Memory-efficient URL representation using a single buffer. More...

#include <url_aggregator.h>

Inheritance diagram for ada::url_aggregator:

Public Member Functions
	url_aggregator ()=default

	url_aggregator (const url_aggregator &u)=default

	url_aggregator (url_aggregator &&u) noexcept=default

url_aggregator &	operator= (url_aggregator &&u) noexcept=default

url_aggregator &	operator= (const url_aggregator &u)=default

	~url_aggregator () override=default

bool	set_href (std::string_view input)

bool	set_host (std::string_view input)

bool	set_hostname (std::string_view input)

bool	set_protocol (std::string_view input)

bool	set_username (std::string_view input)

bool	set_password (std::string_view input)

bool	set_port (std::string_view input)

bool	set_pathname (std::string_view input)

void	set_search (std::string_view input)

void	set_hash (std::string_view input)

bool	has_valid_domain () const noexcept override

std::string	get_origin () const noexcept override

constexpr std::string_view	get_href () const noexcept ada_lifetime_bound

std::string_view	get_username () const noexcept ada_lifetime_bound

std::string_view	get_password () const noexcept ada_lifetime_bound

std::string_view	get_port () const noexcept ada_lifetime_bound

std::string_view	get_hash () const noexcept ada_lifetime_bound

std::string_view	get_host () const noexcept ada_lifetime_bound

std::string_view	get_hostname () const noexcept ada_lifetime_bound

constexpr std::string_view	get_pathname () const noexcept ada_lifetime_bound

ada_really_inline uint32_t	get_pathname_length () const noexcept

std::string_view	get_search () const noexcept ada_lifetime_bound

std::string_view	get_protocol () const noexcept ada_lifetime_bound

ada_really_inline constexpr bool	has_credentials () const noexcept

ada_really_inline const url_components &	get_components () const noexcept

std::string	to_string () const override

std::string	to_diagram () const

constexpr bool	validate () const noexcept

constexpr bool	has_empty_hostname () const noexcept

constexpr bool	has_hostname () const noexcept

constexpr bool	has_non_empty_username () const noexcept

constexpr bool	has_non_empty_password () const noexcept

constexpr bool	has_port () const noexcept

constexpr bool	has_password () const noexcept

constexpr bool	has_hash () const noexcept override

constexpr bool	has_search () const noexcept override

void	clear_port ()

void	clear_hash ()

void	clear_search () override

Public Member Functions inherited from ada::url_base
virtual	~url_base ()=default

ada_really_inline constexpr bool	is_special () const noexcept

Friends
void	helpers::strip_trailing_spaces_from_opaque_path (url_aggregator &url) noexcept

url_aggregator	parser::parse_url (std::string_view, const url_aggregator *)

url_aggregator	parser::parse_url_impl (std::string_view, const url_aggregator *)

url_aggregator	parser::parse_url_impl (std::string_view, const url_aggregator *)

Additional Inherited Members
Public Attributes inherited from ada::url_base
bool	is_valid {true}

bool	has_opaque_path {false}

url_host_type	host_type = url_host_type::DEFAULT

Detailed Description

Memory-efficient URL representation using a single buffer.

The url_aggregator stores the entire normalized URL in a single string buffer and tracks component boundaries using offsets. This design minimizes memory allocations and is ideal for read-mostly access patterns.

Getter methods return std::string_view pointing into the internal buffer. These views are lightweight (no allocation) but become invalid if the url_aggregator is modified or destroyed.

Warning: Views returned by getters (e.g., get_pathname()) are invalidated when any setter is called. Do not use a getter's result as input to a setter on the same object without copying first.

Note: This is the default URL type returned by ada::parse().

See also: url For an alternative using separate std::string instances

Definition at line 46 of file url_aggregator.h.

Constructor & Destructor Documentation

◆ url_aggregator() [1/3]

ada::url_aggregator::url_aggregator ( )

default

Referenced by helpers::strip_trailing_spaces_from_opaque_path, operator=(), operator=(), parser::parse_url, parser::parse_url_impl, url_aggregator(), and url_aggregator().

◆ url_aggregator() [2/3]

ada::url_aggregator::url_aggregator ( const url_aggregator & u )

default

References url_aggregator().

◆ url_aggregator() [3/3]

ada::url_aggregator::url_aggregator ( url_aggregator && u )

defaultnoexcept

References url_aggregator().

◆ ~url_aggregator()

ada::url_aggregator::~url_aggregator ( )

overridedefault

Member Function Documentation

◆ clear_hash()

void ada::url_aggregator::clear_hash ( )

inline

Removes the hash/fragment from the URL.

