A vocabulary and associated APIs for HTML and XHTML
details
elementsummary
element followed by flow content.open
interface HTMLDetailsElement : HTMLElement { attribute boolean open; };
The details
element represents a disclosure widget from which the
user can obtain additional information or controls.
The details
element is not appropriate for footnotes. Please see the section on footnotes for details on how to mark up footnotes.
The first summary
element child of the element, if any,
represents the summary or legend of the details. If there is no
child summary
element, the user agent should provide its own legend (e.g.
"Details").
The rest of the element's contents represents the additional information or controls.
The open
content attribute is a boolean
attribute. If present, it indicates that both the summary and the additional information is
to be shown to the user. If the attribute is absent, only the summary is to be shown.
When the element is created, if the attribute is absent, the additional information should be hidden; if the attribute is present, that information should be shown. Subsequently, if the attribute is removed, then the information should be hidden; if the attribute is added, the information should be shown.
The user agent should allow the user to request that the additional information be shown or
hidden. To honor a request for the details to be shown, the user agent must set the open
attribute on the element to the value open
. To honor a request for the information to be hidden, the user agent must
remove the open
attribute from the element.
Whenever the open
attribute is added to or removed from
a details
element, the user agent must queue a task that runs the
following steps, which are known as the details notification task steps, for this
details
element:
If another task has been queued to run the details notification task steps for this
details
element, then abort these steps.
When the open
attribute is toggled
several times in succession, these steps essentially get coalesced so that only one event is
fired.
Fire a simple event named toggle
at the
details
element.
The task source for this task must be the DOM manipulation task source.
The open
IDL attribute must
reflect the open
content attribute.
The following example shows the details
element being used to hide technical
details in a progress report.
<section class="progress window"> <h1>Copying "Really Achieving Your Childhood Dreams"</h1> <details> <summary>Copying... <progress max="375505392" value="97543282"></progress> 25%</summary> <dl> <dt>Transfer rate:</dt> <dd>452KB/s</dd> <dt>Local filename:</dt> <dd>/home/rpausch/raycd.m4v</dd> <dt>Remote filename:</dt> <dd>/var/www/lectures/raycd.m4v</dd> <dt>Duration:</dt> <dd>01:16:27</dd> <dt>Color profile:</dt> <dd>SD (6-1-6)</dd> <dt>Dimensions:</dt> <dd>320×240</dd> </dl> </details> </section>
The following shows how a details
element can be used to hide some controls by
default:
<details> <summary><label for=fn>Name & Extension:</label></summary> <p><input type=text id=fn name=fn value="Pillar Magazine.pdf"> <p><label><input type=checkbox name=ext checked> Hide extension</label> </details>
One could use this in conjunction with other details
in a list to allow the user
to collapse a set of fields down to a small set of headings, with the ability to open each
one.
In these examples, the summary really just summarises what the controls can change, and not the actual values, which is less than ideal.
Because the open
attribute is added and removed
automatically as the user interacts with the control, it can be used in CSS to style the element
differently based on its state. Here, a stylesheet is used to animate the color of the summary
when the element is opened or closed:
<style> details > summary { transition: color 1s; color: black; } details[open] > summary { color: red; } </style> <details> <summary>Automated Status: Operational</summary> <p>Velocity: 12m/s</p> <p>Direction: North</p> </details>
summary
elementdetails
element.HTMLElement
.The summary
element represents a summary, caption, or legend for the
rest of the contents of the summary
element's parent details
element, if any.
dialog
elementopen
interface HTMLDialogElement : HTMLElement { attribute boolean open; attribute DOMString returnValue; void show(optional (MouseEvent or Element) anchor); void showModal(optional (MouseEvent or Element) anchor); void close(optional DOMString returnValue); };
The dialog
element represents a part of an application that a user interacts with
to perform a task, for example a dialog box, inspector, or window.
The open
attribute is a boolean
attribute. When specified, it indicates that the dialog
element is active and
that the user can interact with it.
A dialog
element without an open
attribute
specified should not be shown to the user. This requirement may be implemented indirectly through
the style layer. For example, user agents that support the suggested
default rendering implement this requirement using the CSS rules described in the rendering section.
show
( [ anchor ] )Displays the dialog
element.
The argument, if provided, provides an anchor point to which the element will be fixed.
showModal
( [ anchor ] )Displays the dialog
element and makes it the top-most modal dialog.
The argument, if provided, provides an anchor point to which the element will be fixed.
This method honors the autofocus
attribute.
close
( [ result ] )Closes the dialog
element.
The argument, if provided, provides a return value.
returnValue
[ = result ]Returns the dialog
's return value.
Can be set, to update the return value.
