Merge branch 'dev' of github.com:ballerine-io/ballerine into dev

alonp99 · alonp99 · commit 799a74e44aa3 · 2025-05-21T17:01:38.000+03:00
diff --git a/apps/backoffice-v2/src/common/constants.ts b/apps/backoffice-v2/src/common/constants.ts
@@ -7,6 +7,7 @@ export const DOWNLOAD_ONLY_MIME_TYPES = [
 
 export const BALLERINE_CALENDLY_LINK = 'https://calendly.com/d/cp53-ryw-4s3/ballerine-intro';
 
+// Taken from https://gist.github.com/dperini/729294, changed a bit to allow only http/https and make protocol optional
 export const URL_REGEX =
   /((https?):\/\/)?([a-zA-Z0-9-_]+\.)+[a-zA-Z0-9]+(\.[a-z]{2})?(\/[a-zA-Z0-9_#-]+)*(\/)?(\?[a-zA-Z0-9_-]+=[a-zA-Z0-9_-]+(&[a-zA-Z0-9_-]+=[a-zA-Z0-9_-]+)*)?(#[a-zA-Z0-9_-]+)?/;
 
diff --git a/packages/common/src/consts/index.ts b/packages/common/src/consts/index.ts
@@ -172,7 +172,7 @@ export const MatchReasonCode = {
 } as const;
 
 export const URL_PATTERN =
-  /^(https?:\/\/)?((([\da-z]([\da-z-]*[\da-z])*)\.)+[a-z]{2,}|((25[0-5]|2[0-4]\d|1\d{2}|\d{1,2})\.){3}(25[0-5]|2[0-4]\d|1\d{2}|\d{1,2})|localhost)(:\d{1,5})?(\/[\w!$%&'()*+,.:;=@~-]*)*(\?([\w!$%&'()*+,.:;=@~-]+=[\w!$%&'()*+,.:;=@~-]*(&[\w!$%&'()*+,.:;=@~-]+=[\w!$%&'()*+,.:;=@~-]*)*)?)?(#[\w!$%&'()*+,.:;=@~-]*)?$/i;
+  /^(?:https?:\/\/)?(?:\S+(?::\S*)?@)?(?:(?!(?:10|127)(?:\.\d{1,3}){3})(?!(?:169\.254|192\.168)(?:\.\d{1,3}){2})(?!172\.(?:1[6-9]|2\d|3[0-1])(?:\.\d{1,3}){2})(?:[1-9]\d?|1\d\d|2[01]\d|22[0-3])(?:\.(?:1?\d{1,2}|2[0-4]\d|25[0-5])){2}(?:\.(?:[1-9]\d?|1\d\d|2[0-4]\d|25[0-4]))|(?:(?:[a-z0-9\u00A1-\uFFFF][a-z0-9\u00A1-\uFFFF_-]{0,62})?[a-z0-9\u00A1-\uFFFF]\.)+(?:[a-z\u00A1-\uFFFF]{2,}\.?))(?::\d{2,5})?(?:[/?#]\S*)?$/i;
 
 export const MERCHANT_REPORT_STATUSES = [
   'in-progress',
diff --git a/services/workflows-service/src/business-report/business-report.service.ts b/services/workflows-service/src/business-report/business-report.service.ts
@@ -1,8 +1,9 @@
 import { BadRequestException, Injectable, UnprocessableEntityException } from '@nestjs/common';
+import uniqBy from 'lodash/uniqBy';
 import { Business } from '@prisma/client';
-import { TProjectId } from '@/types';
+import { type PrismaTransaction, TProjectId } from '@/types';
 import { parseCsv } from '@/common/utils/parse-csv/parse-csv';
-import { BusinessReportRequestSchema } from '@/common/schemas';
+import { BusinessReportRequestSchema, TBusinessReportRequest } from '@/common/schemas';
 import { PrismaService } from '@/prisma/prisma.service';
 import { BusinessService } from '@/business/business.service';
 import { env } from '@/env';
@@ -113,6 +114,27 @@ export class BusinessReportService {
     return await this.merchantMonitoringClient.count(args);
   }
 
+  private async createBusiness(
+    projectId: string,
+    businessReportRequest: TBusinessReportRequest,
+    transaction?: PrismaTransaction,
+  ) {
+    return await this.businessService.create(
+      {
+        data: {
+          ...(businessReportRequest.correlationId
+            ? { correlationId: businessReportRequest.correlationId }
+            : {}),
+          companyName: businessReportRequest.merchantName || 'Not detected',
+          website: businessReportRequest.websiteUrl || '',
+          country: businessReportRequest.countryCode || '',
+          projectId,
+        },
+      },
+      transaction,
+    );
+  }
+
   async processBatchFile({
     type,
     projectId,
@@ -160,27 +182,54 @@ export class BusinessReportService {
 
     await this.prisma.$transaction(
       async transaction => {
-        const businessCreatePromises = businessReportsRequests.map(async businessReportRequest => {
-          let business =
-            businessReportRequest.correlationId &&
-            (await this.businessService.getByCorrelationId(businessReportRequest.correlationId, [
-              projectId,
-            ]));
+        const businessesLookup = new Map<string, Business>();
+
+        const allCorrelationIds = new Set(
+          businessReportsRequests
+            .map(({ correlationId }) => correlationId)
+            .filter(Boolean) as string[],
+        );
 
-          business ||= await this.businessService.create(
+        if (allCorrelationIds.size > 0) {
+          const businesses = await this.businessService.list(
             {
-              data: {
-                ...(businessReportRequest.correlationId
-                  ? { correlationId: businessReportRequest.correlationId }
-                  : {}),
-                companyName: businessReportRequest.merchantName || 'Not detected',
-                website: businessReportRequest.websiteUrl || '',
-                country: businessReportRequest.countryCode || '',
-                projectId,
+              where: {
+                correlationId: {
+                  in: [...allCorrelationIds],
+                },
               },
             },
+            [projectId],
             transaction,
           );
+          for (const business of businesses) {
+            businessesLookup.set(business.correlationId || business.id, business);
+          }
+        }
+
+        const businessesToCreate = uniqBy(
+          businessReportsRequests.filter(
+            ({ correlationId }) => !!correlationId && !businessesLookup.has(correlationId),
+          ),
+          'correlationId',
+        );
+        if (businessesToCreate.length > 0) {
+          const businesses = await Promise.all(
+            businessesToCreate.map(business =>
+              this.createBusiness(projectId, business, transaction),
+            ),
+          );
+          for (const business of businesses) {
+            businessesLookup.set(business.correlationId || business.id, business);
+          }
+        }
+
+        const businessCreatePromises = businessReportsRequests.map(async businessReportRequest => {
+          let business =
+            businessReportRequest.correlationId &&
+            businessesLookup.get(businessReportRequest.correlationId);
+
+          business ||= await this.createBusiness(projectId, businessReportRequest, transaction);
 
           return {
             businessReportRequest,
diff --git a/services/workflows-service/src/business/business.repository.ts b/services/workflows-service/src/business/business.repository.ts
@@ -33,10 +33,9 @@ export class BusinessRepository {
   async findMany<T extends Prisma.BusinessFindManyArgs>(
     args: Prisma.SelectSubset<T, Prisma.BusinessFindManyArgs>,
     projectIds: TProjectIds,
+    transaction: PrismaClient | PrismaTransaction = this.prismaService,
   ) {
-    return await this.prismaService.business.findMany(
-      this.scopeService.scopeFindMany(args, projectIds),
-    );
+    return await transaction.business.findMany(this.scopeService.scopeFindMany(args, projectIds));
   }
 
   async findManyUnscoped<T extends Prisma.BusinessFindManyArgs>(
diff --git a/services/workflows-service/src/business/business.service.ts b/services/workflows-service/src/business/business.service.ts
@@ -42,41 +42,18 @@ export class BusinessService {
     })(async tx => {
       const business = await this.repository.create(args, tx);
 
