Can you add support for having Chinese characters for predicate labels?

[lichteveld]

For our team in China it's a challenge that Chinese characters are not supported as predicate labels. This means they have to translate everything into English which is not ideal if you're importing data from a database with Chinese columns

[srfrog]

Thanks for submitting this issue. Let me get some extra information to make sure I understand correctly.

Dgraph already supports international predicate names via IRI. Simply enclose the predicate inside <>. Is this what you are trying to do?

For example:

// alter
<名字>: string .

// mutate
{
  set {
    _:test1 <名字> "srfrog" .
    <_:測試> <名字> "testing" .
  }
}

// query
{
  q(func: uid(0x1)) {
    uid
    <名字>
  }
}

Response:

{
  "data": {
    "q": [
      {
        "uid": "0x1",
        "名字": "srfrog"
      }
    ]
  },
  "extensions": {
    "server_latency": {
      "parsing_ns": 22581,
      "processing_ns": 1766643,
      "encoding_ns": 438319
    },
    "txn": {
      "start_ts": 6
    }
  }
}

[manishrjain]

Wao! Even I didn't realize this worked. Can you please add this in our documentation?

[manishrjain]

I think we can close this issue after this is documented.

[melixir]

That's great news. I'll pass the message to my coworker who was running into this problem. Thanks guys!