Facing issue in ImageVariation call

[hassanmirza444]

I had integrate this library in my project. It is Working fine for content generation("v1/completions") and Image generation("v1/images/generations"). But When I start using ImageVariation ("v1/images/variations"), I had faced issue that I am unable to fix,

D 1-->file:///storage/emulated/0/DCIM/Camera/IMG_20230125_180236381.jpg 2023-01-25 18:02:38.499 8356-8356 hssn com.sparklab.ai D 2-->file:///data/user/0/com.sparklab.ai/files/sparklab/sparklab_hssnmirza.png 2023-01-25 18:02:38.975 8356-8396 OpenGLRenderer com.sparklab.ai D endAllActiveAnimators on 0xb4000076461a44d0 (RippleDrawable) with handle 0xb400007556108220 2023-01-25 18:02:40.404 8356-8583 System.out com.sparklab.ai I HttpClient: REQUEST: https://api.openai.com/v1/images/variations 2023-01-25 18:02:40.404 8356-8583 System.out com.sparklab.ai I METHOD: HttpMethod(value=POST) 2023-01-25 18:02:40.404 8356-8583 System.out com.sparklab.ai I COMMON HEADERS 2023-01-25 18:02:40.404 8356-8583 System.out com.sparklab.ai I -> Accept: application/json 2023-01-25 18:02:40.404 8356-8583 System.out com.sparklab.ai I -> Accept-Charset: UTF-8 2023-01-25 18:02:40.404 8356-8583 System.out com.sparklab.ai I -> Authorization: Bearer <apiKey> 2023-01-25 18:02:40.404 8356-8583 System.out com.sparklab.ai I CONTENT HEADERS 2023-01-25 18:02:40.404 8356-8583 System.out com.sparklab.ai I -> Content-Type: multipart/form-data; boundary=-651c0b63-3b3247d1-5ce2608bf515725-441b30e6-4507d4a361a52508-1fc4734e- 2023-01-25 18:02:40.535 8356-8585 TrafficStats com.sparklab.ai D tagSocket(96) with statsTag=0xffffffff, statsUid=-1 2023-01-25 18:02:42.753 8356-8585 System.out com.sparklab.ai I HttpClient: RESPONSE: 400 BAD REQUEST 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I METHOD: HttpMethod(value=POST) 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I FROM: https://api.openai.com/v1/images/variations 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I COMMON HEADERS 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> access-control-allow-origin: * 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> connection: keep-alive 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> content-length: 161 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> content-type: application/json 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> date: Wed, 25 Jan 2023 13:02:40 GMT 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> openai-organization: user-zinuwdnhn5tnrbfqaxqlrm8u 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> openai-processing-ms: 147 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> openai-version: 2020-10-01 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> strict-transport-security: max-age=15724800; includeSubDomains 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> x-android-received-millis: 1674651762739 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> x-android-response-source: NETWORK 400 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> x-android-selected-protocol: http/1.1 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> x-android-sent-millis: 1674651761388 --------- beginning of crash 2023-01-25 18:02:42.754 8356-8585 System.out com.sparklab.ai I -> x-request-id: 8ba8f7abb134b74cd3ae5b2d6f41c489 2023-01-25 18:02:42.798 8356-8592 AndroidRuntime com.sparklab.ai E FATAL EXCEPTION: DefaultDispatcher-worker-8 Process: com.sparklab.ai, PID: 8356 com.aallam.openai.api.exception.OpenAIHttpException at com.aallam.openai.client.internal.http.HttpTransport.perform(HttpTransport.kt:24) at com.aallam.openai.client.internal.http.HttpTransport$perform$1.invokeSuspend(Unknown Source:15) at kotlin.coroutines.jvm.internal.BaseContinuationImpl.resumeWith(ContinuationImpl.kt:33) at kotlinx.coroutines.DispatchedTask.run(DispatchedTask.kt:106) at kotlinx.coroutines.internal.LimitedDispatcher.run(LimitedDispatcher.kt:42) at kotlinx.coroutines.scheduling.TaskImpl.run(Tasks.kt:95) at kotlinx.coroutines.scheduling.CoroutineScheduler.runSafely(CoroutineScheduler.kt:570) at kotlinx.coroutines.scheduling.CoroutineScheduler$Worker.executeTask(CoroutineScheduler.kt:750) at kotlinx.coroutines.scheduling.CoroutineScheduler$Worker.runWorker(CoroutineScheduler.kt:677) at kotlinx.coroutines.scheduling.CoroutineScheduler$Worker.run(CoroutineScheduler.kt:664) Suppressed: kotlinx.coroutines.DiagnosticCoroutineContextException: [StandaloneCoroutine{Cancelling}@9c8cf2c, Dispatchers.Default] Caused by: com.aallam.openai.api.exception.OpenAIAPIException: (statusCode=400, body='{ "error": { "code": null, "message": "Uploaded image must be a PNG and less than 4 MB.", "param": null, "type": "invalid_request_error" } } ') at com.aallam.openai.client.internal.http.HttpTransport.bodyOrThrow(HttpTransport.kt:32) at com.aallam.openai.client.internal.http.HttpTransport.perform(HttpTransport.kt:22) ... 9 more 2023-01-25 18:02:42.832 8356-8592 Process com.sparklab.ai I Sending signal. PID: 8356 SIG: 9
My Image is already in .png format and size is in KBs. So, Please let me know how can I fix it.