Definition at line 615 of file url_aggregator-inl.h.

References ADA_ASSERT_EQUAL, ADA_ASSERT_TRUE, get_hash(), ada::url_components::omitted, to_diagram(), and validate().

Referenced by ~url_aggregator().

◆ clear_port()

void ada::url_aggregator::clear_port ( )

inline

Removes the port from the URL.

Definition at line 566 of file url_aggregator-inl.h.

References ADA_ASSERT_TRUE, ada::url_components::omitted, and validate().

Referenced by set_port(), and ~url_aggregator().

◆ clear_search()

void ada::url_aggregator::clear_search ( )

inlineoverridevirtual

Removes the query/search string from the URL.

Implements ada::url_base.

Definition at line 590 of file url_aggregator-inl.h.

References ADA_ASSERT_EQUAL, ADA_ASSERT_TRUE, get_search(), ada::url_components::omitted, to_diagram(), and validate().

Referenced by set_search(), and ~url_aggregator().

◆ get_components()

ada_really_inline const ada::url_components & ada::url_aggregator::get_components ( ) const

nodiscardnoexcept

Returns the URL component offsets for efficient serialization.

The components represent byte offsets into the serialized URL:

https://user:[email protected]:1234/foo/bar?baz#quux
      |     |    |          | ^^^^|       |   |
      |     |    |          | |   |       |   `----- hash_start
      |     |    |          | |   |       `--------- search_start
      |     |    |          | |   `----------------- pathname_start
      |     |    |          | `--------------------- port
      |     |    |          `----------------------- host_end
      |     |    `---------------------------------- host_start
      |     `--------------------------------------- username_end
      `--------------------------------------------- protocol_end

Returns: A constant reference to the url_components struct.

See also: https://github.com/servo/rust-url

Definition at line 726 of file url_aggregator-inl.h.

Referenced by ~url_aggregator().

◆ get_hash()

std::string_view ada::url_aggregator::get_hash ( ) const

nodiscardnoexcept

Returns the URL's fragment prefixed with '#' (e.g., "#section"). Does not allocate memory. Returns empty view if no fragment is set. The returned view becomes invalid if this url_aggregator is modified.

Returns: A string_view of the hash.

See also: https://url.spec.whatwg.org/#dom-url-hash

Definition at line 738 of file url_aggregator.cpp.

References ada_lifetime_bound, and ada::url_components::omitted.

Referenced by clear_hash(), to_string(), and ~url_aggregator().

◆ get_host()

std::string_view ada::url_aggregator::get_host ( ) const

nodiscardnoexcept

Returns the URL's host and port (e.g., "example.com:8080"). Does not allocate memory. Returns empty view if no host is set. The returned view becomes invalid if this url_aggregator is modified.

Returns: A string_view of host:port.

See also: https://url.spec.whatwg.org/#dom-url-host

Definition at line 752 of file url_aggregator.cpp.

References ada_lifetime_bound.

Referenced by get_origin(), to_string(), and ~url_aggregator().

◆ get_hostname()

std::string_view ada::url_aggregator::get_hostname ( ) const

nodiscardnoexcept

Returns the URL's hostname (without port). Does not allocate memory. Returns empty view if no host is set. The returned view becomes invalid if this url_aggregator is modified.

Returns: A string_view of the hostname.

See also: https://url.spec.whatwg.org/#dom-url-hostname

Definition at line 771 of file url_aggregator.cpp.

References ada_lifetime_bound.

Referenced by has_valid_domain(), and ~url_aggregator().

◆ get_href()

std::string_view ada::url_aggregator::get_href ( ) const

nodiscardconstexprnoexcept

Returns the full serialized URL (the href) as a string_view. Does not allocate memory. The returned view becomes invalid if this url_aggregator is modified or destroyed.

Returns: A string_view into the internal buffer.

See also: https://url.spec.whatwg.org/#dom-url-href

Definition at line 843 of file url_aggregator-inl.h.

References ada_lifetime_bound.

Referenced by ~url_aggregator().

◆ get_origin()

std::string ada::url_aggregator::get_origin ( ) const

nodiscardoverridevirtualnoexcept

Returns the URL's origin (scheme + host + port for special URLs).

Returns: A newly allocated string containing the serialized origin.

See also: https://url.spec.whatwg.org/#concept-url-origin

Implements ada::url_base.

Definition at line 681 of file url_aggregator.cpp.

References ada::scheme::FILE, get_host(), get_pathname(), get_protocol(), ada::scheme::HTTP, ada::scheme::HTTPS, ada::url_base::is_special(), and ada::parse().

