From 261e45d74a2f912dd0ab6e356fef90c97b420f3f Mon Sep 17 00:00:00 2001
From: vinodkiran <vinodkiran@usa.net>
Date: Mon, 13 Nov 2023 21:56:18 +0530
Subject: [PATCH] MongoDB Atlas Integration: Adding MongoDB as a Vector Store

---
 .../vectorstores/MongoDB/MongoDBSearchBase.ts | 145 ++++++++++++++++++
 .../vectorstores/MongoDB/MongoDB_Existing.ts  |  41 +++++
 .../vectorstores/MongoDB/MongoDB_Upsert.ts    |  58 +++++++
 .../nodes/vectorstores/MongoDB/mongodb.png    | Bin 0 -> 3741 bytes
 4 files changed, 244 insertions(+)
 create mode 100644 packages/components/nodes/vectorstores/MongoDB/MongoDBSearchBase.ts
 create mode 100644 packages/components/nodes/vectorstores/MongoDB/MongoDB_Existing.ts
 create mode 100644 packages/components/nodes/vectorstores/MongoDB/MongoDB_Upsert.ts
 create mode 100644 packages/components/nodes/vectorstores/MongoDB/mongodb.png

diff --git a/packages/components/nodes/vectorstores/MongoDB/MongoDBSearchBase.ts b/packages/components/nodes/vectorstores/MongoDB/MongoDBSearchBase.ts
new file mode 100644
index 00000000..e9ef8e9a
--- /dev/null
+++ b/packages/components/nodes/vectorstores/MongoDB/MongoDBSearchBase.ts
@@ -0,0 +1,145 @@
+import {
+    getBaseClasses,
+    getCredentialData,
+    getCredentialParam,
+    ICommonObject,
+    INodeData,
+    INodeOutputsValue,
+    INodeParams
+} from '../../../src'
+
+import { Embeddings } from 'langchain/embeddings/base'
+import { VectorStore } from 'langchain/vectorstores/base'
+import { Document } from 'langchain/document'
+import { MongoDBAtlasVectorSearch } from 'langchain/vectorstores/mongodb_atlas'
+import { Collection, MongoClient } from 'mongodb'
+
+export abstract class MongoDBSearchBase {
+    label: string
+    name: string
+    version: number
+    description: string
+    type: string
+    icon: string
+    category: string
+    baseClasses: string[]
+    inputs: INodeParams[]
+    credential: INodeParams
+    outputs: INodeOutputsValue[]
+    mongoClient: MongoClient
+
+    protected constructor() {
+        this.type = 'MongoDB Atlas'
+        this.icon = 'mongodb.png'
+        this.category = 'Vector Stores'
+        this.baseClasses = [this.type, 'VectorStoreRetriever', 'BaseRetriever']
+        this.credential = {
+            label: 'Connect Credential',
+            name: 'credential',
+            type: 'credential',
+            credentialNames: ['mongoDBUrlApi']
+        }
+        this.inputs = [
+            {
+                label: 'Embeddings',
+                name: 'embeddings',
+                type: 'Embeddings'
+            },
+            {
+                label: 'Database',
+                name: 'databaseName',
+                placeholder: '<DB_NAME>',
+                type: 'string'
+            },
+            {
+                label: 'Collection Name',
+                name: 'collectionName',
+                placeholder: '<COLLECTION_NAME>',
+                type: 'string'
+            },
+            {
+                label: 'Index Name',
+                name: 'indexName',
+                placeholder: '<VECTOR_INDEX_NAME>',
+                type: 'string'
+            },
+            {
+                label: 'Content Field',
+                name: 'textKey',
+                description: 'Name of the field (column) that contains the actual content',
+                type: 'string',
+                default: 'text',
+                additionalParams: true,
+                optional: true
+            },
+            {
+                label: 'Embedded Field',
+                name: 'embeddingKey',
+                description: 'Name of the field (column) that contains the Embedding',
+                type: 'string',
+                default: 'embedding',
+                additionalParams: true,
