adjusting semver

[sbussetti]

No description provided.

[ghost]

I also get blocked from using this loader because of the peer dependency requirement.

@sbussetti I think >=0.25.0 <0.30.0 is not appropriate. There's no guarantee that 0.29 - 0.30 will be compatible with this plugin (given that 0.28 is current latest version). Unless you know the development plans of css-loader.

@smithad15 I think you can just use >=0.25.0 if you would not like to keep updating whenever a new version of css-loader is released, although there's no guarantee that the loader will always work with the latest version of css-loader. But since this loader is quite simple, it's much less likely that it no longer works than a new version of css-loader is released.

[sbussetti]

I'm going by the general guideline that developers should not introduce breaking changes in minor version upgrades. So if smithad also follows that generally agreed upon pragma, then yes in a way i do know his development plans, which is why I limited the top end to be below .30

…

-- Steve Bussetti From: Martin Chen <[email protected]> <[email protected]> Reply: smithad15/css-to-string-loader <[email protected]> <[email protected]> Date: April 19, 2017 at 3:26:04 AM To: smithad15/css-to-string-loader <[email protected]> <[email protected]> CC: Steve Bussetti <[email protected]> <[email protected]>, Mention <[email protected]> <[email protected]> Subject: Re: [smithad15/css-to-string-loader] adjusting semver (#6) I also get blocked from using this loader because of the peer dependency

requirement. @sbussetti <https://github.com/sbussetti> I think >=0.25.0 <0.30.0 is not appropriate. There's no guarantee that 0.29 - 0.30 will be compatible with this plugin (given that 0.28 is current latest version). Unless you know the development plans of css-loader. @smithad15 <https://github.com/smithad15> I think you can just use >=0.25.0 if you would not like to keep updating whenever a new version of css-loader is released, although there's no guarantee that the loader will always work with the latest version of css-loader. But since this loader is quite simple, it's much less likely that it no longer works than a new version of css-loader is released. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#6 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA3v0XajQMGNm1IYVzssPaQ8n9YBDfBaks5rxbcMgaJpZM4MnO5n> .

[ghost]

That's only true after major version 1.

Major version zero (0.y.z) is for initial development. Anything may change at any time. The public API should not be considered stable.

quoted from http://semver.org/

[ghost]

And that's why Caret Ranges are defined as:

Allows changes that do not modify the left-most non-zero digit in the [major, minor, patch] tuple.

[smithad15]

Thanks for this. As I've had to deal with some peerDependency version issues before, and since semver works a bit differently when dealing with 0.x.x packages, @sbussetti could you please update the PR to an even more permissive range of >=0.25.0 <1.0.0 and remove the package version bump as I will do that during my release cycle? I have a fairly high confidence that the toString method of css-loader won't be changing significantly any time soon and if it does, I will deal with any upstream API changes at that time.

For more semver references, check out https://docs.npmjs.com/misc/semver#caret-ranges-123-025-004 for some more details of how the caret operator changes once you are <1.0.0, if either of you are interested.

[ghost]

@sbussetti Are you able to change according to smithad15's suggestion recently?

[sbussetti]

@mchencb @smithad15 done and done. Also @mchencb thanks for the semver clarification. I was not aware of the exception for 0.x.x !

[smithad15]

@sbussetti Thank you for this. Apologies for the delay but I'll push out 0.1.3 shortly.

[sbussetti]

No sweat 👌