Mintplex-Labs · timothycarambat · Jul 30, 2025 · Jul 30, 2025
diff --git a/collector/processLink/convert/generic.js b/collector/processLink/convert/generic.js
@@ -62,10 +62,10 @@ async function scrapeGenericUrl({
     token_count_estimate: tokenizeString(content),
   };
 
-  const document = writeToServerDocuments(
+  const document = writeToServerDocuments({
     data,
-    `url-${slugify(filename)}-${data.id}`
-  );
+    filename: `url-${slugify(filename)}-${data.id}`,
+  });
   console.log(`[SUCCESS]: URL ${link} converted & ready for embedding.\n`);
   return { success: true, reason: null, documents: [document] };
 }

diff --git a/collector/processRawText/index.js b/collector/processRawText/index.js
@@ -58,10 +58,10 @@ async function processRawText(textContent, metadata) {
     token_count_estimate: tokenizeString(textContent),
   };
 
-  const document = writeToServerDocuments(
+  const document = writeToServerDocuments({
     data,
-    `raw-${stripAndSlug(metadata.title)}-${data.id}`
-  );
+    filename: `raw-${stripAndSlug(metadata.title)}-${data.id}`,
+  });
   console.log(`[SUCCESS]: Raw text and metadata saved & ready for embedding.\n`);
   return { success: true, reason: null, documents: [document] };
 }

diff --git a/collector/processSingleFile/convert/asAudio.js b/collector/processSingleFile/convert/asAudio.js
@@ -59,10 +59,10 @@ async function asAudio({ fullFilePath = "", filename = "", options = {} }) {
     token_count_estimate: tokenizeString(content),
   };
 
-  const document = writeToServerDocuments(
+  const document = writeToServerDocuments({
     data,
-    `${slugify(filename)}-${data.id}`
-  );
+    filename: `${slugify(filename)}-${data.id}`,
+  });
   trashFile(fullFilePath);
   console.log(
     `[SUCCESS]: ${filename} transcribed, converted & ready for embedding.\n`

diff --git a/collector/processSingleFile/convert/asDocx.js b/collector/processSingleFile/convert/asDocx.js
@@ -45,10 +45,10 @@ async function asDocX({ fullFilePath = "", filename = "" }) {
     token_count_estimate: tokenizeString(content),
   };
 
-  const document = writeToServerDocuments(
+  const document = writeToServerDocuments({
     data,
-    `${slugify(filename)}-${data.id}`
-  );
+    filename: `${slugify(filename)}-${data.id}`,
+  });
   trashFile(fullFilePath);
   console.log(`[SUCCESS]: ${filename} converted & ready for embedding.\n`);
   return { success: true, reason: null, documents: [document] };

diff --git a/collector/processSingleFile/convert/asEPub.js b/collector/processSingleFile/convert/asEPub.js
@@ -43,10 +43,10 @@ async function asEPub({ fullFilePath = "", filename = "" }) {
     token_count_estimate: tokenizeString(content),
   };
 
-  const document = writeToServerDocuments(
+  const document = writeToServerDocuments({
     data,
-    `${slugify(filename)}-${data.id}`
-  );
+    filename: `${slugify(filename)}-${data.id}`,
+  });
   trashFile(fullFilePath);
   console.log(`[SUCCESS]: ${filename} converted & ready for embedding.\n`);
   return { success: true, reason: null, documents: [document] };

diff --git a/collector/processSingleFile/convert/asImage.js b/collector/processSingleFile/convert/asImage.js
@@ -38,10 +38,10 @@ async function asImage({ fullFilePath = "", filename = "", options = {} }) {
     token_count_estimate: tokenizeString(content),
   };
 
