more rubyish

[dpep]

while working on #481 I noticed a few things that could be tightened up and be made more Rushish. there should be no functionality changes. Forgive me if this is unwelcomed feedback

[dpep]

this line took me a bit to figure out, so I figured it was comment worthy. based on the specs, it looks like initialize leverages the validation of self.path done in .to_s. but moving it into .validate causes other test failures...

[dentarg]

Could give this PR a better title?

[dpep]

for sure. sorry, that got copied over from the other branch. better? (otherwise, open to suggestions)

[dentarg]

Looks like this assignment was moved inside if authority != nil where it wasn't before.
Not sure it matters (all tests still pass, but we might lack coverage).

Spotted this when looking at the files instead of the diff.

This PR

addressable/lib/addressable/uri.rb

Lines 134 to 148 in 7379545

	if authority != nil
	# The Regexp above doesn't split apart the authority.
	userinfo = authority[/^([^\[\]]*)@/, 1]
	if userinfo != nil
	user = userinfo.strip[/^([^:]*):?/, 1]
	password = userinfo.strip[/:(.*)$/, 1]
	end
	host = authority.sub(
	/^([^\[\]]*)@/, EMPTY_STR
	).sub(
	/:([^:@\[\]]*?)$/, EMPTY_STR
	)
	port = authority[/:([^:@\[\]]*?)$/, 1]
	port = nil if port == EMPTY_STR
	end

main

addressable/lib/addressable/uri.rb

Lines 134 to 150 in 8657465

	if authority != nil
	# The Regexp above doesn't split apart the authority.
	userinfo = authority[/^([^\[\]]*)@/, 1]
	if userinfo != nil
	user = userinfo.strip[/^([^:]*):?/, 1]
	password = userinfo.strip[/:(.*)$/, 1]
	end
	host = authority.sub(
	/^([^\[\]]*)@/, EMPTY_STR
	).sub(
	/:([^:@\[\]]*?)$/, EMPTY_STR
	)
	port = authority[/:([^:@\[\]]*?)$/, 1]
	end
	if port == EMPTY_STR
	port = nil
	end

[dpep]

correct. port is set to nil on line 133 before the block, so this check only needs to happen if the authority block changes the port value

[dpep]

added a test to ensure proper functionality

[dentarg]

@dpep sorry letting this drag on... #486 changed things related to this PR. Does the changes here still make sense after #486? Feel free to rebase this PR and I'll have a look again

[dpep]

rebased. I think these changes still improve code quality / readability, but feel free to close