FIX: Correct prompt format for img2text used in our AI Bot PDF Rag pipeline (#1323)

xfalcox · web-flow · commit 1b71330cea42 · 2025-05-07T16:48:56.000-03:00
diff --git a/lib/utils/image_to_text.rb b/lib/utils/image_to_text.rb
@@ -127,14 +127,15 @@ def extract_text_from_page(page)
       messages = [
         {
           type: :user,
-          content:
+          content: [
             "The following text was extracted from an image using OCR. Please enhance, correct, and structure this content while maintaining the original text:\n\n#{raw_text}",
-          upload_ids: [page.id],
+            { upload_id: page.id },
+          ],
         },
       ]
     else
       messages = [
-        { type: :user, content: "Please OCR the content in the image.", upload_ids: [page.id] },
+        { type: :user, content: ["Please OCR the content in the image.", { upload_id: page.id }] },
       ]
     end
     prompt = DiscourseAi::Completions::Prompt.new(system_message, messages: messages)

Original file line number	Diff line number	Diff line change
`@@ -127,14 +127,15 @@ def extract_text_from_page(page)`
`127`	`127`	`messages = [`
`128`	`128`	`{`
`129`	`129`	`type: :user,`
`130`		`- content:`
	`130`	`+ content: [`
`131`	`131`	`"The following text was extracted from an image using OCR. Please enhance, correct, and structure this content while maintaining the original text:\n\n#{raw_text}",`
`132`		`- upload_ids: [page.id],`
	`132`	`+ { upload_id: page.id },`
	`133`	`+ ],`
`133`	`134`	`},`
`134`	`135`	`]`
`135`	`136`	`else`
`136`	`137`	`messages = [`
`137`		`- { type: :user, content: "Please OCR the content in the image.", upload_ids: [page.id] },`
	`138`	`+ { type: :user, content: ["Please OCR the content in the image.", { upload_id: page.id }] },`
`138`	`139`	`]`
`139`	`140`	`end`
`140`	`141`	`prompt = DiscourseAi::Completions::Prompt.new(system_message, messages: messages)`