+                optional: true
+            },
+            {
+                label: 'Top K',
+                name: 'topK',
+                description: 'Number of top results to fetch. Default to 4',
+                placeholder: '4',
+                type: 'number',
+                additionalParams: true,
+                optional: true
+            }
+        ]
+        this.outputs = [
+            {
+                label: 'MongoDB Retriever',
+                name: 'retriever',
+                baseClasses: this.baseClasses
+            },
+            {
+                label: 'MongoDB Vector Store',
+                name: 'vectorStore',
+                baseClasses: [this.type, ...getBaseClasses(MongoDBAtlasVectorSearch)]
+            }
+        ]
+    }
+
+    abstract constructVectorStore(
+        embeddings: Embeddings,
+        collection: Collection,
+        indexName: string,
+        textKey: string,
+        embeddingKey: string,
+        docs: Document<Record<string, any>>[] | undefined
+    ): Promise<VectorStore>
+
+    async init(nodeData: INodeData, _: string, options: ICommonObject, docs: Document<Record<string, any>>[] | undefined): Promise<any> {
+        const credentialData = await getCredentialData(nodeData.credential ?? '', options)
+        const databaseName = nodeData.inputs?.databaseName as string
+        const collectionName = nodeData.inputs?.collectionName as string
+        const indexName = nodeData.inputs?.indexName as string
+        let textKey = nodeData.inputs?.textKey as string
+        let embeddingKey = nodeData.inputs?.embeddingKey as string
+        const embeddings = nodeData.inputs?.embeddings as Embeddings
+        const topK = nodeData.inputs?.topK as string
+        const k = topK ? parseFloat(topK) : 4
+        const output = nodeData.outputs?.output as string
+
+        let mongoDBConnectUrl = getCredentialParam('mongoDBConnectUrl', credentialData, nodeData)
+
+        this.mongoClient = new MongoClient(mongoDBConnectUrl)
+        const collection = this.mongoClient.db(databaseName).collection(collectionName)
+        if (!textKey || textKey === '') textKey = 'text'
+        if (!embeddingKey || embeddingKey === '') embeddingKey = 'embedding'
+        const vectorStore = await this.constructVectorStore(embeddings, collection, indexName, textKey, embeddingKey, docs)
+
+        if (output === 'retriever') {
+            return vectorStore.asRetriever(k)
+        } else if (output === 'vectorStore') {
+            ;(vectorStore as any).k = k
+            return vectorStore
+        }
+        return vectorStore
+    }
+}
diff --git a/packages/components/nodes/vectorstores/MongoDB/MongoDB_Existing.ts b/packages/components/nodes/vectorstores/MongoDB/MongoDB_Existing.ts
new file mode 100644
index 00000000..3cbb36b8
--- /dev/null
+++ b/packages/components/nodes/vectorstores/MongoDB/MongoDB_Existing.ts
@@ -0,0 +1,41 @@
+import { ICommonObject, INode, INodeData } from '../../../src/Interface'
+import { Embeddings } from 'langchain/embeddings/base'
+import { VectorStore } from 'langchain/vectorstores/base'
+import { Document } from 'langchain/document'
+
+import { MongoDBSearchBase } from './MongoDBSearchBase'
+import { Collection } from 'mongodb'
+import { MongoDBAtlasVectorSearch } from 'langchain/vectorstores/mongodb_atlas'
+
+class MongoDBExisting_VectorStores extends MongoDBSearchBase implements INode {
+    constructor() {
+        super()
+        this.label = 'MongoDB Atlas Load Existing Index'
+        this.name = 'MongoDBIndex'
+        this.version = 1.0
+        this.description = 'Load existing data from MongoDB Atlas (i.e: Document has been upserted)'
+    }
+
+    async init(nodeData: INodeData, _: string, options: ICommonObject): Promise<any> {