-  const document = writeToServerDocuments(
+  const document = writeToServerDocuments({
     data,
-    `${slugify(filename)}-${data.id}`
-  );
+    filename: `${slugify(filename)}-${data.id}`,
+  });
   trashFile(fullFilePath);
   console.log(`[SUCCESS]: ${filename} converted & ready for embedding.\n`);
   return { success: true, reason: null, documents: [document] };

diff --git a/collector/processSingleFile/convert/asMbox.js b/collector/processSingleFile/convert/asMbox.js
@@ -57,10 +57,10 @@ async function asMbox({ fullFilePath = "", filename = "" }) {
     };
 
     item++;
-    const document = writeToServerDocuments(
+    const document = writeToServerDocuments({
       data,
-      `${slugify(filename)}-${data.id}-msg-${item}`
-    );
+      filename: `${slugify(filename)}-${data.id}-msg-${item}`,
+    });
     documents.push(document);
   }
 

diff --git a/collector/processSingleFile/convert/asOfficeMime.js b/collector/processSingleFile/convert/asOfficeMime.js
@@ -41,10 +41,10 @@ async function asOfficeMime({ fullFilePath = "", filename = "" }) {
     token_count_estimate: tokenizeString(content),
   };
 
-  const document = writeToServerDocuments(
+  const document = writeToServerDocuments({
     data,
-    `${slugify(filename)}-${data.id}`
-  );
+    filename: `${slugify(filename)}-${data.id}`,
+  });
   trashFile(fullFilePath);
   console.log(`[SUCCESS]: ${filename} converted & ready for embedding.\n`);
   return { success: true, reason: null, documents: [document] };

diff --git a/collector/processSingleFile/convert/asPDF/index.js b/collector/processSingleFile/convert/asPDF/index.js
@@ -62,10 +62,10 @@ async function asPdf({ fullFilePath = "", filename = "", options = {} }) {
     token_count_estimate: tokenizeString(content),
   };
 
-  const document = writeToServerDocuments(
+  const document = writeToServerDocuments({
     data,
-    `${slugify(filename)}-${data.id}`
-  );
+    filename: `${slugify(filename)}-${data.id}`,
+  });
   trashFile(fullFilePath);
   console.log(`[SUCCESS]: ${filename} converted & ready for embedding.\n`);
   return { success: true, reason: null, documents: [document] };

diff --git a/collector/processSingleFile/convert/asTxt.js b/collector/processSingleFile/convert/asTxt.js
@@ -41,10 +41,10 @@ async function asTxt({ fullFilePath = "", filename = "" }) {
     token_count_estimate: tokenizeString(content),
   };
 
-  const document = writeToServerDocuments(
+  const document = writeToServerDocuments({
     data,
-    `${slugify(filename)}-${data.id}`
-  );
+    filename: `${slugify(filename)}-${data.id}`,
+  });
   trashFile(fullFilePath);
   console.log(`[SUCCESS]: ${filename} converted & ready for embedding.\n`);
   return { success: true, reason: null, documents: [document] };

diff --git a/collector/processSingleFile/convert/asXlsx.js b/collector/processSingleFile/convert/asXlsx.js
@@ -64,11 +64,11 @@ async function asXlsx({ fullFilePath = "", filename = "" }) {
           token_count_estimate: tokenizeString(content),
         };
 
-        const document = writeToServerDocuments(
-          sheetData,
-          `sheet-${slugify(name)}`,
-          outFolderPath
-        );
+        const document = writeToServerDocuments({
+          data: sheetData,
+          filename: `sheet-${slugify(name)}`,
+          destinationOverride: outFolderPath,
+        });
         documents.push(document);
         console.log(
           `[SUCCESS]: Sheet "${name}" converted & ready for embedding.`

diff --git a/collector/utils/extensions/Confluence/index.js b/collector/utils/extensions/Confluence/index.js
@@ -116,7 +116,11 @@ async function loadConfluence(
     const fileName = sanitizeFileName(
       `${slugify(doc.metadata.title)}-${data.id}`
     );
-    writeToServerDocuments(data, fileName, outFolderPath);
+    writeToServerDocuments({
+      data,
+      filename: fileName,
+      destinationOverride: outFolderPath,
+    });
   });
 
   return {

diff --git a/collector/utils/extensions/DrupalWiki/DrupalWiki/index.js b/collector/utils/extensions/DrupalWiki/DrupalWiki/index.js
@@ -202,7 +202,11 @@ class DrupalWiki {
     console.log(
       `[DrupalWiki Loader]: Saving page '${page.title}' (${page.id}) to '${this.storagePath}/${fileName}'`
     );
-    writeToServerDocuments(data, fileName, this.storagePath);
+    writeToServerDocuments({
+      data,
+      filename: fileName,
+      destinationOverride: this.storagePath,
+    });
   }
 
