@paul burke的答案适用于API级别19及以上的相机和画廊图片，但如果你的Android项目的最低SDK设置为19以下，它就不起作用，上面提到的一些答案并不适用于画廊和相机。好吧，我已经修改了@paul burke的代码，它适用于API级别低于19的代码。下面是代码。

public static String getPath(final Context context, final Uri uri) {

    final boolean isKitKat = Build.VERSION.SDK_INT >=
                             Build.VERSION_CODES.KITKAT;
    Log.i("URI",uri+"");
    String result = uri+"";

    // DocumentProvider
    // if (isKitKat && DocumentsContract.isDocumentUri(context, uri)) {
    if (isKitKat && (result.contains("media.documents"))) {

        String[] ary = result.split("/");
        int length = ary.length;
        String imgary = ary[length-1];
        final String[] dat = imgary.split("%3A");

        final String docId = dat[1];
        final String type = dat[0];

        Uri contentUri = null;
        if ("image".equals(type)) {
            contentUri = MediaStore.Images.Media.EXTERNAL_CONTENT_URI;
        }
        else if ("video".equals(type)) {
        }
        else if ("audio".equals(type)) {
        }

        final String selection = "_id=?";
        final String[] selectionArgs = new String[] {
            dat[1]
        };

        return getDataColumn(context, contentUri, selection, selectionArgs);
    }
    else
    if ("content".equalsIgnoreCase(uri.getScheme())) {
        return getDataColumn(context, uri, null, null);
    }
    // File
    else if ("file".equalsIgnoreCase(uri.getScheme())) {
        return uri.getPath();
    }

    return null;
}

public static String getDataColumn(Context context, Uri uri, String selection,
                                   String[] selectionArgs) {
    Cursor cursor = null;
    final String column = "_data";
    final String[] projection = {
            column
    };

    try {
        cursor = context.getContentResolver().query(uri, projection, selection, selectionArgs,
                null);
        if (cursor != null && cursor.moveToFirst()) {
            final int column_index = cursor.getColumnIndexOrThrow(column);
            return cursor.getString(column_index);
        }
    }
    finally {
        if (cursor != null)
            cursor.close();
    }
    return null;
}

请尽量避免使用takePersistableUriPermission方法，因为它会为我引发运行时异常。 ／** 从图库中选择。 * /

public void selectFromGallery() {
    if (Build.VERSION.SDK_INT < AppConstants.KITKAT_API_VERSION) {

        Intent intent = new Intent(); 
        intent.setType("image/*");
        intent.setAction(Intent.ACTION_GET_CONTENT);
        ((Activity)mCalledContext).startActivityForResult(intent,AppConstants.GALLERY_INTENT_CALLED);

    } else {

        Intent intent = new Intent(Intent.ACTION_GET_CONTENT);
        intent.setType("image/*");
        ((Activity)mCalledContext).startActivityForResult(intent, AppConstants.GALLERY_AFTER_KITKAT_INTENT_CALLED);
    }
}

OnActivity的结果处理图像数据:

@Override onActivityResult(int requestCode, int resultCode, Intent data) {

    //gallery intent result handling before kit-kat version
    if(requestCode==AppConstants.GALLERY_INTENT_CALLED 
            && resultCode == RESULT_OK) {

        Uri selectedImage = data.getData();
        String[] filePathColumn = {MediaStore.Images.Media.DATA};
        Cursor cursor = getContentResolver().query(selectedImage,filePathColumn, null, null, null);
        cursor.moveToFirst();
        int columnIndex = cursor.getColumnIndex(filePathColumn[0]);
        String filePath = cursor.getString(columnIndex);
        cursor.close();
        photoFile = new File(filePath);
        mImgCropping.startCropImage(photoFile,AppConstants.REQUEST_IMAGE_CROP);

    }
    //gallery intent result handling after kit-kat version
    else if (requestCode == AppConstants.GALLERY_AFTER_KITKAT_INTENT_CALLED 
            && resultCode == RESULT_OK) {

        Uri selectedImage = data.getData();
        InputStream input = null;
        OutputStream output = null;

        try {
            //converting the input stream into file to crop the 
            //selected image from sd-card.
            input = getApplicationContext().getContentResolver().openInputStream(selectedImage);
            try {
                photoFile = mImgCropping.createImageFile();
            } catch (IOException e) {
                e.printStackTrace();
            }catch(Exception e) {
                e.printStackTrace();
            }
            output = new FileOutputStream(photoFile);

            int read = 0;
            byte[] bytes = new byte[1024];

            while ((read = input.read(bytes)) != -1) {
                try {
                    output.write(bytes, 0, read);
                } catch (IOException e) {
                    e.printStackTrace();
                }
            }
        } catch (FileNotFoundException e) {
            e.printStackTrace();
        } catch (IOException e) {
            e.printStackTrace();
        } catch (Exception e) {
            e.printStackTrace();
        }


