From 7ce0f71e2fc53cf9180c5af8659611b1b39d4c0f Mon Sep 17 00:00:00 2001
From: ivalkshfoeif <linsichen2018@gmail.com>
Date: Wed, 21 Jun 2023 18:56:37 -0700
Subject: [PATCH 1/2] add JSONLines Loader

---
 .../documentloaders/Jsonlines/Jsonlines.ts    | 105 ++++++++++++++++++
 .../documentloaders/Jsonlines/jsonlines.svg   |  16 +++
 2 files changed, 121 insertions(+)
 create mode 100644 packages/components/nodes/documentloaders/Jsonlines/Jsonlines.ts
 create mode 100644 packages/components/nodes/documentloaders/Jsonlines/jsonlines.svg

diff --git a/packages/components/nodes/documentloaders/Jsonlines/Jsonlines.ts b/packages/components/nodes/documentloaders/Jsonlines/Jsonlines.ts
new file mode 100644
index 00000000000..e92a97aa2f3
--- /dev/null
+++ b/packages/components/nodes/documentloaders/Jsonlines/Jsonlines.ts
@@ -0,0 +1,105 @@
+import { INode, INodeData, INodeParams } from '../../../src/Interface'
+import { TextSplitter } from 'langchain/text_splitter'
+import { JSONLinesLoader } from 'langchain/document_loaders/fs/json'
+
+class Jsonlines_DocumentLoaders implements INode {
+    label: string
+    name: string
+    description: string
+    type: string
+    icon: string
+    category: string
+    baseClasses: string[]
+    inputs: INodeParams[]
+
+    constructor() {
+        this.label = 'Json Lines File'
+        this.name = 'jsonlinesFile'
+        this.type = 'Document'
+        this.icon = 'jsonlines.svg'
+        this.category = 'Document Loaders'
+        this.description = `Load data from JSON Lines files`
+        this.baseClasses = [this.type]
+        this.inputs = [
+            {
+                label: 'Jsonlines File',
+                name: 'jsonlinesFile',
+                type: 'file',
+                fileType: '.jsonl'
+            },
+            {
+                label: 'Text Splitter',
+                name: 'textSplitter',
+                type: 'TextSplitter',
+                optional: true
+            },
+            {
+                label: 'Pointer Extraction',
+                name: 'pointerName',
+                type: 'string',
+                description: 'Extracting the pointer',
+                placeholder: 'Enter pointer name',
+                optional: true
+            },
+            {
+                label: 'Metadata',
+                name: 'metadata',
+                type: 'json',
+                optional: true,
+                additionalParams: true
+            }
+        ]
+    }
+
+    async init(nodeData: INodeData): Promise<any> {
+        const textSplitter = nodeData.inputs?.textSplitter as TextSplitter
+        const jsonLinesFileBase64 = nodeData.inputs?.jsonlinesFile as string
+        const pointerName = nodeData.inputs?.pointerName as string
+        const metadata = nodeData.inputs?.metadata
+
+        let alldocs = []
+        let files: string[] = []
+
+        if (jsonLinesFileBase64.startsWith('[') && jsonLinesFileBase64.endsWith(']')) {
+            files = JSON.parse(jsonLinesFileBase64)
+        } else {
+            files = [jsonLinesFileBase64]
+        }
+
+        for (const file of files) {
+            const splitDataURI = file.split(',')
+            splitDataURI.pop()
+            const bf = Buffer.from(splitDataURI.pop() || '', 'base64')
+            const blob = new Blob([bf])
+            const loader = new JSONLinesLoader(blob, pointerName)
+
+            if (textSplitter) {
+                const docs = await loader.loadAndSplit(textSplitter)
+                alldocs.push(...docs)
+            } else {
+                const docs = await loader.load()
+                alldocs.push(...docs)
+            }
+        }
+
+        if (metadata) {
+            const parsedMetadata = typeof metadata === 'object' ? metadata : JSON.parse(metadata)
+            let finaldocs = []
+            for (const doc of alldocs) {
+                const newdoc = {
+                    ...doc,
+                    metadata: {
+                        ...doc.metadata,
+                        ...parsedMetadata
+                    }
+                }
+                finaldocs.push(newdoc)
+            }
+            return finaldocs
+        }
+
+        return alldocs
+    }
+}
+
+module.exports = { nodeClass: Jsonlines_DocumentLoaders }
diff --git a/packages/components/nodes/documentloaders/Jsonlines/jsonlines.svg b/packages/components/nodes/documentloaders/Jsonlines/jsonlines.svg
new file mode 100644
index 00000000000..f3686f0c983
--- /dev/null
+++ b/packages/components/nodes/documentloaders/Jsonlines/jsonlines.svg
@@ -0,0 +1,16 @@
+<svg width="24" height="24" xmlns="http://www.w3.org/2000/svg">
+ <!-- Created with Method Draw - http://github.com/duopixel/Method-Draw/ -->
+
+ <g>
+  <title>background</title>
+  <rect fill="none" id="canvas_background" height="26" width="26" y="-1" x="-1"/>
+  <g display="none" overflow="visible" y="0" x="0" height="100%" width="100%" id="canvasGrid">
+   <rect fill="url(#gridpattern)" stroke-width="0" y="0" x="0" height="100%" width="100%"/>
+  </g>
+ </g>
+ <g>
+  <title>Layer 1</title>
+  <text font-weight="bold" stroke="#000" transform="matrix(8.682896011956823,0,0,10.412942243751806,-30.866304860177404,-63.784276261342) " xml:space="preserve" text-anchor="start" font-family="Helvetica, Arial, sans-serif" font-size="1" id="svg_2" y="7.062874" x="3.579384" stroke-opacity="null" stroke-width="0" fill="#000000">JSON</text>
+  <text font-weight="bold" stroke="#000" transform="matrix(9.059566511875573,0,0,9.893934811310315,-1.3962337706973242,-106.08964247698567) " xml:space="preserve" text-anchor="start" font-family="Helvetica, Arial, sans-serif" font-size="1" id="svg_3" y="12.90427" x="0.172236" stroke-opacity="null" stroke-width="0" fill="#000000">Lines</text>
+ </g>
+</svg>
\ No newline at end of file

