Merge branch 'main' of https://github.com/FlowiseAI/Flowise

2026-06-29 05:01:10 +03:00 · 2023-05-09 00:49:37 +01:00
parent 9a8a3fc67f 6001ed6ecf
commit f0097c7f57
21 changed files with 4188 additions and 2725 deletions
@@ -0,0 +1,63 @@
+import { INode, INodeData, INodeParams } from '../../../src/Interface'
+import { getBaseClasses } from '../../../src/utils'
+import { CharacterTextSplitter, CharacterTextSplitterParams } from 'langchain/text_splitter'
+
+class CharacterTextSplitter_TextSplitters implements INode {
+    label: string
+    name: string
+    description: string
+    type: string
+    icon: string
+    category: string
+    baseClasses: string[]
+    inputs: INodeParams[]
+
+    constructor() {
+        this.label = 'Character Text Splitter'
+        this.name = 'characterTextSplitter'
+        this.type = 'CharacterTextSplitter'
+        this.icon = 'textsplitter.svg'
+        this.category = 'Text Splitters'
+        this.description = `splits only on one type of character (defaults to "\\n\\n").`
+        this.baseClasses = [this.type, ...getBaseClasses(CharacterTextSplitter)]
+        this.inputs = [
+            {
+                label: 'Separator',
+                name: 'separator',
+                type: 'string',
+                optional: true
+            },
+            {
+                label: 'Chunk Size',
+                name: 'chunkSize',
+                type: 'number',
+                default: 1000,
+                optional: true
+            },
+            {
+                label: 'Chunk Overlap',
+                name: 'chunkOverlap',
+                type: 'number',
+                optional: true
+            }
+        ]
+    }
+
+    async init(nodeData: INodeData): Promise<any> {
+        const separator = nodeData.inputs?.separator as string
+        const chunkSize = nodeData.inputs?.chunkSize as string
+        const chunkOverlap = nodeData.inputs?.chunkOverlap as string
+
+        const obj = {} as CharacterTextSplitterParams
+
+        if (separator) obj.separator = separator
+        if (chunkSize) obj.chunkSize = parseInt(chunkSize, 10)
+        if (chunkOverlap) obj.chunkOverlap = parseInt(chunkOverlap, 10)
+
+        const splitter = new CharacterTextSplitter(obj)
+
+        return splitter
+    }
+}
+
+module.exports = { nodeClass: CharacterTextSplitter_TextSplitters }
@@ -0,0 +1,7 @@
+<svg xmlns="http://www.w3.org/2000/svg" class="icon icon-tabler icon-tabler-abc" width="24" height="24" viewBox="0 0 24 24" stroke-width="2" stroke="currentColor" fill="none" stroke-linecap="round" stroke-linejoin="round">
+   <path stroke="none" d="M0 0h24v24H0z" fill="none"></path>
+   <path d="M3 16v-6a2 2 0 1 1 4 0v6"></path>
+   <path d="M3 13h4"></path>
+   <path d="M10 8v6a2 2 0 1 0 4 0v-1a2 2 0 1 0 -4 0v1"></path>
+   <path d="M20.732 12a2 2 0 0 0 -3.732 1v1a2 2 0 0 0 3.726 1.01"></path>
+</svg>
@@ -0,0 +1,55 @@
+import { INode, INodeData, INodeParams } from '../../../src/Interface'
+import { getBaseClasses } from '../../../src/utils'
+import { MarkdownTextSplitter, MarkdownTextSplitterParams } from 'langchain/text_splitter'
+
+class MarkdownTextSplitter_TextSplitters implements INode {
+    label: string
+    name: string
+    description: string
+    type: string
+    icon: string
+    category: string
+    baseClasses: string[]
+    inputs: INodeParams[]
+
+    constructor() {
+        this.label = 'Markdown Text Splitter'
+        this.name = 'markdownTextSplitter'
+        this.type = 'MarkdownTextSplitter'
+        this.icon = 'markdownTextSplitter.svg'
+        this.category = 'Text Splitters'
+        this.description = `Split your content into documents based on the Markdown headers`
+        this.baseClasses = [this.type, ...getBaseClasses(MarkdownTextSplitter)]
+        this.inputs = [
+            {
+                label: 'Chunk Size',
+                name: 'chunkSize',
+                type: 'number',
+                default: 1000,
+                optional: true
+            },
+            {
+                label: 'Chunk Overlap',
+                name: 'chunkOverlap',
+                type: 'number',
+                optional: true
+            }
+        ]
+    }
+
+    async init(nodeData: INodeData): Promise<any> {
+        const chunkSize = nodeData.inputs?.chunkSize as string
+        const chunkOverlap = nodeData.inputs?.chunkOverlap as string
+
+        const obj = {} as MarkdownTextSplitterParams
+
+        if (chunkSize) obj.chunkSize = parseInt(chunkSize, 10)
+        if (chunkOverlap) obj.chunkOverlap = parseInt(chunkOverlap, 10)
+
+        const splitter = new MarkdownTextSplitter(obj)
+
+        return splitter
+    }
+}
+
+module.exports = { nodeClass: MarkdownTextSplitter_TextSplitters }
@@ -0,0 +1,6 @@