[hassanmirza444]

Logs in organised way.

D loaded /vendor/lib64/egl/libGLES_mali.so D maxBitmapSize: 3243 D Uri scheme: file D Image size: [393:524] I Should crop: true D 1-->file:///storage/emulated/0/DCIM/Camera/IMG_20230125_180236381.jpg D 2-->file:///data/user/0/com.sparklab.ai/files/sparklab/sparklab_hssnmirza.png D endAllActiveAnimators on 0xb4000076461a44d0 (RippleDrawable) with handle 0xb400007556108220 I HttpClient: REQUEST: https://api.openai.com/v1/images/variations I METHOD: HttpMethod(value=POST) I COMMON HEADERS I -> Accept: application/json I -> Accept-Charset: UTF-8 I -> Authorization: Bearer <apiKey> I CONTENT HEADERS I -> Content-Type: multipart/form-data; boundary=-651c0b63-3b3247d1-5ce2608bf515725-441b30e6-4507d4a361a52508-1fc4734e- D tagSocket(96) with statsTag=0xffffffff, statsUid=-1 I HttpClient: RESPONSE: 400 BAD REQUEST I METHOD: HttpMethod(value=POST) I FROM: https://api.openai.com/v1/images/variations I COMMON HEADERS I -> access-control-allow-origin: * I -> connection: keep-alive I -> content-length: 161 I -> content-type: application/json I -> date: Wed, 25 Jan 2023 13:02:40 GMT I -> openai-organization: user-zinuwdnhn5tnrbfqaxqlrm8u I -> openai-processing-ms: 147 I -> openai-version: 2020-10-01 I -> strict-transport-security: max-age=15724800; includeSubDomains I -> x-android-received-millis: 1674651762739 I -> x-android-response-source: NETWORK 400 I -> x-android-selected-protocol: http/1.1 I -> x-android-sent-millis: 1674651761388 --------- beginning of crash I -> x-request-id: 8ba8f7abb134b74cd3ae5b2d6f41c489 E FATAL EXCEPTION: DefaultDispatcher-worker-8 Process: com.sparklab.ai, PID: 8356 com.aallam.openai.api.exception.OpenAIHttpException at com.aallam.openai.client.internal.http.HttpTransport.perform(HttpTransport.kt:24) at com.aallam.openai.client.internal.http.HttpTransport$perform$1.invokeSuspend(Unknown Source:15) at kotlin.coroutines.jvm.internal.BaseContinuationImpl.resumeWith(ContinuationImpl.kt:33) at kotlinx.coroutines.DispatchedTask.run(DispatchedTask.kt:106) at kotlinx.coroutines.internal.LimitedDispatcher.run(LimitedDispatcher.kt:42) at kotlinx.coroutines.scheduling.TaskImpl.run(Tasks.kt:95) at kotlinx.coroutines.scheduling.CoroutineScheduler.runSafely(CoroutineScheduler.kt:570) at kotlinx.coroutines.scheduling.CoroutineScheduler$Worker.executeTask(CoroutineScheduler.kt:750) at kotlinx.coroutines.scheduling.CoroutineScheduler$Worker.runWorker(CoroutineScheduler.kt:677) at kotlinx.coroutines.scheduling.CoroutineScheduler$Worker.run(CoroutineScheduler.kt:664) Suppressed: kotlinx.coroutines.DiagnosticCoroutineContextException: [StandaloneCoroutine{Cancelling}@9c8cf2c, Dispatchers.Default] Caused by: com.aallam.openai.api.exception.OpenAIAPIException: (statusCode=400, body='{ "error": { "code": null, "message": "Uploaded image must be a PNG and less than 4 MB.", "param": null, "type": "invalid_request_error" } } ') at com.aallam.openai.client.internal.http.HttpTransport.bodyOrThrow(HttpTransport.kt:32) at com.aallam.openai.client.internal.http.HttpTransport.perform(HttpTransport.kt:22) ... 9 more I Sending signal. PID: 8356 SIG: 9