From 7646e973e3a77d8f4500f82d3e70ba51bd203048 Mon Sep 17 00:00:00 2001
From: ivalkshfoeif <linsichen2018@gmail.com>
Date: Wed, 21 Jun 2023 19:34:30 -0700
Subject: [PATCH 2/2] update pointer attribute and logic

---
 .../nodes/documentloaders/Jsonlines/Jsonlines.ts           | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/packages/components/nodes/documentloaders/Jsonlines/Jsonlines.ts b/packages/components/nodes/documentloaders/Jsonlines/Jsonlines.ts
index e92a97aa2f3..4af8c2ce926 100644
--- a/packages/components/nodes/documentloaders/Jsonlines/Jsonlines.ts
+++ b/packages/components/nodes/documentloaders/Jsonlines/Jsonlines.ts
@@ -37,9 +37,8 @@ class Jsonlines_DocumentLoaders implements INode {
                 label: 'Pointer Extraction',
                 name: 'pointerName',
                 type: 'string',
-                description: 'Extracting the pointer',
                 placeholder: 'Enter pointer name',
-                optional: true
+                optional: false
             },
             {
                 label: 'Metadata',
@@ -60,6 +59,8 @@ class Jsonlines_DocumentLoaders implements INode {
         let alldocs = []
         let files: string[] = []
 
+        let pointer = '/' + pointerName.trim()
+
         if (jsonLinesFileBase64.startsWith('[') && jsonLinesFileBase64.endsWith(']')) {
             files = JSON.parse(jsonLinesFileBase64)
         } else {
@@ -71,7 +72,7 @@ class Jsonlines_DocumentLoaders implements INode {
             splitDataURI.pop()
             const bf = Buffer.from(splitDataURI.pop() || '', 'base64')
             const blob = new Blob([bf])
-            const loader = new JSONLinesLoader(blob, pointerName)
+            const loader = new JSONLinesLoader(blob, pointer)
 
             if (textSplitter) {
                 const docs = await loader.loadAndSplit(textSplitter)