This library allows for abstract declaration of available CSS and Javascript resources and their dependencies using html_resource/2. Based on these declarations, html generating code can declare that it depends on specific CSS or Javascript functionality, after which this library ensures that the proper links appear in the HTML head. The implementation is based on mail system implemented by html_post/2 of library html_write.pl.

Declarations come in two forms. First of all http locations are declared using the http_path.pl library. Second, html_resource/2 specifies HTML resources to be used in the head and their dependencies. Resources are currently limited to Javascript files (.js) and style sheets (.css). It is trivial to add support for other material in the head. See html_include//1.

For usage in HTML generation, there is the DCG rule html_requires//1 that demands named resources in the HTML head.

About resource ordering

All calls to html_requires//1 for the page are collected and duplicates are removed. Next, the following steps are taken:

1. Add all dependencies to the set
2. Replace multiple members by `aggregate' scripts or css files. see use_agregates/4.
3. Order all resources by demanding that their dependencies preceede the resource itself. Note that the ordering of resources in the dependency list is ignored. This implies that if the order matters the dependency list must be split and only the primary dependency must be added.

Debugging dependencies

Use ?- debug(html(script)). to see the requested and final set of resources. All declared resources are in html_resource/3. The edit/1 command recognises the names of HTML resources.

Predicates

To be done

- Possibly we should add img//2 to include images from symbolic path notation.

- It would be nice if the HTTP file server could use our location declarations.

 html_resource(+About, +Properties) is det

Register an HTML head resource. About is either an atom that specifies an HTTP location or a term Alias(Sub). This works similar to absolute_file_name/2. See location_path/2 for details. Recognised properties are:

requires(+Requirements)

Other required script and css files. If this is a plain file name, it is interpreted relative to the declared resource. Requirements can be a list, which is equivalent to multiple requires properties.

virtual(+Bool)

If true (default false), do not include About itself, but only its dependencies. This allows for defining an alias for one or more resources.

ordered(+Bool)

Defines that the list of requirements is ordered, which means that each requirement in the list depends on its predecessor.

aggregate(+List)

States that About is an aggregate of the resources in List. This means that if both About and one of the elements of List appears in the dependencies, About is kept and the smaller one is dropped. If there are a number of dependencies on the small members, these are replaced with dependency on the big (aggregate) one, for example, to specify that a big javascript is actually the composition of a number of smaller ones.

mime_type(-Mime)

May be specified for non-virtual resources to specify the mime-type of the resource. By default, the mime type is derived from the file name using file_mime_type/2.

Registering the same About multiple times extends the properties defined for About. In particular, this allows for adding additional dependencies to a (virtual) resource.

 html_current_resource(?About) is nondet

True when About is a currently known resource.

 html_requires(+ResourceOrList)// is det

Include ResourceOrList and all dependencies derived from it and add them to the HTML head using html_post/2. The actual dependencies are computed during the HTML output phase by html_insert_resource//1.

 html_insert_resource(+ResourceOrList)// is det

Actually include HTML head resources. Called through html_post//2 from html_requires//1 after rewrite by html_head_expansion/2. We are guaranteed we will only get one call that is passed a flat list of requested requirements. We have three jobs:

1. Figure out all indirect requirements
2. See whether we can use any `aggregate' resources
3. Put required resources before their requiree.

 use_agregates(+Paths, -Aggregated, -AggregatedBy) is det[private]

Try to replace sets of resources by an `aggregate', a large javascript or css file that combines the content of multiple small ones to reduce the number of files that must be transferred to the client. The current rule says that aggregates are used if at least half of the members are used.

 current_aggregate(-Aggregate, -Parts, -Size) is nondet[private]

True if Aggregate is a defined aggregate with Size Parts. All parts are canonical absolute HTTP locations and Parts is sorted to allow for processing using ordered set predicates.

 requires(+Spec)// is det[private]

 requires(+Spec, +Base)// is det[private]

True if Files is the set of files that need to be loaded for Spec. Note that Spec normally appears in Files, but this is not necessary (i.e. virtual resources or the usage of aggregate resources).

 order_html_resources(+Requirements, +AggregatedBy, -Ordered) is det[private]

Establish a proper order for the collected (sorted and unique) list of Requirements.

 requirements_graph(+Requirements, +AggregatedBy, -Graph) is det[private]

Produce an S-graph (see library(ugraphs)) that represents the dependencies in the list of Requirements. Edges run from required to requirer.

 prop_edges(+Properties)//[private]

Subscribes a list of dependencies from resources that declare their requirements with ordered(true).

 connect_graph(+Graph, -Start, -Connected) is det[private]

Turn Graph into a connected graph by putting a shared starting point before all vertices.

 before(+Term, -Before) is det[private]

Unify Before to a term that comes before Term in the standard order of terms.

Errors

- instantiation_error if Term is unbound.

 res_properties(+Spec, -Properties) is det[private]

True if Properties is the set of defined properties on Spec.

 same_resource(+R1, +R2) is semidet[private]

True if R1 an R2 represent the same resource. R1 and R2 are resource specifications are defined by http_absolute_location/3.

 html_include(+PathOrList)// is det[private]

Include to HTML resources that must be in the HTML <head> element. Currently onlu supports .js and .css files. Extend this to support more header material. Do not use this predicate directly. html_requires//1 is the public interface to include HTML resources.

Arguments:

HTTP	- location or list of these.

 mime_include(+Mime, +Path)// is semidet[multifile]

Hook called to include a link to an HTML resource of type Mime into the HTML head. The Mime type is computed from Path using file_mime_type/2. If the hook fails, two built-in rules for text/css and text/javascript are tried. For example, to include a =.pl= files as a Prolog script, use:

:- multifile
    html_head:mime_include//2.

html_head:mime_include(text/'x-prolog', Path) --> !,
    html(script([ type('text/x-prolog'),
                  src(Path)
                ],  [])).