   /**

diff --git a/collector/utils/extensions/ObsidianVault/index.js b/collector/utils/extensions/ObsidianVault/index.js
@@ -66,7 +66,11 @@ async function loadObsidianVault({ files = [] }) {
       const targetFileName = sanitizeFileName(
         `${slugify(file.name)}-${data.id}`
       );
-      writeToServerDocuments(data, targetFileName, outFolderPath);
+      writeToServerDocuments({
+        data,
+        filename: targetFileName,
+        destinationOverride: outFolderPath,
+      });
       results.push({ file: file.path, status: "success" });
     } catch (e) {
       console.error(`Failed to process ${file.path}:`, e);

diff --git a/collector/utils/extensions/RepoLoader/GithubRepo/index.js b/collector/utils/extensions/RepoLoader/GithubRepo/index.js
@@ -71,11 +71,11 @@ async function loadGithubRepo(args, response) {
     console.log(
       `[GitHub Loader]: Saving ${doc.metadata.source} to ${outFolder}`
     );
-    writeToServerDocuments(
+    writeToServerDocuments({
       data,
-      `${slugify(doc.metadata.source)}-${data.id}`,
-      outFolderPath
-    );
+      filename: `${slugify(doc.metadata.source)}-${data.id}`,
+      destinationOverride: outFolderPath,
+    });
   }
 
   return {

diff --git a/collector/utils/extensions/RepoLoader/GitlabRepo/index.js b/collector/utils/extensions/RepoLoader/GitlabRepo/index.js
@@ -98,11 +98,11 @@ async function loadGitlabRepo(args, response) {
       `[GitLab Loader]: Saving ${doc.metadata.source} to ${outFolder}`
     );
 
-    writeToServerDocuments(
+    writeToServerDocuments({
       data,
-      sanitizeFileName(`${slugify(doc.metadata.source)}-${data.id}`),
-      outFolderPath
-    );
+      filename: sanitizeFileName(`${slugify(doc.metadata.source)}-${data.id}`),
+      destinationOverride: outFolderPath,
+    });
   }
 
   return {

diff --git a/collector/utils/extensions/WebsiteDepth/index.js b/collector/utils/extensions/WebsiteDepth/index.js
@@ -125,7 +125,11 @@ async function bulkScrapePages(links, outFolderPath) {
         token_count_estimate: tokenizeString(content),
       };
 
-      writeToServerDocuments(data, data.title, outFolderPath);
+      writeToServerDocuments({
+        data,
+        filename: data.title,
+        destinationOverride: outFolderPath,
+      });
       scrapedData.push(data);
 
       console.log(`Successfully scraped ${link}.`);

diff --git a/collector/utils/extensions/YoutubeTranscript/index.js b/collector/utils/extensions/YoutubeTranscript/index.js
@@ -116,11 +116,11 @@ async function loadYouTubeTranscript({ url }) {
   };
 
   console.log(`[YouTube Loader]: Saving ${metadata.title} to ${outFolder}`);
-  writeToServerDocuments(
+  writeToServerDocuments({
     data,
-    sanitizeFileName(`${slugify(metadata.title)}-${data.id}`),
-    outFolderPath
-  );
+    filename: sanitizeFileName(`${slugify(metadata.title)}-${data.id}`),
+    destinationOverride: outFolderPath,
+  });
 
   return {
     success: true,

diff --git a/collector/utils/files/index.js b/collector/utils/files/index.js
@@ -96,11 +96,21 @@ function createdDate(filepath) {
   }
 }
 
-function writeToServerDocuments(
+/**
+ * Writes a document to the server documents folder.
+ * @param {Object} params - The parameters for the function.
+ * @param {Object} params.data - The data to write to the file. Must look like a document object.
+ * @param {string} params.filename - The name of the file to write to.
+ * @param {string|null} params.destinationOverride - A forced destination to write to - will be honored if provided.
+ * @returns {Object} - The data with the location added.
+ */
+function writeToServerDocuments({
   data = {},
-  filename,
-  destinationOverride = null
-) {
+  filename = null,
+  destinationOverride = null,
+}) {
+  if (!filename) throw new Error("Filename is required!");
+
   let destination = null;
   if (destinationOverride) destination = path.resolve(destinationOverride);
   else destination = path.resolve(documentsFolder, "custom-documents");