[
https://issues.apache.org/jira/browse/THRIFT-4529?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16682477#comment-16682477
]
Allen George commented on THRIFT-4529:
--------------------------------------
Humorously (I guess this is the story of all programming) this turns out to be
more complicated than expected.
There are a few cases:
1. All lowercase. "foo" => "Foo"
2. All uppercase. "FOO" => "Foo"
3. Screaming snake case (most common, I'd assume). "SCREAMING_SNAKE_CASE" =>
"ScreamingSnakeCase"
4. Snake case. "snake_case" => "SnakeCase"
4. Camel case. "CamelCase" => "CamelCase"
The problem is, we have to figure out which one of the situations we're dealing
with, because the conversions we'll perform will differ for every one, which is
supremely annoying. I tried to avoid having to differentiate between the cases,
but, the default implementations of {{t_generator::camelcase}} and
{{t_generator::uppercase}} make this impossible.
> Rust generation should include #![allow(non_snake_case)] or force conform to
> Rust style guidelines
> --------------------------------------------------------------------------------------------------
>
> Key: THRIFT-4529
> URL: https://issues.apache.org/jira/browse/THRIFT-4529
> Project: Thrift
> Issue Type: Improvement
> Components: Rust - Compiler
> Affects Versions: 0.11.0
> Reporter: Joshua
> Assignee: Allen George
> Priority: Minor
>
> Without this, building a project using a thrift file meant for multiple
> languages may end up with many compiler warnings similar to the following:
> {code:sh}
> warning: variant `EXAMPLE_NAME` should have a camel case name such as
> `ExampleName`
> {code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