+        return super.init(nodeData, _, options, undefined)
+    }
+
+    constructVectorStore(
+        embeddings: Embeddings,
+        collection: Collection,
+        indexName: string,
+        textKey: string,
+        embeddingKey: string,
+        _: Document<Record<string, any>>[] | undefined
+    ): Promise<VectorStore> {
+        const mongoDBAtlasVectorSearch = new MongoDBAtlasVectorSearch(embeddings, {
+            collection: collection,
+            indexName: indexName,
+            textKey: textKey,
+            embeddingKey: embeddingKey
+        })
+        return Promise.resolve(mongoDBAtlasVectorSearch)
+    }
+}
+
+module.exports = { nodeClass: MongoDBExisting_VectorStores }
diff --git a/packages/components/nodes/vectorstores/MongoDB/MongoDB_Upsert.ts b/packages/components/nodes/vectorstores/MongoDB/MongoDB_Upsert.ts
new file mode 100644
index 00000000..80dfbf19
--- /dev/null
+++ b/packages/components/nodes/vectorstores/MongoDB/MongoDB_Upsert.ts
@@ -0,0 +1,58 @@
+import { ICommonObject, INode, INodeData } from '../../../src/Interface'
+import { Embeddings } from 'langchain/embeddings/base'
+import { Document } from 'langchain/document'
+
+import { flatten } from 'lodash'
+import { VectorStore } from 'langchain/vectorstores/base'
+import { MongoDBSearchBase } from './MongoDBSearchBase'
+import { Collection } from 'mongodb'
+import { MongoDBAtlasVectorSearch } from 'langchain/vectorstores/mongodb_atlas'
+
+class MongoDBUpsert_VectorStores extends MongoDBSearchBase implements INode {
+    constructor() {
+        super()
+        this.label = 'MongoDB Upsert Document'
+        this.name = 'MongoDBUpsert'
+        this.version = 1.0
+        this.description = 'Upsert documents to MongoDB Atlas'
+        this.inputs.unshift({
+            label: 'Document',
+            name: 'document',
+            type: 'Document',
+            list: true
+        })
+    }
+
+    constructVectorStore(
+        embeddings: Embeddings,
+        collection: Collection,
+        indexName: string,
+        textKey: string,
+        embeddingKey: string,
+        docs: Document<Record<string, any>>[]
+    ): Promise<VectorStore> {
+        return MongoDBAtlasVectorSearch.fromDocuments(docs, embeddings, {
+            collection: collection,
+            indexName: indexName,
+            textKey: textKey,
+            embeddingKey: embeddingKey
+        })
+    }
+
+    async init(nodeData: INodeData, _: string, options: ICommonObject): Promise<any> {
+        const docs = nodeData.inputs?.document as Document[]
+
+        const flattenDocs = docs && docs.length ? flatten(docs) : []
+        const finalDocs = []
+        for (let i = 0; i < flattenDocs.length; i += 1) {
+            if (flattenDocs[i] && flattenDocs[i].pageContent) {
+                const document = new Document(flattenDocs[i])
+                finalDocs.push(document)
+            }
+        }
+
+        return super.init(nodeData, _, options, flattenDocs)
+    }
+}
+
+module.exports = { nodeClass: MongoDBUpsert_VectorStores }
diff --git a/packages/components/nodes/vectorstores/MongoDB/mongodb.png b/packages/components/nodes/vectorstores/MongoDB/mongodb.png
new file mode 100644
index 0000000000000000000000000000000000000000..5586fe0ac672f7997014d814389c1d6c436d9d0c
GIT binary patch
literal 3741
zcmb`Jc{J4R`^U$YrHqk1Gck`mMA>7^WSwCm4MMgVON56fyHH5hDNHC^_9aWUvCB4O
z8H(&h8tVuN^<;@+e)IkLp80jo@0{;n-_N<u{dwKj`?{|CzV35>&iOnv!JxUh&Ts($
z0B(aTsO$81EC9fCfrEukvneL|008rziIKS;+S};wO8sMVEzz+`Q>L?XktQA7i?7g`
z4#!S9-+HcfJoPC3m&8yR0hXysbjD@Fj@91-{(6BtcDfX69~WV$#}dQI@c(CEs2-Ji
z3})=awe;V;{9g9oVd={Dx$4JE9~t0QrfZX~!l-ZzjFG-=Mld6Wk$WuB8?nt$!DXp2
zA{jtW`8~__*`?#Ddj$%csd`+85x@YC3Ozd0jZxDs8+NSP=lov_4!^bIn2(@Wb1KmI
zeQ5?C!|_NzH{U9ZTY-Lf&CPp74mVO26Y2xPik)wyDx@_=WH-kby-bNL@t`kP+VU*!