Referenced by ~url_aggregator().

◆ get_password()

std::string_view ada::url_aggregator::get_password ( ) const

nodiscardnoexcept

Returns the URL's password component. Does not allocate memory. The returned view becomes invalid if this url_aggregator is modified or destroyed.

Returns: A string_view of the password.

See also: https://url.spec.whatwg.org/#dom-url-password

Definition at line 718 of file url_aggregator.cpp.

References ada_lifetime_bound, and has_non_empty_password().

Referenced by to_string(), and ~url_aggregator().

◆ get_pathname()

std::string_view ada::url_aggregator::get_pathname ( ) const

nodiscardconstexprnoexcept

Returns the URL's path component. Does not allocate memory. The returned view becomes invalid if this url_aggregator is modified or destroyed.

Returns: A string_view of the pathname.

See also: https://url.spec.whatwg.org/#dom-url-pathname

Definition at line 1088 of file url_aggregator-inl.h.

References ada_lifetime_bound, and ada::url_components::omitted.

Referenced by get_origin(), set_pathname(), to_string(), and ~url_aggregator().

◆ get_pathname_length()

ada_really_inline uint32_t ada::url_aggregator::get_pathname_length ( ) const

nodiscardnoexcept

Returns the byte length of the pathname without creating a string.

Returns: Size of the pathname in bytes.

See also: https://url.spec.whatwg.org/#dom-url-pathname

Definition at line 154 of file url_aggregator-inl.h.

References ada::url_components::omitted.

Referenced by ~url_aggregator().

◆ get_port()

std::string_view ada::url_aggregator::get_port ( ) const

nodiscardnoexcept

Returns the URL's port as a string (e.g., "8080"). Does not allocate memory. Returns empty view if no port is set. The returned view becomes invalid if this url_aggregator is modified.

Returns: A string_view of the port.

See also: https://url.spec.whatwg.org/#dom-url-port

Definition at line 728 of file url_aggregator.cpp.

References ada_lifetime_bound, and ada::url_components::omitted.

Referenced by ~url_aggregator().

◆ get_protocol()

std::string_view ada::url_aggregator::get_protocol ( ) const

nodiscardnoexcept

Returns the URL's scheme followed by a colon (e.g., "https:"). Does not allocate memory. The returned view becomes invalid if this url_aggregator is modified or destroyed.

Returns: A string_view of the protocol.

See also: https://url.spec.whatwg.org/#dom-url-protocol

Definition at line 804 of file url_aggregator.cpp.

References ada_lifetime_bound.

Referenced by get_origin(), to_string(), and ~url_aggregator().

◆ get_search()

std::string_view ada::url_aggregator::get_search ( ) const

nodiscardnoexcept

Returns the URL's query string prefixed with '?' (e.g., "?foo=bar"). Does not allocate memory. Returns empty view if no query is set. The returned view becomes invalid if this url_aggregator is modified.

Returns: A string_view of the search/query.

See also: https://url.spec.whatwg.org/#dom-url-search

Definition at line 786 of file url_aggregator.cpp.

References ada_lifetime_bound, and ada::url_components::omitted.

Referenced by clear_search(), to_string(), and ~url_aggregator().

◆ get_username()

std::string_view ada::url_aggregator::get_username ( ) const

nodiscardnoexcept

Returns the URL's username component. Does not allocate memory. The returned view becomes invalid if this url_aggregator is modified or destroyed.

Returns: A string_view of the username.

See also: https://url.spec.whatwg.org/#dom-url-username

Definition at line 708 of file url_aggregator.cpp.

References ada_lifetime_bound, and has_non_empty_username().

Referenced by to_string(), and ~url_aggregator().

◆ has_credentials()

bool ada::url_aggregator::has_credentials ( ) const

nodiscardconstexprnoexcept

Checks if the URL has credentials (non-empty username or password).

Returns: true if username or password is non-empty, false otherwise.

Definition at line 714 of file url_aggregator-inl.h.

References has_non_empty_password(), and has_non_empty_username().

Referenced by to_string(), and ~url_aggregator().

◆ has_empty_hostname()

bool ada::url_aggregator::has_empty_hostname ( ) const

nodiscardconstexprnoexcept

Checks if the URL has an empty hostname (host is set but empty string).

Returns: true if host exists but is empty, false otherwise.

Definition at line 787 of file url_aggregator-inl.h.

References has_hostname().

Referenced by ~url_aggregator().

