Send file DataContent as a data URI in OpenAI chat requests

[PratikDhanave]

In buildMessageParam (chat completions, RoleUser), a non-image/non-audio DataContent (e.g. a PDF) was sent with the raw base64 payload:

default:
    contents = append(contents, openai.FileContentPart(openai.ChatCompletionContentPartFileFileParam{
        FileData: openai.String(c.Data),   // raw base64 — missing the data-URI prefix
        Filename: openai.String(c.Name),
    }))

DataContent.Data is the raw base64 (see message/content.go), but the OpenAI Chat Completions API requires file_data to be a full data URI — data:<mediatype>;base64,<data> (per OpenAI’s PDF-files guide: "file_data": "data:application/pdf;base64,${base64String}"). Sent as raw base64, the file is rejected/mishandled.

This is the outlier:

• The image branch two cases up already uses c.URI() (chat.go:410).
• The Responses provider sends file_data as c.URI() (responses.go:509), asserted in responses_test.go.
• DataContent.URI() returns exactly data:<mediatype>;base64,<data>.

No test covered the chat non-image DataContent path (only the image case), so it went unnoticed.

Fix

FileData: openai.String(c.URI()). The adjacent audio branch keeps raw c.Data — the input_audio API genuinely expects raw base64, so it is correct and left unchanged.

Test

TestChatDataContentMessage_File_NonStreaming sends a user message with a PDF DataContent and asserts the request body’s file_data is data:application/pdf;base64,cGRmZGF0YQ==. It fails on the old code (raw cGRmZGF0YQ==) and passes with the fix.

[Copilot]

Pull request overview

This PR fixes how message.DataContent files (non-image/non-audio, e.g. PDFs) are serialized into OpenAI Chat Completions requests by sending file_data as a full RFC 2397 data URI (data:<mediatype>;base64,<data>), aligning with the existing image handling and the Responses provider behavior.

Changes:

• Update Chat Completions message building to use DataContent.URI() for file_data in the non-image/non-audio file branch.
• Add a non-streaming chat test that asserts PDF DataContent is sent as a data URI (not raw base64).

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
provider/openaiprovider/chat.go	Switch non-image/non-audio DataContent file_data from raw base64 to c.URI() (data URI).
provider/openaiprovider/chat_test.go	Add coverage ensuring PDF DataContent is emitted as a data URI in chat requests.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.