avro-schema
avro-schema copied to clipboard
tarantool
Apache Avro schema tools for Tarantool
I have a problem that happens only under load. Problem occurs only on MacOS (seems GC64 is enabled). Avro-schema version: 3.0.3 Tarantool version: 2.2.1 Also we use only "validate" method...
Since 2.3 Tarantool supports "decimal" data type. I think that avro-schema should also supports it. (https://avro.apache.org/docs/1.8.1/spec.html#Decimal). Also it could be reasonable to support "uuid" type (at least as extension)
Please add documentation on `options` argument for `avro.create` function in README. Maybe reflect changes in the wiki
It may give some performance on success paths. We replaced `debug.getlocal()` approach with fstack in order to fix #11. Later, @mejedi [investigated](https://gist.github.com/mejedi/d61752c5fd582d2507360d375513c6b8) the problem further and proposed a work around....
``` do avro = require('avro_schema') schematext = [[ { "name":"test", "type":"record", "fields": [ { "name": "usageType", "type": { "name": "usageTypeEnum", "type": "enum*", "symbols": [ "monetary", "voice", "data", "sms", "other" ]...
### Problem statement Schema definition requires type declaration, which is for now is string, e.g.: ``` {name = "field", type = "int"}``` which results in no help from ide preventing...
Prohibit any non-related to the Avro and non-mentioned in `preserve_in_ast` field.
This type can store any scalar type (null, int , double, string...)
1. Should we check utf-8 validity? - I think yes, because is seems that there are no way to ban certain symbols in encoding-unaware way. - But once we checked...
since we have nullable records, unions, maps, and arrays, one may want to update just some subelement of those complex types, however, it is impossible without json path. The feature...