[/
    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 http://www.boost.org/LICENSE_1_0.txt)

    Official repository: https://github.com/cppalliance/url
]

[/-----------------------------------------------------------------------------]

[section Sanitizing URLs]

This example parses a non-strict or invalid URL
into path components according to its delimiters.
This pattern can be adapted to the requirements of other
applications.

Once the non-strict components are determined, a new URL is
created and its parts are set with the `set_encoded_X`
functions, which will encode any invalid chars accordingly.

This sort of transformation is useful in applications that are
extremely loose in what kinds of URLs they accept, such as
browsers. The sanitized URL can later be used for machine-to-machine
communication.

Using non-strict URLs directly is a security concern in
machine-to-machine communication, is ambiguous, and also
involve an extra cost for the transformations.

Different transformations are required by different applications to
construct a valid URL appropriate for machine-to-machine communication.
For instance, if an invalid relative reference includes something that
looks like a host in the first path segment, browsers usually interpret
that as the host with an implicit "https" scheme. Other applications
also have other implicit schemes.

The example also identifies whether the input url is already valid.
It includes diagnostics that can be used to help the user determine
if a URL is invalid and why it's invalid.

Once all transformations are applied, the result is a URL
appropriate for machine-to-machine communication.

[example_sanitize_url]
[endsect]