zMIb%Pv(oTIb53&Q1Isj}>b9axVtid|c~O0OV6{b48__O9y|ts!@tIafM@w^Ct!tLP
zWvXIytw$HB-7DAhRVOLAA*8#rGq%*Hr?ZRrvLu9v!=|f6mw2XC#nAn|UEO!9ZRkL^
znzqto_?N%4o5BArxOnsMDg8@0^uLZV1DGYbzZVang)ybf%be<_9mO-NZPHei@kg{w
zHt0v%mS!o3<|=L5u&?PPf90M<(P1Y@_@@(bP(<<27xj*E(R!&8=pXi-w!S;5z;lpZ
zOXG?cjMy*qqtS95zNKLB@E}&n0XL+l!8$}{>@<zWUgnd$+`8255UfI_sXe06EJhJg
zi#7x~+J|$r={j0;9&N7OmIn4zr5V$14I1~88nKzmI#RUXA3GY&^!NGy<EBL=%Fht{
zqp>qm%Pwq{-uR7OZQg<R!l?~kZ^W%%graA-NU8#Y96zl904^&7l#Y22(;$T{aHjuE
z-?tKM!1|5JyJK*bF78&Ju{^U+;h9P?B3s<985tQOq~+*I>+G|2+k{V(G`?S`M#-Ug
z9ZxcbwE)+;C(f}!X`0gpz=>NABj@G)SBb$r%=e6Ac;X@o=ikBv)h}?xJ?g?5l>N#r
ztN^!JlmDOv;vX3jZ&+MhHlz7XOhnWr!~a5{LRo&e|D8;O6_7<lh7`$A8=J0)hcsm{
z+7=Ncvxe^{gA3|Tf&;{A1qc$eEHRjzWCNhj!6|>&Yz-+h@ZvAF)yg1Ct74yeTzy{M
z>#J$YPPlzCwCbI%Q@&HXQaf^c2WiWK;8S{ntp76F*PNP|0l7)MDkSk|o%-N++^JJD
zK9oCdQBlAiXO~#j{)9}<42mPtv+MmMueqdPSX6KmucixSJ!P$hCO!aN<gs0_?JI<4
zlpuRzMdvLgCY8Km-9#&+v=UMbg@C+}+;FjL7z0b7sq$^@&;-cXb0k&+xNNZ*|MCuA
z$|Vox)M(ifSV|L6P?K+mq24f?H#VPZS4Zi|PBeUk*6mi`clGaiT%m2j*W{D~k*S`A
zAjevf{1dv@;!4i-LJ!_*>*ZS*_~8$1^IR(^`S8VcS#h;SXc>?<yV$-qGkFtayPS6m
z_o$414dnhA*OmnO1w{v$Pry(^<E$~?jC50wS(~vdIiZ*r-X^)DQO}%n2rJ*c>vx2|
zzALm#0oLp=U$&?{mjXwH%3gME7dD;G6U(RxE}Mcj9xSwoUQ^n#yQ{7hTiVRj)IPQz
zDe?JWyrKqkFb}HUe9-#tbO&rug3sG*;QU$~5*x}br)2l|2Ne{_-B0ROE3zl7X7(f2
zEAd{y(cqK&ySp*l!@MN?<+?kDf$d9u(AiMqolS4>R;y2Xq2R(7lNFGyE5F?Z)>z>j
z?alZZ-dxCjg5ddm?Sl=F9$){o&ES3Vq)=jc-}$mmojIuMmbCZCaN=v8=fGI2ro@I9
zP<g^*ETR#3T60VQKKM8R9GSZf>uJ50_#CGQA!J3W=w@C|L;9<petSKd(5RMleyMn3
z1r({4n9p|=4!KngMItuoCAHpE7#E?*B{|@dJ@MO;xoZJEB}t8F&UWCpV3skJ8ktK7
z_tyev1tX9ZqE|NJqFl*oXCa{{spXQ9kw+N^lk5kOf&$@zgpZEi{9u%?m%&!>ogR|@
zbrv%)(Dc5<z_)AZy*@i`$xf=T78S^AV#WTk-={Gh?i(FDo6?9!WW;p&hz?9)7?loQ
z{4N?kj(S4%#co&z+=T%T(AJZ;RAXTz-5=W~t5bJ}9ffnRXrgC)UZut9#tN$fT6^6w
zXjfT>J)CaxK@L^=Ue`YVlPuthW}ka%7Fq9OL)Q3Hzx#Mg(c5IIUj=XlvF9vbEivnf
z2*ZDc_YgOl&u2g`asw1byEdjiumD)Jd=Ul8025!rjAIx#z$xt|WcT5LVLatquw?kX
z<PROXe^4oM00iAF^|?oeRJb7}?vI~y+y&TYF9zZR9KNZyAWa}Eai#*P6Tl}Bt)|74
