Implement float parsing fixes #6

[terrarier2111]

Note that things like 1.2 when made to_string spit out incorrect results (like 1.199999999) but this seems to be an issue in the to_string impl as the same happens with from_f64(1.2)

[terrarier2111]

The test failure seems to be spurious and unrelated to this feature

[nadavrot]

@terrarier2111 Thanks for working on it. I love it. Let's move forward in a few steps.
Can you first submit the cleanup commits as a separate pull request? The stuff for adding const and the other cleanups should be easy to review and land.

I read the code and I have a few comments. First, we'll need a little bit more comments and a few tests. And second, I'd like to talk about the API for constructing a float number. I noticed that you added a default semantics, but I think that we should have something like Float::from_string. What do you think?

[terrarier2111]

@terrarier2111 Thanks for working on it. I love it. Let's move forward in a few steps. Can you first submit the cleanup commits as a separate pull request? The stuff for adding const and the other cleanups should be easy to review and land.

I read the code and I have a few comments. First, we'll need a little bit more comments and a few tests. And second, I'd like to talk about the API for constructing a float number. I noticed that you added a default semantics, but I think that we should have something like Float::from_string. What do you think?

Hey, thank you for your reponse, i cleaned up the impl a bit, fixed some issues and added tests, tomorrow i will move all the code in a function that takes a semantic as an argument and let the try_from impl call this function with some default, at least for now until i implement autodetecting semantics. Are you okay with this moving forward or do you have some changes that you want? Regarding the additional changes that are unrelated to this pr i just found them randomly, i wouldn't mind removing them but i can't promise that i'll do them again as i am not really good with git and it is probably a bit cumbersome. Also i named the method try_from_str idk if this is better or just from_str/from_string

[nadavrot]

Can you please land the cleanup commits as a separate PR, one by one? It would be easier to review and land.

Afterwards I'll be able to review this diff. I think that you are really close, and I only have small suggestions (such as moving the tests into a new function, no big deal). You are almost there, just split the PR into smaller diffs.

[terrarier2111]

Can you please land the cleanup commits as a separate PR, one by one? It would be easier to review and land.

Afterwards I'll be able to review this diff. I think that you are really close, and I only have small suggestions (such as moving the tests into a new function, no big deal). You are almost there, just split the PR into smaller diffs.

I removed the cleanup changed and moved the tests

[terrarier2111]

@nadavrot are you still reviewing this or can i close this and just use my fork?

[nadavrot]

I would love for you to land the chances, but not in this pull request. Please close this PR and split the changes here into separate pull requests with cleanups and afterwards with the changes for landing the float parse. Each PR should have one commit. Thank you. I really appreciate your contribution and would love to see it land, one PR at a time.

[terrarier2111]

I would love for you to land the chances, but not in this pull request. Please close this PR and split the changes here into separate pull requests with cleanups and afterwards with the changes for landing the float parse. Each PR should have one commit. Thank you. I really appreciate your contribution and would love to see it land, one PR at a time.

I think i did it, it's now only a single commit!

[nadavrot]

Please submit a new and clean pull request (without the previous discussion).