fix: deep virtual package #829
Conversation
Potential file name extraction issue in virtual MBTI file path constructionCategory if mod_desc.name.is_empty() {
anyhow::bail!("Module name cannot be empty");
}
rel.file_name()
.map(|x| x.to_string_lossy())
.unwrap_or_else(|| {
mod_desc.name
.rsplit('/')
.next()
.ok_or_else(|| anyhow::anyhow!("Invalid module name format"))?
.into()
})Reasoning Missing inline documentation for virtual package handling logicCategory // Constructs the path to the virtual MBTI file by either:
// 1. Using the relative path's filename if available
// 2. Falling back to the last component of the module name
virtual_mbti_file: if pkg.virtual_pkg.is_some() {Reasoning Consistent package naming convention in test filesCategory package "test/virtual/deep/lib"
...
package "test/virtual"
**Reasoning**
Using 'username' as a package prefix in test files might be confusing. A more explicit naming convention would make it clearer that these are test packages and how they relate to each other in the hierarchy.
</details> |
| .rsplit('/') | ||
| .next() | ||
| .expect("Empty module name") |
There was a problem hiding this comment.
TODO: This is obviously not the most elegant way to do the task, but is there any existing aux functions that does the task?
Related Issues
When handling the deep nested virtual package, the relative path is misused.
Type of Pull Request
Does this PR change existing behavior?
Does this PR introduce new dependencies?
Checklist: