -57,11 +62,14 @@ class LocalAiEmbedder {
        .map((res) => res.error)
        .flat();
      if (errors.length > 0) {
+        let uniqueErrors = new Set();
+        errors.map((error) =>
+          uniqueErrors.add(`[${error.type}]: ${error.message}`)
+        );
+
        return {
          data: [],
-          error: `(${errors.length}) Embedding Errors! ${errors
-            .map((error) => `[${error.type}]: ${error.message}`)
-            .join(", ")}`,
+          error: Array.from(uniqueErrors).join(", "),
        };
      }
      return {
--- a/server/utils/EmbeddingEngines/openAi/index.js
+++ b/server/utils/EmbeddingEngines/openAi/index.js
@ -37,7 +37,12 @@ class OpenAiEmbedder {
              resolve({ data: res.data?.data, error: null });
            })
            .catch((e) => {
-              resolve({ data: [], error: e?.error });
+              e.type =
+                e?.response?.data?.error?.code ||
+                e?.response?.status ||
+                "failed_to_embed";
+              e.message = e?.response?.data?.error?.message || e.message;
+              resolve({ data: [], error: e });
            });
        })
      );
@ -53,11 +58,14 @@ class OpenAiEmbedder {
        .map((res) => res.error)
        .flat();
      if (errors.length > 0) {
+        let uniqueErrors = new Set();
+        errors.map((error) =>
+          uniqueErrors.add(`[${error.type}]: ${error.message}`)
+        );
+
        return {
          data: [],
-          error: `(${errors.length}) Embedding Errors! ${errors
-            .map((error) => `[${error.type}]: ${error.message}`)
-            .join(", ")}`,
+          error: Array.from(uniqueErrors).join(", "),
        };
      }
      return {
--- a/server/utils/vectorDbProviders/chroma/index.js
+++ b/server/utils/vectorDbProviders/chroma/index.js
@ -171,7 +171,7 @@ const Chroma = {
        }

        await DocumentVectors.bulkInsert(documentVectors);
-        return true;
+        return { vectorized: true, error: null };
      }

      // If we are here then we are going to embed and store a novel document.
@ -242,11 +242,10 @@ const Chroma = {
      }

      await DocumentVectors.bulkInsert(documentVectors);
-      return true;
+      return { vectorized: true, error: null };
    } catch (e) {
-      console.error(e);
      console.error("addDocumentToNamespace", e.message);
-      return false;
+      return { vectorized: false, error: e.message };
    }
  },
  deleteDocumentFromNamespace: async function (namespace, docId) {
--- a/server/utils/vectorDbProviders/lance/index.js
+++ b/server/utils/vectorDbProviders/lance/index.js
@ -172,7 +172,7 @@ const LanceDb = {

        await this.updateOrCreateCollection(client, submissions, namespace);
        await DocumentVectors.bulkInsert(documentVectors);
-        return true;
+        return { vectorized: true, error: null };
      }

      // If we are here then we are going to embed and store a novel document.
@ -229,11 +229,10 @@ const LanceDb = {
      }

      await DocumentVectors.bulkInsert(documentVectors);
-      return true;
+      return { vectorized: true, error: null };
    } catch (e) {
-      console.error(e);
      console.error("addDocumentToNamespace", e.message);
-      return false;
+      return { vectorized: false, error: e.message };
    }
  },
  performSimilaritySearch: async function ({
--- a/server/utils/vectorDbProviders/milvus/index.js
+++ b/server/utils/vectorDbProviders/milvus/index.js
@ -167,7 +167,7 @@ const Milvus = {
        }
        await DocumentVectors.bulkInsert(documentVectors);
        await client.flushSync({ collection_names: [namespace] });
-        return true;
+        return { vectorized: true, error: null };
      }

      const textSplitter = new RecursiveCharacterTextSplitter({
@ -231,11 +231,10 @@ const Milvus = {
      }

      await DocumentVectors.bulkInsert(documentVectors);
-      return true;
+      return { vectorized: true, error: null };
    } catch (e) {
-      console.error(e);
      console.error("addDocumentToNamespace", e.message);
-      return false;
+      return { vectorized: false, error: e.message };
    }
  },
  deleteDocumentFromNamespace: async function (namespace, docId) {
--- a/server/utils/vectorDbProviders/pinecone/index.js
+++ b/server/utils/vectorDbProviders/pinecone/index.js
@ -125,7 +125,7 @@ const Pinecone = {
        }

        await DocumentVectors.bulkInsert(documentVectors);
-        return true;
+        return { vectorized: true, error: null };
      }

      // If we are here then we are going to embed and store a novel document.
@ -183,11 +183,10 @@ const Pinecone = {
      }

      await DocumentVectors.bulkInsert(documentVectors);
-      return true;
+      return { vectorized: true, error: null };
    } catch (e) {
-      console.error(e);
      console.error("addDocumentToNamespace", e.message);
-      return false;
+      return { vectorized: false, error: e.message };
    }
  },
  deleteDocumentFromNamespace: async function (namespace, docId) {
--- a/server/utils/vectorDbProviders/qdrant/index.js
+++ b/server/utils/vectorDbProviders/qdrant/index.js
@ -190,7 +190,7 @@ const QDrant = {
        }

        await DocumentVectors.bulkInsert(documentVectors);
-        return true;
+        return { vectorized: true, error: null };
      }

      // If we are here then we are going to embed and store a novel document.
@ -272,11 +272,10 @@ const QDrant = {
      }

      await DocumentVectors.bulkInsert(documentVectors);
-      return true;
+      return { vectorized: true, error: null };
    } catch (e) {
-      console.error(e);
      console.error("addDocumentToNamespace", e.message);
-      return false;
+      return { vectorized: false, error: e.message };
    }
  },
  deleteDocumentFromNamespace: async function (namespace, docId) {
--- a/server/utils/vectorDbProviders/weaviate/index.js
+++ b/server/utils/vectorDbProviders/weaviate/index.js
@ -233,7 +233,7 @@ const Weaviate = {
        }

        await DocumentVectors.bulkInsert(documentVectors);
-        return true;
+        return { vectorized: true, error: null };
      }

      // If we are here then we are going to embed and store a novel document.
@ -316,11 +316,10 @@ const Weaviate = {
      }

      await DocumentVectors.bulkInsert(documentVectors);
-      return true;
+      return { vectorized: true, error: null };
    } catch (e) {
-      console.error(e);
      console.error("addDocumentToNamespace", e.message);
-      return false;
+      return { vectorized: false, error: e.message };
    }
  },
  deleteDocumentFromNamespace: async function (namespace, docId) {
--- a/server/utils/vectorDbProviders/zilliz/index.js
+++ b/server/utils/vectorDbProviders/zilliz/index.js
@ -168,7 +168,7 @@ const Zilliz = {
        }
        await DocumentVectors.bulkInsert(documentVectors);
        await client.flushSync({ collection_names: [namespace] });
-        return true;
+        return { vectorized: true, error: null };
      }

      const textSplitter = new RecursiveCharacterTextSplitter({
@ -232,11 +232,10 @@ const Zilliz = {
      }

      await DocumentVectors.bulkInsert(documentVectors);
-      return true;
+      return { vectorized: true, error: null };
    } catch (e) {
-      console.error(e);
      console.error("addDocumentToNamespace", e.message);
-      return false;
+      return { vectorized: false, error: e.message };
    }
  },
  deleteDocumentFromNamespace: async function (namespace, docId) {