+<svg xmlns="http://www.w3.org/2000/svg" class="icon icon-tabler icon-tabler-markdown" width="24" height="24" viewBox="0 0 24 24" stroke-width="2" stroke="currentColor" fill="none" stroke-linecap="round" stroke-linejoin="round">
+   <path stroke="none" d="M0 0h24v24H0z" fill="none"></path>
+   <path d="M3 5m0 2a2 2 0 0 1 2 -2h14a2 2 0 0 1 2 2v10a2 2 0 0 1 -2 2h-14a2 2 0 0 1 -2 -2z"></path>
+   <path d="M7 15v-6l2 2l2 -2v6"></path>
+   <path d="M14 13l2 2l2 -2m-2 2v-6"></path>
+</svg>
@@ -18,7 +18,7 @@ class RecursiveCharacterTextSplitter_TextSplitters implements INode {
        this.type = 'RecursiveCharacterTextSplitter'
        this.icon = 'textsplitter.svg'
        this.category = 'Text Splitters'
-        this.description = `Split documents recursively by different characters - starting with "\n\n", then "\n", then " "`
+        this.description = `Split documents recursively by different characters - starting with "\\n\\n", then "\\n", then " "`
        this.baseClasses = [this.type, ...getBaseClasses(RecursiveCharacterTextSplitter)]
        this.inputs = [
            {
@@ -0,0 +1,86 @@
+import { INode, INodeData, INodeParams } from '../../../src/Interface'
+import { getBaseClasses } from '../../../src/utils'
+import { TokenTextSplitter, TokenTextSplitterParams } from 'langchain/text_splitter'
+import { TiktokenEncoding } from '@dqbd/tiktoken'
+
+class TokenTextSplitter_TextSplitters implements INode {
+    label: string
+    name: string
+    description: string
+    type: string
+    icon: string
+    category: string
+    baseClasses: string[]
+    inputs: INodeParams[]
+
+    constructor() {
+        this.label = 'Token Text Splitter'
+        this.name = 'tokenTextSplitter'
+        this.type = 'TokenTextSplitter'
+        this.icon = 'tiktoken.svg'
+        this.category = 'Text Splitters'
+        this.description = `Splits a raw text string by first converting the text into BPE tokens, then split these tokens into chunks and convert the tokens within a single chunk back into text.`
+        this.baseClasses = [this.type, ...getBaseClasses(TokenTextSplitter)]
+        this.inputs = [
+            {
+                label: 'Encoding Name',
+                name: 'encodingName',
+                type: 'options',
+                options: [
+                    {
+                        label: 'gpt2',
+                        name: 'gpt2'
+                    },
+                    {
+                        label: 'r50k_base',
+                        name: 'r50k_base'
+                    },
+                    {
+                        label: 'p50k_base',
+                        name: 'p50k_base'
+                    },
+                    {
+                        label: 'p50k_edit',
+                        name: 'p50k_edit'
+                    },
+                    {
+                        label: 'cl100k_base',
+                        name: 'cl100k_base'
+                    }
+                ],
+                default: 'gpt2'
+            },
+            {
+                label: 'Chunk Size',
+                name: 'chunkSize',
+                type: 'number',
+                default: 1000,
+                optional: true
+            },
+            {
+                label: 'Chunk Overlap',
+                name: 'chunkOverlap',
+                type: 'number',
+                optional: true
+            }
+        ]
+    }
+
+    async init(nodeData: INodeData): Promise<any> {
+        const encodingName = nodeData.inputs?.encodingName as string
+        const chunkSize = nodeData.inputs?.chunkSize as string
+        const chunkOverlap = nodeData.inputs?.chunkOverlap as string
+
+        const obj = {} as TokenTextSplitterParams
+
+        obj.encodingName = encodingName as TiktokenEncoding
+        if (chunkSize) obj.chunkSize = parseInt(chunkSize, 10)
+        if (chunkOverlap) obj.chunkOverlap = parseInt(chunkOverlap, 10)
+
+        const splitter = new TokenTextSplitter(obj)
+
+        return splitter
+    }
+}
+
+module.exports = { nodeClass: TokenTextSplitter_TextSplitters }
@@ -0,0 +1,7 @@
+<svg xmlns="http://www.w3.org/2000/svg" class="icon icon-tabler icon-tabler-hourglass" width="24" height="24" viewBox="0 0 24 24" stroke-width="2" stroke="currentColor" fill="none" stroke-linecap="round" stroke-linejoin="round">
+   <path stroke="none" d="M0 0h24v24H0z" fill="none"></path>
+   <path d="M6.5 7h11"></path>
+   <path d="M6.5 17h11"></path>
+   <path d="M6 20v-2a6 6 0 1 1 12 0v2a1 1 0 0 1 -1 1h-10a1 1 0 0 1 -1 -1z"></path>
+   <path d="M6 4v2a6 6 0 1 0 12 0v-2a1 1 0 0 0 -1 -1h-10a1 1 0 0 0 -1 1z"></path>
+</svg>
@@ -16,7 +16,7 @@
    },
    "license": "SEE LICENSE IN LICENSE.md",
    "dependencies": {
-        "@dqbd/tiktoken": "^1.0.4",
+        "@dqbd/tiktoken": "^1.0.7",
        "@huggingface/inference": "^1.6.3",
        "@pinecone-database/pinecone": "^0.0.12",
        "@supabase/supabase-js": "^2.21.0",