Implementation of OkHttp's RequestBody that supports Android's content:// URIs

ContentUriRequestBody.kt

import android.content.ContentResolver
import android.net.Uri
import okhttp3.MediaType
import okhttp3.RequestBody
import okio.BufferedSink
import okio.Okio
import java.lang.IllegalStateException

class ContentUriRequestBody(
        private val contentResolver: ContentResolver,
        private val contentUri: Uri
) : RequestBody() {

    override fun contentType(): MediaType? {
        val contentType = contentResolver.getType(contentUri) ?: return null
        return MediaType.parse(contentType)
    }

    override fun writeTo(sink: BufferedSink) {
        val inputStream = contentResolver.openInputStream(contentUri) 
                ?: throw IllegalStateException("Couldn't open content URI for reading: $contentUri")
        
        Okio.source(inputStream).use { source ->
            sink.writeAll(source)
        }
    }
}

I'm looking forward to trying it out since unfortunately it didn't work out for me previously in my initial attempt ....sent from my phone

…

On Fri, May 22, 2020, 7:43 PM cketti ***@***.***> wrote: ***@***.**** commented on this gist. ------------------------------ I apologise. I didn't mean "why didn't you make a blog about it" as much as "this seems so cool why aren't more people discussing it" @neiljaywarner <https://github.com/neiljaywarner>: No worries, I didn't read it like that. I was asking because I was in the mood for writing. So I created this blog post <https://cketti.de/2020/05/23/content-uris-and-okhttp/> and a sample app <https://github.com/cketti/OkHttpWithContentUri>. I hope it helps. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <https://gist.github.com/8ac927509787d7085a5ef8f866806f0f#gistcomment-3314989>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAXBGYMCVNWHC22AGJRDIJDRS4L3PANCNFSM4NG6TGZQ> .

ah. I didn't notice put vs post. (s3). That doesn't give me the answer yet but it does make me think your stuff is fine :)

I believe the issue is that S3 doesn't support the 'chunked' transfer encoding for uploads. This is what OkHttp will use for large files when the RequestBody doesn't specify a size. You can help OkHttp out by implementing the contentLength() method. To get the size of a document you could use DocumentFile.length().

@cketti What if we want to know the file size to get the progress of upload?