    }

这不需要特殊的权限，并且与存储访问框架以及非官方的ContentProvider模式(在_data字段中的文件路径)一起工作。

/**
 * Get a file path from a Uri. This will get the the path for Storage Access
 * Framework Documents, as well as the _data field for the MediaStore and
 * other file-based ContentProviders.
 *
 * @param context The context.
 * @param uri The Uri to query.
 * @author paulburke
 */
public static String getPath(final Context context, final Uri uri) {

    final boolean isKitKat = Build.VERSION.SDK_INT >= Build.VERSION_CODES.KITKAT;

    // DocumentProvider
    if (isKitKat && DocumentsContract.isDocumentUri(context, uri)) {
        // ExternalStorageProvider
        if (isExternalStorageDocument(uri)) {
            final String docId = DocumentsContract.getDocumentId(uri);
            final String[] split = docId.split(":");
            final String type = split[0];

            if ("primary".equalsIgnoreCase(type)) {
                return Environment.getExternalStorageDirectory() + "/" + split[1];
            }

            // TODO handle non-primary volumes
        }
        // DownloadsProvider
        else if (isDownloadsDocument(uri)) {

            final String id = DocumentsContract.getDocumentId(uri);
            final Uri contentUri = ContentUris.withAppendedId(
                    Uri.parse("content://downloads/public_downloads"), Long.valueOf(id));

            return getDataColumn(context, contentUri, null, null);
        }
        // MediaProvider
        else if (isMediaDocument(uri)) {
            final String docId = DocumentsContract.getDocumentId(uri);
            final String[] split = docId.split(":");
            final String type = split[0];

            Uri contentUri = null;
            if ("image".equals(type)) {
                contentUri = MediaStore.Images.Media.EXTERNAL_CONTENT_URI;
            } else if ("video".equals(type)) {
                contentUri = MediaStore.Video.Media.EXTERNAL_CONTENT_URI;
            } else if ("audio".equals(type)) {
                contentUri = MediaStore.Audio.Media.EXTERNAL_CONTENT_URI;
            }

            final String selection = "_id=?";
            final String[] selectionArgs = new String[] {
                    split[1]
            };

            return getDataColumn(context, contentUri, selection, selectionArgs);
        }
    }
    // MediaStore (and general)
    else if ("content".equalsIgnoreCase(uri.getScheme())) {

        // Return the remote address
        if (isGooglePhotosUri(uri))
            return uri.getLastPathSegment();

        return getDataColumn(context, uri, null, null);
    }
    // File
    else if ("file".equalsIgnoreCase(uri.getScheme())) {
        return uri.getPath();
    }

    return null;
}

/**
 * Get the value of the data column for this Uri. This is useful for
 * MediaStore Uris, and other file-based ContentProviders.
 *
 * @param context The context.
 * @param uri The Uri to query.
 * @param selection (Optional) Filter used in the query.
 * @param selectionArgs (Optional) Selection arguments used in the query.
 * @return The value of the _data column, which is typically a file path.
 */
public static String getDataColumn(Context context, Uri uri, String selection,
        String[] selectionArgs) {

    Cursor cursor = null;
    final String column = "_data";
    final String[] projection = {
            column
    };

    try {
        cursor = context.getContentResolver().query(uri, projection, selection, selectionArgs,
                null);
        if (cursor != null && cursor.moveToFirst()) {
            final int index = cursor.getColumnIndexOrThrow(column);
            return cursor.getString(index);
        }
    } finally {
        if (cursor != null)
            cursor.close();
    }
    return null;
}


/**
 * @param uri The Uri to check.
 * @return Whether the Uri authority is ExternalStorageProvider.
 */
public static boolean isExternalStorageDocument(Uri uri) {
    return "com.android.externalstorage.documents".equals(uri.getAuthority());
}

/**
 * @param uri The Uri to check.
 * @return Whether the Uri authority is DownloadsProvider.
 */
public static boolean isDownloadsDocument(Uri uri) {
    return "com.android.providers.downloads.documents".equals(uri.getAuthority());
}

/**
 * @param uri The Uri to check.
 * @return Whether the Uri authority is MediaProvider.
 */
public static boolean isMediaDocument(Uri uri) {
    return "com.android.providers.media.documents".equals(uri.getAuthority());
}

