//
// Copyright (c) 2023 Alan de Freitas (alandefreitas@gmail.com)
//
// Distributed under the Boost Software License, Version 1.0. (See accompanying
// file LICENSE_1_0.txt or copy at https://www.boost.org/LICENSE_1_0.txt)
//
// Official repository: https://github.com/boostorg/url
//


= Parsing

Algorithms which parse URLs return a view which references the
underlying character buffer without taking ownership, avoiding
memory allocations and copies. The following example parses a
string literal containing a
https://datatracker.ietf.org/doc/html/rfc3986#section-3[__URI__,window=blank_]:


[source,cpp]
----
include::example$unit/snippets.cpp[tag=code_urls_parsing_1,indent=0]
----


The function returns a `result` which holds a `url_view`
if the string is a valid URL. Otherwise it holds an `error_code`.
It is impossible to construct a `url_view` which refers to an
invalid URL.

[WARNING]
====
The caller is responsible for ensuring that the lifetime
of the character buffer extends until it is no longer
referenced by the view. These are the same semantics
as that of cpp:std::string_view[].
====

For convenience, a URL view can be constructed directly from the character
buffer in a `string_view`. In this case, it parses the string according
to the
https://datatracker.ietf.org/doc/html/rfc3986#section-4.1[__URI-reference__,window=blank_]
grammar, throwing an exception upon failure. The following two statements
are equivalent:

[source,cpp]
----
include::example$unit/snippets.cpp[tag=code_urls_parsing_2,indent=0]
----


In this library, free functions which parse things are named with the
word "parse" followed by the name of the grammar used to match the string.
There are several varieties of URLs, and depending on the use-case a
particular grammar may be needed. In the target of an HTTP GET request
for example, the scheme and fragment are omitted. This corresponds to the
https://datatracker.ietf.org/doc/html/rfc7230#section-5.3.1[__origin-form__,window=blank_]
production rule described in https://tools.ietf.org/html/rfc7230[rfc7230,window=blank_]. The function
`parse_origin_form`
is suited for this purpose. All the URL parsing functions are listed here:

[cols="a,a,a,a"]
|===
// Headers
|Function|Grammar|Example|Notes

// Row 1, Column 1
|`parse_absolute_uri`
// Row 1, Column 2
|https://datatracker.ietf.org/doc/html/rfc3986#section-4.3[__absolute-URI__,window=blank_]
// Row 1, Column 3
|`pass:[http://www.boost.org/index.html?field=value]`
// Row 1, Column 4
|No fragment

// Row 2, Column 1
|`parse_origin_form`
// Row 2, Column 2
|https://datatracker.ietf.org/doc/html/rfc7230#section-5.3.1[__origin-form__,window=blank_]
// Row 2, Column 3
|`pass:[/index.html?field=value]`
// Row 2, Column 4
|Used in HTTP

// Row 3, Column 1
|`parse_relative_ref`
// Row 3, Column 2
|https://datatracker.ietf.org/doc/html/rfc3986#section-4.2[__relative-ref__,window=blank_]
// Row 3, Column 3
|`pass:[//www.boost.org/index.html?field=value#downloads]`
// Row 3, Column 4
|

// Row 4, Column 1
|`parse_uri`
// Row 4, Column 2
|https://datatracker.ietf.org/doc/html/rfc3986#section-3[__URI__,window=blank_]
// Row 4, Column 3
|`pass:[http://www.boost.org/index.html?field=value#downloads]`
// Row 4, Column 4
|

// Row 5, Column 1
|`parse_uri_reference`
// Row 5, Column 2
|https://datatracker.ietf.org/doc/html/rfc3986#section-4.1[__URI-reference__,window=blank_]
// Row 5, Column 3
|`pass:[http://www.boost.org/index.html]`
// Row 5, Column 4
|Any __URI__ or __relative-ref__

|===


The URL is stored in its serialized form. Therefore, it can
always be easily output, sent, or embedded as part of a
protocol:

// snippet_parsing_url_1bb
[source,cpp]
----
include::example$unit/snippets.cpp[tag=snippet_parsing_url_1bb,indent=0]
----


A `url` is an allocating container which owns its character buffer.
Upon construction from `url_view`, it allocates dynamic storage
to hold a copy of the string.

// snippet_parsing_url_1bc
[source,cpp]
----
include::example$unit/snippets.cpp[tag=snippet_parsing_url_1bc,indent=0]
----


A `static_url` is a container which owns its character buffer for
a URL whose maximum size is known. Upon construction from
`url_view`, it does not perform any dynamic memory allocations.

// snippet_parsing_url_1bd
[source,cpp]
----
include::example$unit/snippets.cpp[tag=snippet_parsing_url_1bd,indent=0]
----


== Result Type

These functions have a return type which uses the `result` alias
template. This class allows the parsing algorithms to report
errors without referring to exceptions.

The functions `result::operator bool()` and `result::operator*`
can be used to check if the result contains an error.

// snippet_parsing_url_1
[source,cpp]
----
include::example$unit/snippets.cpp[tag=snippet_parsing_url_1,indent=0]
----


Since `result::operator bool()` is already checking if `result` contains an
error, `result::operator*` provides an unchecked alternative to get a value
from `result`. In contexts where it is acceptable to throw errors,
`result::value` can be used directly.

// snippet_parsing_url_1b
[source,cpp]
----
include::example$unit/snippets.cpp[tag=snippet_parsing_url_1b,indent=0]
----


Check the reference for `result` for a synopsis of the type. For complete
information please consult the full
https://www.boost.org/doc/libs/1_83_0//libs/system/doc/html/system.html#ref_resultt_e[`result`,window=blank_]
documentation in https://www.boost.org/doc/libs/1_83_0//libs/system/doc/html/system.html[Boost.System,window=blank_].