◆ has_hash()

bool ada::url_aggregator::has_hash ( ) const

nodiscardconstexproverridevirtualnoexcept

Checks if the URL has a fragment/hash component.

Returns: true if hash is present, false otherwise.

Implements ada::url_base.

Definition at line 704 of file url_aggregator-inl.h.

References ada::url_components::omitted.

Referenced by ~url_aggregator().

◆ has_hostname()

bool ada::url_aggregator::has_hostname ( ) const

nodiscardconstexprnoexcept

Checks if the URL has a hostname (including empty hostnames).

Returns: true if host is present, false otherwise.

Definition at line 800 of file url_aggregator-inl.h.

Referenced by has_empty_hostname(), has_port(), and ~url_aggregator().

◆ has_non_empty_password()

bool ada::url_aggregator::has_non_empty_password ( ) const

nodiscardconstexprnoexcept

Checks if the URL has a non-empty password.

Returns: true if password is non-empty, false otherwise.

Definition at line 775 of file url_aggregator-inl.h.

Referenced by get_password(), has_credentials(), and ~url_aggregator().

◆ has_non_empty_username()

bool ada::url_aggregator::has_non_empty_username ( ) const

nodiscardconstexprnoexcept

Checks if the URL has a non-empty username.

Returns: true if username is non-empty, false otherwise.

Definition at line 770 of file url_aggregator-inl.h.

Referenced by get_username(), has_credentials(), and ~url_aggregator().

◆ has_password()

bool ada::url_aggregator::has_password ( ) const

nodiscardconstexprnoexcept

Checks if the URL has a password component (may be empty).

Returns: true if password is present, false otherwise.

Definition at line 780 of file url_aggregator-inl.h.

Referenced by ~url_aggregator().

◆ has_port()

bool ada::url_aggregator::has_port ( ) const

nodiscardconstexprnoexcept

Checks if the URL has a non-default port explicitly specified.

Returns: true if a port is present, false otherwise.

Definition at line 804 of file url_aggregator-inl.h.

References has_hostname().

Referenced by ~url_aggregator().

◆ has_search()

bool ada::url_aggregator::has_search ( ) const

nodiscardconstexproverridevirtualnoexcept

Checks if the URL has a query/search component.

Returns: true if query is present, false otherwise.

Implements ada::url_base.

Definition at line 709 of file url_aggregator-inl.h.

References ada::url_components::omitted.

Referenced by ~url_aggregator().

◆ has_valid_domain()

bool ada::url_aggregator::has_valid_domain ( ) const

nodiscardoverridevirtualnoexcept

Validates whether the hostname is a valid domain according to RFC 1034.

Returns: true if the domain is valid, false otherwise.

Implements ada::url_base.

Definition at line 903 of file url_aggregator.cpp.

References get_hostname().

◆ operator=() [1/2]

url_aggregator & ada::url_aggregator::operator= ( const url_aggregator & u )

default

References url_aggregator().

◆ operator=() [2/2]

url_aggregator & ada::url_aggregator::operator= ( url_aggregator && u )

defaultnoexcept

References url_aggregator().

◆ set_hash()

void ada::url_aggregator::set_hash ( std::string_view input )

Definition at line 400 of file url_aggregator.cpp.

References ADA_ASSERT_TRUE, ada::url_components::omitted, and validate().

◆ set_host()

bool ada::url_aggregator::set_host ( std::string_view input )

Definition at line 667 of file url_aggregator.cpp.

References ADA_ASSERT_TRUE, and validate().

◆ set_hostname()

bool ada::url_aggregator::set_hostname ( std::string_view input )

Definition at line 674 of file url_aggregator.cpp.

References ADA_ASSERT_TRUE, and validate().

◆ set_href()

bool ada::url_aggregator::set_href ( std::string_view input )

The setter functions follow the steps defined in the URL Standard.

The url_aggregator has a single buffer that contains the entire normalized URL. The various components are represented as offsets into that buffer. When you call get_pathname(), for example, you get a std::string_view that points into that buffer. If the url_aggregator is modified, the buffer may be reallocated, and the std::string_view you obtained earlier may become invalid. In particular, this implies that you cannot modify the URL using a setter function with a std::string_view that points into the url_aggregator E.g., the following is incorrect: url->set_hostname(url->get_pathname()). You must first copy the pathname to a separate string. std::string pathname(url->get_pathname()); url->set_hostname(pathname);

The caller is responsible for ensuring that the url_aggregator is not modified while any std::string_view obtained from it is in use.

Definition at line 420 of file url_aggregator.cpp.

References ADA_ASSERT_TRUE, and ada::parse< url_aggregator >().