[rust] Save to directory

[oriongonza]

Allows the user to download the webdriver to a specific directory.

Description

Adds the --save-path flag, which allows the user to download the selected driver to a folder.

Also a few refactors that I didn't see in the other PR

Motivation and Context

I had the need for this so I implemented it here.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Checklist

• I have read the contributing document.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.
• I have added tests to cover my changes.
• All new and existing tests passed.

[CLAassistant]

All committers have signed the CLA.

[oriongonza]

@bonigarcia you might want to take a look at this two PRs #11832

[bonigarcia]

This PR should not be merged. The flag --save-path is not necessary, since the feature to change the binary path will be implemented through the flag --cache-path (planned in #11688, which has a current PR under review: #11840).

[oriongonza]

This might still be useful (it is at least for me) since the webdriver will be saved under a specific version folder eg ~/mycache/111.xxx.xxx/chromedriver.

[oriongonza]

Still, even if this gets pushed back there are a few refactors worth considering, especially using #[Derive(Clone)] attribute macro.
By the way, why does the get_config clone the configuration struct instead of returning a mutable reference to it? That way you can simplify (and speed up but that doesn't really matter) the addding of a new configuration.

Also it might be worth considering using optionals instead of empty strings for the configurations that don't exist since rust provides really powerful tools for working with options,

[bonigarcia]

This might still be useful (it is at least for me) since the webdriver will be saved under a specific version folder eg ~/mycache/111.xxx.xxx/chromedriver.

I see your point now. Still, I think having both flags --cache-path and --save-path can be confusing. Instead, I propose having only --cache-path (which it has already another PR), and created another extra flag e.g. called --avoid-driver-tree to store the driver without using a tree folder for the drivers. @dev-ardi What do you think?

[bonigarcia]

Still, even if this gets pushed back there are a few refactors worth considering, especially using #[Derive(Clone)] attribute macro. By the way, why does the get_config clone the configuration struct instead of returning a mutable reference to it? That way you can simplify (and speed up but that doesn't really matter) the addding of a new configuration.

Yes, I have just seen your PR about it and it seems a handy way to avoid the cloning method, thanks a lot. I think we should merge it, thanks.

Also it might be worth considering using optionals instead of empty strings for the configurations that don't exist since rust provides really powerful tools for working with options,

Yes, that is a good idea too.

[diemol]

I got lost a bit. What should we do with this PR? I believe it needs to be split to handle the refactor apart?

[oriongonza]

I see your point now. Still, I think having both flags --cache-path and --save-path can be confusing. Instead, I propose having only --cache-path (which it has already another PR), and created another extra flag e.g. called --avoid-driver-tree to store the driver without using a tree folder for the drivers. @dev-ardi What do you think?

That looks like the correct approach

[oriongonza]

I got lost a bit. What should we do with this PR? I believe it needs to be split to handle the refactor apart?

I guess so. I don't know how to select only some of the changes from each commit, I'm very good with git. I could use some help with this :)

[diemol]

You can create a complete new branch from trunk and only apply the refactoring changes to send a PR.