fix: when source is rss-feed or csv-import, do not parse items in the backend

packages/api/src/services/save_page.ts

@@ -36,6 +36,7 @@ const FORCE_PUPPETEER_URLS = [
   TWEET_URL_REGEX,
   /^((?:https?:)?\/\/)?((?:www|m)\.)?((?:youtube\.com|youtu.be))(\/(?:[\w-]+\?v=|embed\/|v\/)?)([\w-]+)(\S+)?$/,
 ]
+const ALREADY_PARSED_SOURCES = ['puppeteer-parse', 'csv-importer', 'rss-feeder']
 
 const createSlug = (url: string, title?: Maybe<string> | undefined) => {
   const { pathname } = new URL(url)
@@ -52,7 +53,7 @@ const createSlug = (url: string, title?: Maybe<string> | undefined) => {
 
 const shouldParseInBackend = (input: SavePageInput): boolean => {
   return (
-    input.source !== 'puppeteer-parse' &&
+    ALREADY_PARSED_SOURCES.indexOf(input.source) === -1 &&
     FORCE_PUPPETEER_URLS.some((regex) => regex.test(input.url))
   )
 }
@@ -99,7 +100,7 @@ export const savePage = async (
       await createPageSaveRequest({
         userId: user.id,
         url: itemToSave.originalUrl,
-        articleSavingRequestId: clientRequestId,
+        articleSavingRequestId: clientRequestId || undefined,
         state: input.state || undefined,
         labels: input.labels || undefined,
       })

packages/api/test/resolvers/article.test.ts

@@ -185,15 +185,17 @@ const savePageQuery = (
   title: string,
   originalContent: string,
   state: ArticleSavingRequestStatus | null = null,
-  labels: string[] | null = null
+  labels: string[] | null = null,
+  clientRequestId = generateFakeUuid(),
+  source = 'puppeteer-parse'
 ) => {
   return `
     mutation {
       savePage(
         input: {
           url: "${url}",
-          source: "test",
-          clientRequestId: "${generateFakeUuid()}",
+          source: "${source}",
+          clientRequestId: "${clientRequestId}",
           title: "${title}",
           originalContent: "${originalContent}"
           state: ${state}
@@ -605,6 +607,29 @@ describe('Article API', () => {
         expect(savedItem?.labels?.map((l) => l.name)).to.eql(labels)
       })
     })
+
+    context('when the source is rss-feeder', () => {
+      const source = 'rss-feeder'
+      const stub = sinon.stub(createTask, 'enqueueParseRequest')
+
+      before(() => {
+        url = 'https://blog.omnivore.app/new-url-3'
+      })
+
+      after(async () => {
+        await deleteLibraryItemByUrl(url, user.id)
+        sinon.restore()
+      })
+
+      it('does not parse in the backend', async () => {
+        await graphqlRequest(
+          savePageQuery(url, title, originalContent, null, null, '', source),
+          authToken
+        ).expect(200)
+
+        expect(stub).not.to.have.been.called
+      })
+    })
   })
 
   describe('SaveUrl', () => {