std::net::url cleanups for lint check

This commit is contained in:
Daniel Patterson 2012-07-31 23:14:50 -04:00 committed by Brian Anderson
parent b57f6b73ba
commit 21187206c3

View file

@ -30,8 +30,9 @@ type userinfo = {
type query = ~[(~str, ~str)];
fn url(-scheme: ~str, -user: option<userinfo>, -host: ~str, -port: option<~str>,
-path: ~str, -query: query, -fragment: option<~str>) -> url {
fn url(-scheme: ~str, -user: option<userinfo>, -host: ~str,
-port: option<~str>, -path: ~str, -query: query,
-fragment: option<~str>) -> url {
{ scheme: scheme, user: user, host: host, port: port,
path: path, query: query, fragment: fragment }
}
@ -79,9 +80,9 @@ fn encode_inner(s: ~str, full_url: bool) -> ~str {
}
}
/**
* Encodes a URI by replacing reserved characters with percent encoded character
* sequences.
/**
* Encodes a URI by replacing reserved characters with percent encoded
* character sequences.
*
* This function is compliant with RFC 3986.
*/
@ -89,9 +90,9 @@ fn encode(s: ~str) -> ~str {
encode_inner(s, true)
}
/**
* Encodes a URI component by replacing reserved characters with percent encoded
* character sequences.
/**
* Encodes a URI component by replacing reserved characters with percent
* encoded character sequences.
*
* This function is compliant with RFC 3986.
*/
@ -139,14 +140,14 @@ fn decode_inner(s: ~str, full_url: bool) -> ~str {
/**
* Decode a string encoded with percent encoding.
*
*
* This will only decode escape sequences generated by encode_uri.
*/
fn decode(s: ~str) -> ~str {
decode_inner(s, true)
}
/**
/**
* Decode a string encoded with percent encoding.
*/
fn decode_component(s: ~str) -> ~str {
@ -172,7 +173,7 @@ fn encode_plus(s: ~str) -> ~str {
}
}
/**
/**
* Encode a hashmap to the 'application/x-www-form-urlencoded' media type.
*/
fn encode_form_urlencoded(m: hashmap<~str, @dvec<@~str>>) -> ~str {
@ -190,18 +191,19 @@ fn encode_form_urlencoded(m: hashmap<~str, @dvec<@~str>>) -> ~str {
first = false;
}
out += #fmt("%s=%s", key, encode_plus(*value));
out += #fmt("%s=%s", key, encode_plus(*value));
}
}
out
}
/**
/**
* Decode a string encoded with the 'application/x-www-form-urlencoded' media
* type into a hashmap.
*/
fn decode_form_urlencoded(s: ~[u8]) -> hashmap<~str, @dvec<@~str>> {
fn decode_form_urlencoded(s: ~[u8]) ->
map::hashmap<~str, @dvec::dvec<@~str>> {
do io::with_bytes_reader(s) |rdr| {
let m = str_hash();
let mut key = ~"";
@ -282,7 +284,8 @@ fn split_char_first(s: ~str, c: char) -> (~str, ~str) {
if index+match_ == len {
return (str::slice(s, 0, index), ~"");
} else {
return (str::slice(s, 0, index), str::slice(s, index + match_, str::len(s)));
return (str::slice(s, 0, index),
str::slice(s, index + match_, str::len(s)));
}
}
@ -355,7 +358,7 @@ fn get_scheme(rawurl: ~str) -> result::result<(~str, ~str), @~str> {
// returns userinfo, host, port, and unparsed part, or an error
// currently doesn't handle IPv6 addresses.
fn get_authority(rawurl: ~str) ->
fn get_authority(rawurl: ~str) ->
result::result<(option<userinfo>, ~str, option<~str>, ~str), @~str> {
if !str::starts_with(rawurl, ~"//") {
// there is no authority.