/**
 * @param uri The Uri to check.
 * @return Whether the Uri authority is Google Photos.
 */
public static boolean isGooglePhotosUri(Uri uri) {
    return "com.google.android.apps.photos.content".equals(uri.getAuthority());
}

请在这里查看此方法的最新版本。

对于这种类型的uri 内容:/ / % 3 a19298 com.android.providers.media.documents /文件/文档 或者uri.getAuthority()是这些中的任何一个

"com.google.android.apps.docs.storage".equals(uri.getAuthority()) || "com.google.android.apps.docs.storage.legacy".equals(uri.getAuthority());

使用这个函数

private static String getDriveFilePath(Uri uri, Context context) {
        Uri returnUri = uri;
        Cursor returnCursor = context.getContentResolver().query(returnUri, null, null, null, null);

        int nameIndex = returnCursor.getColumnIndex(OpenableColumns.DISPLAY_NAME);
        int sizeIndex = returnCursor.getColumnIndex(OpenableColumns.SIZE);
        returnCursor.moveToFirst();
        String name = (returnCursor.getString(nameIndex));
        String size = (Long.toString(returnCursor.getLong(sizeIndex)));
        File file = new File(context.getCacheDir(), name);
        try {
            InputStream inputStream = context.getContentResolver().openInputStream(uri);
            FileOutputStream outputStream = new FileOutputStream(file);
            int read = 0;
            int maxBufferSize = 1 * 1024 * 1024;
            int bytesAvailable = inputStream.available();

            //int bufferSize = 1024;
            int bufferSize = Math.min(bytesAvailable, maxBufferSize);

            final byte[] buffers = new byte[bufferSize];
            while ((read = inputStream.read(buffers)) != -1) {
                outputStream.write(buffers, 0, read);
            }
            Log.e("File Size", "Size " + file.length());
            inputStream.close();
            outputStream.close();
            Log.e("File Path", "Path " + file.getPath());
            Log.e("File Size", "Size " + file.length());
        } catch (Exception e) {
            Log.e("Exception", e.getMessage());
        }
        return file.getPath();
    }

Just wanted to say that this answer is brilliant and I'm using it for a long time without problems. But some time ago I've stumbled upon a problem that DownloadsProvider returns URIs in format content://com.android.providers.downloads.documents/document/raw%3A%2Fstorage%2Femulated%2F0%2FDownload%2Fdoc.pdf and hence app is crashed with NumberFormatException as it's impossible to parse its uri segments as long. But raw: segment contains direct uri which can be used to retrieve a referenced file. So I've fixed it by replacing isDownloadsDocument(uri) if content with following:

final String id = DocumentsContract.getDocumentId(uri);
if (!TextUtils.isEmpty(id)) {
if (id.startsWith("raw:")) {
    return id.replaceFirst("raw:", "");
}
try {
    final Uri contentUri = ContentUris.withAppendedId(
            Uri.parse("content://downloads/public_downloads"), Long.valueOf(id));
    return getDataColumn(context, contentUri, null, null);
} catch (NumberFormatException e) {
    Log.e("FileUtils", "Downloads provider returned unexpected uri " + uri.toString(), e);
    return null;
}
}

这是一个完全的hack，但这是我所做的…

因此，在设置DocumentsProvider时，我注意到样例代码(在getDocIdForFile中，大约在第450行)根据文件相对于指定根的路径(即在第96行中设置mBaseDir的路径)为所选文档生成了唯一的id。

所以URI最终看起来像这样:

内容:/ / com.example.provider /文档/根:路径/ / /文件

正如文档所说，它假设只有一个根(在我的情况下是Environment.getExternalStorageDirectory()，但你可以在其他地方使用…然后它获取文件路径，从根目录开始，并使其成为唯一的ID，前置“root:”。所以我可以通过消除uri.getPath()中的“/document/root:”部分来确定路径，通过这样做来创建一个实际的文件路径:

public void onActivityResult(int requestCode, int resultCode, Intent data) {
// check resultcodes and such, then...
uri = data.getData();
if (uri.getAuthority().equals("com.example.provider"))  {
    String path = Environment.getExternalStorageDirectory(0.toString()
                 .concat("/")
                 .concat(uri.getPath().substring("/document/root:".length())));
    doSomethingWithThePath(path); }
else {
    // another provider (maybe a cloud-based service such as GDrive)
    // created this uri.  So handle it, or don't.  You can allow specific
    // local filesystem providers, filter non-filesystem path results, etc.
}

我知道。这很可耻，但它奏效了。同样，这依赖于您在应用程序中使用自己的文档提供程序来生成文档ID。

(此外，还有一种更好的方法来构建路径，不假设“/”是路径分隔符，等等。但你懂的。)