[aallam]

Hi @hassanmirza444, I think you are trying to upload a jpg image, while the API only accepts png:

1. The error from openai API is: Uploaded image must be a PNG and less than 4 MB.
2. The image you are trying to upload is: IMG_20230125_180236381.jpg

PS: I've deleted your API key from the stacktraces above (and edit history)

[hassanmirza444]

Yes brother, What I am doing is I select a file from gallery, and Logged it as 1.
1-->file:///storage/emulated/0/DCIM/Camera/IMG_20230125_180236381.jpg
Then I had converted it in png. Also, I have show it in an imageView. its completely fine.
2-->file:///data/user/0/com.sparklab.ai/files/sparklab/sparklab_hssnmirza.png

[aallam]

Can you share how did you convert it from jpg to png please ?

[hassanmirza444]

@aallam Here is code, on Image selected from gallery and conversion to png

` override fun onActivityResult(requestCode: Int, resultCode: Int, data: Intent?) {
super.onActivityResult(requestCode, resultCode, data)
if (resultCode == Activity.RESULT_OK) {
//Image Uri will not be null for RESULT_OK

        Log.d("hssn", "1-->" + data?.data!!.toString())
        seletedPngFileUri = convertJpgUriToPngUri(data.data!!, "hssnmirza")!!
        Log.d("hssn", "2-->" + seletedPngFileUri.toString())
        binding.ivInput.setImageURI(seletedPngFileUri)



    } else if (resultCode == ImagePicker.RESULT_ERROR) {
        Toast.makeText(requireContext(), ImagePicker.getError(data), Toast.LENGTH_SHORT)
            .show()
    } else {
        Toast.makeText(requireContext(), "Task Cancelled", Toast.LENGTH_SHORT).show()
    }
}

 private fun convertJpgUriToPngUri(sourceuri: Uri, fName: String): Uri? {
    val sourceFilename = sourceuri.path
    val folderpath = context?.getFilesDir().toString() + File.separator + "sparklab"
    val folder = File(folderpath)
    var destinationFilename = File(folderpath + File.separator + "sparklab_" + fName + ".png")
    try {
        // Make sure the Pictures directory exists.
        folder.mkdirs()
        destinationFilename.createNewFile()
    } catch (e: java.lang.Exception) {
        Log.d("hssn---0", e.toString())
    }


    var bis: BufferedInputStream? = null
    var bos: BufferedOutputStream? = null
    try {
        bis = BufferedInputStream(FileInputStream(sourceFilename))
        bos = BufferedOutputStream(FileOutputStream(destinationFilename, false))
        val buf = ByteArray(1024)
        bis.read(buf)
        do {
            bos.write(buf)
        } while (bis.read(buf) != -1)
    } catch (e: IOException) {
        e.printStackTrace()
        Log.d("hssn", e.toString())
        // destinationFilename = null
    } finally {
        try {
            bis?.close()
            bos?.close()
        } catch (e: IOException) {
            e.printStackTrace()
        }
    }
    return Uri.fromFile(destinationFilename)
}

`

[aallam]

Correct me if I am wrong, in this code snippet you make a copy of the file with different extension, not actually converting it, right ?

[hassanmirza444]

No , I am wrong. You are right. Issue was on my side. Now, it is fixed using below function.

` private fun convertJpgToPng(sourceuri: Uri): Uri {
val bitmap = MediaStore.Images.Media.getBitmap(context?.getContentResolver(), sourceuri);

    //folder and file creation
    val folderpath = context?.getFilesDir().toString() + File.separator + "sparklab"
    val folder = File(folderpath)
    val destinationFilename =
        File(folderpath + File.separator + "sparklab_" + "hssn_mirza" + ".png")
    try {
        // Make sure the directory exists.
        folder.mkdirs()
        destinationFilename.createNewFile()
    } catch (e: java.lang.Exception) {
        Log.d("hssn---10", e.toString())
    }

    val outputStream = FileOutputStream(destinationFilename)
    bitmap.compress(Bitmap.CompressFormat.PNG, 100, outputStream)
    outputStream.close()

    return Uri.fromFile(destinationFilename)
}`


Thanks you so much.