@ -373,7 +376,7 @@ fn get_authority(rawurl: ~str) ->
enum input {
digit, // all digits
hex, // digits and letters a-f
unreserved // all other legal characters in usernames, passwords, hosts
unreserved // all other legal characters
}
let len = str::len(rawurl);
let mut st : state = start;
@ -399,8 +402,8 @@ fn get_authority(rawurl: ~str) ->
alt c {
'0' to '9' { }
'A' to 'F' | 'a' to 'f' {
if in == digit {
in = hex;
if in == digit {
in = hex;
}
}
'G' to 'Z' | 'g' to 'z' | '-' | '.' | '_' | '~' | '%' |
@ -415,7 +418,7 @@ fn get_authority(rawurl: ~str) ->
}
}
// now state machine
// now process states
alt c {
':' {
colon_count += 1;
@ -465,14 +468,14 @@ fn get_authority(rawurl: ~str) ->
alt st {
start {
let user = str::slice(rawurl, begin, i);
userinfo = option::some({user : user,
userinfo = option::some({user : user,
pass: option::none});
st = in_host;
}
pass_host_port {
let user = str::slice(rawurl, begin, pos);
let pass = str::slice(rawurl, pos+1, i);
userinfo = option::some({user: user,
userinfo = option::some({user: user,
pass: option::some(pass)});
st = in_host;
}
@ -482,7 +485,7 @@ fn get_authority(rawurl: ~str) ->
}
begin = i+1;
}
'?' | '#' | '/' {
break;
}
@ -517,14 +520,14 @@ fn get_authority(rawurl: ~str) ->
}
}
let rest = if i+1 == len { ~"" }
let rest = if i+1 == len { ~"" }
else { str::slice(rawurl, i, len) };
return result::ok((userinfo, host, port, rest));
}
// returns the path and unparsed part of url, or an error
fn get_path(rawurl: ~str, authority : bool) ->
fn get_path(rawurl: ~str, authority : bool) ->
result::result<(~str, ~str), @~str> {
let len = str::len(rawurl);
let mut end = len;
@ -548,27 +551,27 @@ fn get_path(rawurl: ~str, authority : bool) ->
'/' in presence of authority.");
}
}
return result::ok((decode_component(str::slice(rawurl, 0, end)),
return result::ok((decode_component(str::slice(rawurl, 0, end)),
str::slice(rawurl, end, len)));
}
// returns the parsed query and the fragment, if present
fn get_query_fragment(rawurl: ~str) ->
fn get_query_fragment(rawurl: ~str) ->
result::result<(query, option<~str>), @~str> {
if !str::starts_with(rawurl, ~"?") {
if str::starts_with(rawurl, ~"#") {
let f = decode_component(str::slice(rawurl,
1,
let f = decode_component(str::slice(rawurl,
1,
str::len(rawurl)));
return result::ok((~[], option::some(f)));
} else {
return result::ok((~[], option::none));
}
}
let (q, r) = split_char_first(str::slice(rawurl, 1,
let (q, r) = split_char_first(str::slice(rawurl, 1,
str::len(rawurl)), '#');
let f = if str::len(r) != 0 {
let f = if str::len(r) != 0 {
option::some(decode_component(r)) } else { option::none };
return result::ok((query_from_str(q), f));
}
@ -606,7 +609,7 @@ fn from_str(rawurl: ~str) -> result::result<url, ~str> {
let mut pth = get_path(rest, has_authority);
if result::is_err(pth) {
return result::err(copy *result::get_err(pth));
}
}
let (path, rest) = result::unwrap(pth);
// query and fragment
@ -616,7 +619,7 @@ fn from_str(rawurl: ~str) -> result::result<url, ~str> {
}
let (query, fragment) = result::unwrap(qry);
return result::ok(url(scheme, userinfo, host,
return result::ok(url(scheme, userinfo, host,
port, path, query, fragment));
}
@ -647,7 +650,8 @@ fn to_str(url: url) -> ~str {
str::concat(~[~"?", query_to_str(url.query)])
};
let fragment = if option::is_some(url.fragment) {
