[FORMATTING] Querying a semi-structured data with an element name that is a reserved clause is incorrect

[chaupow]

Hi! 👋

Input data

Which SQL and options did you provide as input? with "language": "snowflake"

SELECT
  details:type
FROM
  transactions

Expected Output

SELECT
  details:type
FROM
  transactions

Actual Output

SELECT
  details:
type
FROM
  transactions

Usage

We're using snowflake and unfortunately some of our JSON fields have the key type which is identified as a reserved clause and incorrectly put in the next line.

We can prevent this by using details:"type" but it would be great if it would work without quotes as well.

This is especially an issue because we would love to use "keywordCase": "upper" but this breaks the query because key names are case sensitive and details:TYPE fails to query the intended lookup.

• How are you calling / using the library?

  sqlFormatter.format(sqlContent, {"language": "snowflake"});

• What SQL language(s) does this apply to? snowflake

• Which SQL Formatter version are you using? 15.1.2

Thank youuuuuuuuu 🙏

[nene]

Thanks for reporting.

Although I apparently have documented this :-syntax of Redshift, I haven't actually implemented any support for it.

Interestingly I just recently fixed a similar issue for Transact-SQL. So most likely some similar solution could be done in here.

[nene]

I was mistaken. This syntax was supported, but as the issue title says, the problem was only the case when a keyword occurred after :.

Now treating : similarly to . that's commonly used as a table.column separator.

[nene]

Fixed in 15.1.3

[chaupow]

Oh wow this was amazing quick!! ⚡✨

Thanks so much for fixing so promptly and pushing a new version out 🥹 I really appreciate it!