-      const businessPayload = (await this.repository.findByIdUnscoped(
-        business.id,
-        {
-          select: {
-            id: true,
-            correlationId: true,
-            companyName: true,
-            metadata: true,
-            createdAt: true,
-            updatedAt: true,
-            project: {
-              select: {
-                customer: {
-                  select: {
-                    id: true,
-                    config: true,
-                  },
-                },
-              },
-            },
-          },
-        },
-        tx,
-      )) as unknown as BusinessPayload;
-
-      if (env.SYNC_UNIFIED_API) {
-        await retry(() => this.unifiedApiClient.createOrUpdateBusiness(businessPayload));
-      }
+      await this.syncBusiness(business, tx);
 
       return business;
     });
   }
 
-  async list(args: Parameters<BusinessRepository['findMany']>[0], projectIds: TProjectIds) {
-    return (await this.repository.findMany(args, projectIds)) as Array<
+  async list(
+    args: Parameters<BusinessRepository['findMany']>[0],
+    projectIds: TProjectIds,
+    transaction?: PrismaTransaction,
+  ) {
+    return (await this.repository.findMany(args, projectIds, transaction)) as Array<
       Business & {
         metadata?: {
           featureConfig?: TCustomerWithFeatures['features'];
@@ -117,38 +94,43 @@ export class BusinessService {
     })(async tx => {
       const business = await this.repository.updateById(id, args, tx);
 
-      const businessPayload = (await this.repository.findByIdUnscoped(
-        business.id,
-        {
-          select: {
-            id: true,
-            correlationId: true,
-            companyName: true,
-            metadata: true,
-            createdAt: true,
-            updatedAt: true,
-            project: {
-              select: {
-                customer: {
-                  select: {
-                    id: true,
-                    config: true,
-                  },
+      await this.syncBusiness(business, tx);
+
+      return business;
+    });
+  }
+
+  private syncBusiness = async (business: { id: string }, tx: PrismaTransaction) => {
+    if (!env.SYNC_UNIFIED_API) {
+      return;
+    }
+
+    const businessPayload = (await this.repository.findByIdUnscoped(
+      business.id,
+      {
+        select: {
+          id: true,
+          correlationId: true,
+          companyName: true,
+          metadata: true,
+          createdAt: true,
+          updatedAt: true,
+          project: {
+            select: {
+              customer: {
+                select: {
+                  id: true,
+                  config: true,
                 },
               },
             },
           },
         },
-        tx,
-      )) as unknown as BusinessPayload;
-
-      if (env.SYNC_UNIFIED_API) {
-        await retry(() => this.unifiedApiClient.createOrUpdateBusiness(businessPayload));
-      }
-
-      return business;
-    });
-  }
+      },
+      tx,
+    )) as unknown as BusinessPayload;
+    await retry(() => this.unifiedApiClient.createOrUpdateBusiness(businessPayload));
+  };
 
   async fetchCompanyInformation({
     registrationNumber,
diff --git a/services/workflows-service/src/common/utils/parse-csv/parse-csv.ts b/services/workflows-service/src/common/utils/parse-csv/parse-csv.ts
@@ -1,6 +1,7 @@
 import { parse } from 'csv-parse';
 import { z, ZodSchema } from 'zod';
 import { AppLoggerService } from '@/common/app-logger/app-logger.service';
+import { ForbiddenException } from '@/errors';
 import fs from 'fs';
 
 export const parseCsv = async <TSchema extends ZodSchema>(
@@ -43,16 +44,22 @@ export const parseCsv = async <TSchema extends ZodSchema>(
           reject(err);
         }
 
+        let hadErrors = false;
         for (const record of records) {
           try {
             const validatedRecord = schema.parse(record);
             results.push(validatedRecord);
           } catch (error) {
-            logger.error('Validation error:', { error });
+            logger.error('Validation error:', { error, record });
+            hadErrors = true;
           }
         }
 
-        resolve(results);
+        if (hadErrors) {
+          reject(new ForbiddenException('Schema errors in CSV'));
+        } else {
+          resolve(results);
+        }
       },
     );
   });