zSpaO75{at;fYTxhn<1Q?N2O)GmqY;QnQ=F+h;A@hUdJ0RVZ{}-B&T^Y$_X@?{L##x
zton&%A?T7gVVp@_jx9(6L{JI=aiF<VnHTT{a^i#`ZP=n!3bkqk!u+0j5he&C&su@N
zK9>zyyLfxA$Wwf9T`N;oFb;y!1x+DVvPH;_4hU>(m6JU2<>6w{tlo09j<^3SH(<n)
z-@FS5yz=0ljyFG(eGl`zFdJs~AH<?YC;bG_GhUsRx!(}(E?@SzKN2Fs@_ryQ6kkO-
z?{)XzUkUv6k!$`mX(>c89IbeV+;TYkl;SgbcC&MXNDj|}^==SjFPnS%A~5n(-M-C`
z8SWqV)0!bX+`mY5iY{cB1F35{p#G403g;Ry)%|p>nrnlNdsU)1-q;HA76856{aN82
zN920}xVsBKcc}z?7MI7P;Qwa@-x=%!Nc@}#<Wm^z5VYina-E-meDZ6XVJjJdYNC5t
zv%Jq2jOe7Rs;c%ZH-F$6;W<Sa9n%&YHNFG0tQ;5newpyh_p?#JtEe3+!Z7^ERNpUM
ztUN;k6Hs&njSUFzt_futI1IKXb%QNcZL_J;t7vzvbM=U~)#44+n6*3ghcW&Xg4><#
z04Od_)JCe}q~C;gONEC&`>h*=<5pJXg*VyOWIf-l;g6z>4%LfBM}3(xY^*0&+&tK^
zUqBI&#M9G;b#mx7_URW~9=_Wb#cg{@>((RUnuIR_ULz1H#gXQeeggfg;FTUfC&98c
zv2HqQi_rKN)pK8HPUyyzC+h4F_x_hsb)eKg4OhLhNDtZ$%FUlg5<jn0A-IH8#494;
zU-KUmIyCxQrPTQ09OcwAG|5fy#L9_9X}FmCOylkUwFI7qou=EB#dXUwDmIU62aLA(
zvgQMqvR>Ee`4}lYFP%Qr2a$Q7Y!07Tda0<HF)yrA_RN~hxiyHT{ff>l{0U(CbV9?G
z6c)EID<kw^Ed)F?z$+>w?^eF5J{*|!GN*G$G4Q5&@sr}iAcZQae|(A89tp<m-OVg=
z)+ETC7k`ay9qLYc*4JzLK4Mj(9hxa2p)C7n<_df)VzS`qTkqe5ENqU@Q;)1h>iw5<
z@zzek`)>q*NropSjqU4TTwKa=>0pU>HWunUq@hLWx3Z4DMnq5V5APnDu0OxEEKm{m
zGgGxWILA2O7nb$BO1Mw+?x>*r0=0LdJ8ot_!A|`4PyD#v3D{*&Sc&iUS?f8u2BCfM
z{P+$ks#ZoC@j37At~zt>+2RSfZas5Z>nn`L7TJ>u*XzG3^6}-a&86y%Ki&yP^(L`M
z45DTnRC1Qkic!=D!4`Z|**uk;J512mkGj0lPyWH$Re1JvpJ2hSTq}Y|$pie~_wWc=
z6xS7K<FqT5Rr_@RPP}<a_d|qUdtL>%wfO7l*oX!FbhG^YYp8=kFQ+~|1G}D>HG4k%
zU<)p3sh|gP=dE2}m0RS_S5GLI6MlkALrS#%CKQzY*h^Q=4m5H74E7dobHJoSYhy;9
znE6TEDVoZD<I2*|PilXgkkOtP?sXpTN;MaYbBqe*X9o-Sp;U><V^P7DdS9{8&<lyF
z2P?MhHWYx4on^>(ecb1ZtV-i~whn5SBHHG{2nK&~Qi~^#rVj?tBWKPZYVchduFxy5
zy7y5*m^y6#RjhWwzAoUll_Cgby8Yacl7tB4d=e_CtemNn_{;^v78txW%u~<(5-UJ#
z#<?LU{l6*<Qv9tPZX)XQ*X9>r)SGA!)%WW;bcqjd+MGK-B`tE%S2W~8N{6$9zs%|z
zf%q|Jbva4lY4t51YxC_gPXMRr&Qye%&`7K%TS>7U0tHUDV=r-hI_k61zpDLb(UMFd
zvc{YjA}y~3Z_(B#oA<A}xr;zHO!Z@D->jxz5`iTB0;Uq2WD5&Do&8y%sE{Z*$)IS0
tLACkG!7qJ7*HW-wMKKxE3+#3P=^7KM_f;;R=|9^513e6?O4lLce*i)hLc9O~

literal 0
HcmV?d00001