Question

我正在尝试构建一个包含public Blob getImage() { Uri uri = Uri.fromFile(new File(mCurrentPhotoPath)); File f1 = new File(mCurrentPhotoPath); URL myURL = null; try { myURL = f1.toURI().toURL(); } catch (MalformedURLException e) { e.printStackTrace(); } Bitmap bitmap = null; BitmapFactory.Options options = new BitmapFactory.Options(); options.inPreferredConfig = Bitmap.Config.ARGB_8888; try { if (myURL != null) { bitmap = BitmapFactory.decodeStream( myURL.openConnection().getInputStream()); String wBlob = encodeToBase64(bitmap, Bitmap.CompressFormat.JPEG, 100); blob = fromBase64String(wBlob); } } catch (java.io.IOException e) { e.printStackTrace(); } return blob; }成对比较的Hive表。例如，假设我们从表ids开始：

我想创建一个表id val a 10 b 11 c 12：

我认为这需要自我加入，但我不知道该怎么做。非常感谢任何想法或引导。

我试过了：

id_A  id_B  val_A  val_B  val_diff
a     b     10     11     -1
a     c     10     12     -2
b     c     11     12     -1

但它不起作用：

CREATE TABLE C STORED AS ORC AS
SELECT a.id         AS id_A, 
       b.id         AS id_B,
       a.val         AS val_A,
       b.val         AS val_B,
       a.val - b.val AS val_diff
FROM A AS a
INNER JOIN B as b
ON a.id != b.id;

更新问题扩展。

假设我有表FAILED: SemanticException [Error 10017]: Line 4:3 Both left and right aliases encountered in JOIN 'id'：

A'

我如何仅在每个id val category a 10 X b 11 X c 12 X d 13 Y e 14 Y f 15 Y的元素之间进行成对比较？

即。表category：

B'

Answer 1

您使用自我加入：

select a.id as id_a, b.id as id_b, a.val as val_a, b.val as val_b,
       (a.val - b.val) as val_diff
from t a join
     t b
     on a.id < b.id;

在Hive中，您可能需要这样做：

select a.id as id_a, b.id as id_b, a.val as val_a, b.val as val_b,
       (a.val - b.val) as val_diff
from t a cross join
     t b
where a.id < b.id;

HiveQL：构建成对比较表

1 个答案: