bonitoo-io/flux

Flux is a lightweight scripting language for querying databases (like InfluxDB) and working with data. It's part of InfluxDB 1.7 and 2.0, but can be run independently of those.

flux's Issues

live DBs tests for sql package

AWS machine ("Flux DBs") with

• MySQL
• PostgreSQL
  https://phoenixnap.com/kb/how-to-install-postgresql-on-ubuntu
• SQLServer (standalone)
  https://docs.microsoft.com/en-us/sql/linux/quickstart-install-connect-ubuntu?view=sql-server-ver15
• SAP Hana
  https://blogs.sap.com/2018/07/21/how-to-install-sap-hana-express-edition-2.0-on-ubuntu-18.04-bionic-beaver/
  ./HXEDownloadManager_linux.bin linuxx86_64 installer hxe.tgz

Tests against

• MySQL
• PostgreSQL
• SQLServer (standalone)
• SQLServer (Azure)
• AWS Athena
• BigQuery
• SAP Hana

improve sql package test coverage

Improve code coverage by using datasets with type info in tests. PR #231 merged to master. Now waiting for release.

2020-08-17: DATA-DOG/go-sqlmock 1.5.0 with column metadata released

I guess it is better we wait for at least influxdata#3116 PR merged (influxdata#3098 would be nice too), and then update parsing tests for all DBs in single PR.

fix and improve `sql` package

Branch: https://github.com/bonitoo-io/flux/tree/fix/sql-misc

Bugs:

• Snowflake: translationFunc - missing outgoing TUint mapping
• Snowflake: InitColumnTypes - only when scale == 0 then integer can be assumed
• MySQL: translationFunc: TUint should map to BIGINT UNSIGNED
• all: translationFunc: unsupported type error whould be reported with codes.Invalid instead of codes.Internal

Refactor:

• mysql, postgres, sqlite - map should be outside of translationFunc and not use String() just like in recently added db supports
• to.go: CreateInsertComponents() - getTranslationFunc() called in loop, label map not used anywhere

Enhancements:

• Azure: use new public API to get tokens, at least for MSI token
  • go.mod NOTE There already is a pending PR #3172 for Azure dep update
    • autorest: v0.11.4
    • adal: v0.9.2