str::concat(~[~"#", encode_component(option::unwrap(copy url.fragment))])
str::concat(~[~"#", encode_component(
option::unwrap(copy url.fragment))])
} else {
~""
};
@ -674,7 +678,7 @@ mod tests {
let (u,v) = split_char_first(~"hello, sweet world", ',');
assert u == ~"hello";
assert v == ~" sweet world";
let (u,v) = split_char_first(~"hello sweet world", ',');
assert u == ~"hello sweet world";
assert v == ~"";
@ -684,7 +688,7 @@ mod tests {
fn test_get_authority() {
let (u, h, p, r) = result::unwrap(get_authority(
~"//user:pass@rust-lang.org/something"));
assert u == option::some({user: ~"user",
assert u == option::some({user: ~"user",
pass: option::some(~"pass")});
assert h == ~"rust-lang.org";
assert option::is_none(p);
@ -696,7 +700,7 @@ mod tests {
assert h == ~"rust-lang.org";
assert p == option::some(~"8000");
assert r == ~"?something";
let (u, h, p, r) = result::unwrap(get_authority(
~"//rust-lang.org#blah"));
assert option::is_none(u);
@ -718,50 +722,57 @@ mod tests {
~"//us:p@2001:0db8:85a3:0042:0000:8a2e:0370:7334:8000#blah"));
assert u == option::some({user: ~"us", pass : option::some(~"p")});
assert h == ~"2001:0db8:85a3:0042:0000:8a2e:0370:7334";
assert p == option::some(~"8000");
assert p == option::some(~"8000");
// invalid authorities;
assert result::is_err(get_authority(~"//user:pass@rust-lang:something"));
assert result::is_err(get_authority(~"//user@rust-lang:something:/path"));
// invalid authorities;
assert result::is_err(get_authority(
~"//user:pass@rust-lang:something"));
assert result::is_err(get_authority(
~"//user@rust-lang:something:/path"));
assert result::is_err(get_authority(
~"//2001:0db8:85a3:0042:0000:8a2e:0370:7334:800a"));
assert result::is_err(get_authority(
~"//2001:0db8:85a3:0042:0000:8a2e:0370:7334:8000:00"));
// these parse as empty, because they don't start with '//'
let (_, h, _, _) = result::unwrap(get_authority(~"user:pass@rust-lang"));
let (_, h, _, _) = result::unwrap(
get_authority(~"user:pass@rust-lang"));
assert h == ~"";
let (_, h, _, _) = result::unwrap(get_authority(~"rust-lang.org"));
let (_, h, _, _) = result::unwrap(
get_authority(~"rust-lang.org"));
assert h == ~"";
}
#[test]
fn test_get_path() {
let (p, r) = result::unwrap(get_path(~"/something+%20orother", true));
let (p, r) = result::unwrap(get_path(
~"/something+%20orother", true));
assert p == ~"/something+ orother";
assert r == ~"";
let (p, r) = result::unwrap(get_path(~"test@email.com#fragment", false));
let (p, r) = result::unwrap(get_path(
~"test@email.com#fragment", false));
assert p == ~"test@email.com";
assert r == ~"#fragment";
let (p, r) = result::unwrap(get_path(~"/gen/:addr=?q=v", false));
assert p == ~"/gen/:addr=";
assert r == ~"?q=v";
//failure cases
assert result::is_err(get_path(~"something?q", true));
}
#[test]
fn test_url_parse() {
let url = ~"http://user:pass@rust-lang.org/doc?s=v#something";
let up = from_str(url);
let u = result::unwrap(up);
assert u.scheme == ~"http";
assert option::unwrap(copy u.user).user == ~"user";
assert option::unwrap(copy option::unwrap(copy u.user).pass) == ~"pass";
assert option::unwrap(copy option::unwrap(copy u.user).pass)
== ~"pass";
assert u.host == ~"rust-lang.org";
assert u.path == ~"/doc";
assert u.query.find(|kv| kv.first() == ~"s").get().second() == ~"v";