add std::net::url - with fixes (#1748)

* add std::net::url for parsing/generating URLs

* Move String.index_of_chars into std

* Fix param contract

* Idiomatic type naming, Allman formatting, slicing, document functions

* Use String.tokenize

* Don't return str_view() from freed dstring

* Change indentation to tabs

* Variable casing according to guidlelines

* Updated API and added line to the releasenotes.

---------

Co-authored-by: Christoffer Lerno <christoffer@aegik.com>
This commit is contained in:
Louis Brauer
2025-01-02 21:13:42 +01:00
committed by GitHub
parent a16316d7b4
commit d027a15b4a
4 changed files with 615 additions and 0 deletions

277
lib/std/net/url.c3 Normal file
View File

@@ -0,0 +1,277 @@
module std::net::url;
import std::io, std::collections::map, std::collections::list;
def UrlQueryValueList = List(<String>);
struct UrlQueryValues
{
inline HashMap(<String, UrlQueryValueList>) map;
}
struct Url(Printable)
{
String scheme;
String host;
uint port;
String username;
String password;
String path;
String query;
String fragment;
}
<*
Parse a URL string into a Url struct.
@param [in] url_string
@require url_string.len > 0 "the url_string must be len 1 or more"
@return "the parsed Url"
*>
fn Url! parse(String url_string)
{
Url url;
url_string = url_string.trim();
if (!url_string.len)
{
return url;
}
// Parse scheme
if (try pos = url_string.index_of("://"))
{
url.scheme = url_string[:pos];
url_string = url_string[url.scheme.len + 3 ..];
}
else if (url_string.contains(":"))
{
// Handle schemes without authority like 'mailto:'
url.scheme = url_string[:url_string.index_of(":")!];
url_string = url_string[url.scheme.len + 1 ..];
url.path = url_string;
return url;
}
// Parse host, port
if (url.scheme != "urn")
{
usz! authority_end = url_string.index_of_chars("/?#");
if (catch authority_end)
{
authority_end = url_string.len;
}
String authority = url_string[:authority_end]!;
if (try usz userInfo_end = url_string.index_of_char('@'))
{
String userinfo = authority[:userInfo_end];
String[] userpass = userinfo.split(":");
defer free(userpass);
url.username = userpass[0];
if (userpass.len > 1)
{
url.password = userpass[1];
}
authority = authority[userInfo_end + 1 ..];
}
// Check for IPv6 address in square brackets
if (authority.starts_with("[") && authority.contains("]"))
{
usz ipv6_end = authority.index_of("]")!;
url.host = authority[0 .. ipv6_end]; // Includes closing bracket
if ((ipv6_end + 1) < authority.len && authority[.. ipv6_end] == ":")
{
url.port = authority[.. ipv6_end + 1].to_uint()!;
}
}
else
{
String[] host_port = authority.split(":");
defer mem::free(host_port);
if (host_port.len > 1)
{
url.host = host_port[0];
url.port = host_port[1].to_uint()!;
}
else
{
url.host = authority;
}
}
url_string = url_string[authority_end ..]!;
}
// Parse path
long query_index = (long)url_string.index_of_char('?') ?? -1;
long fragment_index = (long)url_string.index_of_char('#') ?? -1;
if (query_index != -1 || fragment_index != -1)
{
long pathEnd = min(query_index == -1 ? url_string.len : query_index,
fragment_index == -1 ? url_string.len : fragment_index,
url_string.len);
url.path = url_string[:pathEnd];
}
else
{
url.path = url_string;
}
// Remove the path part from url for further parsing
url_string = url_string[url.path.len ..];
// Parse query
if (url_string.starts_with("?"))
{
fragment_index = (long)url_string.index_of_char('#') ?? -1;
if (fragment_index == -1)
{
fragment_index = url_string.len;
}
url.query = url_string[1 .. fragment_index - 1];
url_string = url_string[fragment_index ..];
}
// Parse fragment
if (url_string.starts_with("#"))
{
url.fragment = url_string[1 ..];
}
return url;
}
<*
Stringify a Url struct.
@param [in] self
@param [inout] allocator
@return "Url as a string"
*>
fn String Url.to_string(&self, Allocator allocator = allocator::heap()) @dynamic
{
@pool(allocator)
{
DString builder = dstring::temp_new();
// Add scheme if it exists
if (self.scheme != "")
{
builder.append_chars(self.scheme);
builder.append_char(':');
if (self.host.len > 0) builder.append_chars("//");
}
// Add username and password if they exist
if (self.username != "")
{
builder.append_chars(self.username);
if (self.password != "")
{
builder.append_char(':');
builder.append_chars(self.password);
}
builder.append_char('@');
}
// Add host
builder.append_chars(self.host);
// Add port
if (self.port != 0)
{
builder.append_char(':');
builder.appendf("%d", self.port);
}
// Add path
builder.append_chars(self.path);
// Add query if it exists
if (self.query != "")
{
builder.append_char('?');
builder.append_chars(self.query);
}
// Add fragment if it exists
if (self.fragment != "")
{
builder.append_char('#');
builder.append_chars(self.fragment);
}
return builder.copy_str(allocator);
};
}
<*
Parse the query parameters of the Url into a UrlQueryValues map.
@param [in] self
@param [inout] allocator
@return "a UrlQueryValues HashMap"
*>
fn UrlQueryValues Url.query_values(&self, Allocator allocator)
{
UrlQueryValues vals;
vals.init(allocator);
Splitter raw_vals = self.query.tokenize("&");
while (try String rv = raw_vals.next())
{
@pool(allocator)
{
String[] parts = rv.tsplit("=", 2);
if (try existing = vals.get_ref(parts[0]))
{
existing.push(parts[1]);
}
else
{
UrlQueryValueList new_list;
new_list.new_init_with_array({ parts[1] }, allocator);
vals[parts[0]] = new_list;
}
};
}
return vals;
}
<*
Parse the query parameters of the Url into a UrlQueryValues map,
to be freed using values.free()
@param [in] self
@return "a UrlQueryValues map"
*>
fn UrlQueryValues Url.new_query_values(&self)
{
return self.query_values(allocator::heap()) @inline;
}
<*
Parse the query parameters of the Url into a UrlQueryValues map.
stored on the temp allocator.
@param [in] self
@return "a UrlQueryValues map"
*>
fn UrlQueryValues Url.temp_query_values(&self)
{
return self.query_values(allocator::temp()) @inline;
}
fn void UrlQueryValues.free(&self)
{
self.map.@each(;String key, UrlQueryValueList value)
{
value.free();
};
self.map.free();
}