[bug] Encountered Chinese garbled code issue during Iceberg rest HTTP transmission

[hxgdada]

Apache Iceberg version

ALL

Query engine

None

Please describe the bug 🐞

org.apache.iceberg.rest.HTTPClient

origin：
private StringEntity toJson(Object requestBody) {
try {
return new StringEntity(mapper.writeValueAsString(requestBody));
} catch (JsonProcessingException e) {
throw new RESTException(e, "Failed to write request body: %s", requestBody);
}
}

After modification：
private StringEntity toJson(Object requestBody) {
try {
return new StringEntity(mapper.writeValueAsString(requestBody), Charset.forName("UTF-8"));
} catch (JsonProcessingException e) {
throw new RESTException(e, "Failed to write request body: %s", requestBody);
}
}

[nastra]

This probably makes sense. @hxgdada do you have a small reproducible example?

[hxgdada]

This probably makes sense. @hxgdada do you have a small reproducible example?
This is a case study where I used iceberg rest to create a table in trino and found that the Chinese fields became garbled. Through debugging, I found that the iceberg rest service converted the Chinese to a special encoding during HTTP protocol transmission. The screenshot is shown below. Subsequently, I added Charset. forName ("UTF-8") to the HTTP method of iceberg rest to solve this problem

Before modification

After modification

@nastra

[nastra]

@hxgdada thanks for reporting this. I've opened #8046 that should fix it.