When the show()
method is invoked, the user
agent must run the following steps:
If the element already has an open
attribute, then
abort these steps.
Add an open
attribute to the dialog
element, whose value is the empty string.
If the show()
method was invoked with an argument,
set up the position of the dialog
element, using that argument as the
anchor. Otherwise, set up the default static position of the dialog
element.
Each Document
has a stack of dialog
elements known as the
pending dialog stack. When a Document
is created, this stack must be
initialized to be empty.
When an element is added to the pending dialog stack, it must also be added to the top layer layer. When an element is removed from the pending dialog stack, it must be removed from the top layer. [FULLSCREEN]
When the showModal()
method is invoked,
the user agent must run the following steps:
Let subject be the dialog
element on which the method was
invoked.
If subject already has an open
attribute, then throw an InvalidStateError
exception and abort these steps.
If subject is not in a Document
, then throw
an InvalidStateError
exception and abort these steps.
Add an open
attribute to subject, whose value is the empty string.
If the showModal()
method was invoked with an
argument, set up the position of subject, using that argument as
the anchor. Otherwise, set up the default static position of the dialog
element.
Let subject's Document
be blocked by the modal dialog subject.
Push subject onto subject's
Document
's pending dialog stack.
Let control be the first element in tree order that has an autofocus
attribute specified and whose nearest ancestor
dialog
element is subject, if any.
If there isn't one, then let control be the first element in tree order
that is focusable and whose nearest ancestor dialog
element is subject, if any.
If there isn't one of those either but subject itself is focusable, then let control be subject. (This is unusual.)
If there is no control, then abort these steps.
Run the focusing steps for control.
If at any time a dialog
element is removed from a Document
, then if that dialog
is in that
Document
's pending dialog stack, the following steps must be run:
Let subject be that dialog
element and document be the Document
from which it is being removed.
Remove subject from document's pending dialog stack.
If document's pending dialog stack is not empty, then let document be blocked by the modal dialog that is at the top of document's pending dialog stack. Otherwise, let document be no longer blocked by a modal dialog at all.
When the close()
method is invoked, the user
agent must close the dialog that the method was invoked on. If the method was invoked
with an argument, that argument must be used as the return value; otherwise, there is no return
value.
When a dialog
element subject is to be closed, optionally with a return value result, the user agent
must run the following steps:
If subject does not have an open
attribute, then throw an InvalidStateError
exception and abort these steps.
Remove subject's open
attribute.
If the argument result was provided, then set the returnValue
attribute to the value of result.
If subject is in its Document
's pending dialog
stack, then run these substeps:
Remove subject from that pending dialog stack.
If that pending dialog stack is not empty, then let subject's Document
be blocked by the modal dialog that is at the top of the pending dialog
stack. Otherwise, let document be no longer blocked by a modal
dialog at all.
Queue a task to fire a simple event named close
at subject.
The returnValue
IDL attribute, on
getting, must return the last value to which it was set. On setting, it must be set to the new
value. When the element is created, it must be set to the empty string.
Canceling dialogs: When a Document
's pending dialog
stack is not empty, user agents may provide a user interface that, upon activation, queues a task to fire a simple event named cancel
that is cancelable at the top dialog
element on the Document
's pending dialog stack. The default action of
this event must be to check if that element has an open
attribute, and if it does, close the dialog with no return value.
An example of such a UI mechanism would be the user pressing the "Escape" key.
The containing block of all dialog
elements that are absolutely positioned
must be the initial containing block.
All dialog
elements are always in one of two modes: mundanely aligned,
or magically aligned. When a dialog
element is created, it must be placed
in the mundanely aligned mode and the user agent must set up the default static
position for that element, without an anchor.
When a user agent is to set up the default static position of an element subject without an anchor, if that element is being rendered, it must set up the element such that its top static position, for the purposes of calculating the used value of the 'top' property, is the value that would place the element's top margin edge as far from the top of the viewport as the element's bottom margin edge from the bottom of the viewport, if the element's height is less than the height of the viewport, and otherwise is the value that would place the element's top margin edge at the top of the viewport.
If there is a dialog
element that is mundanely aligned and that is
being rendered when its browsing context changes viewport width (as
measured in CSS pixels), then the user agent must set up the default static position
of all such elements in that browsing context again, still without anchors.
When a dialog
element that is mundanely aligned starts being
rendered, the user agent must set up the default static position of that
element, without an anchor.
This top static position of a mundanely aligned dialog
element must
remain the element's top static position until the set up the default static position
algorithm is once again invoked for that element. (The element's static position is only used in
calculating the used value of the 'top' property in certain situations; it's not used, for
instance, to position the element if its 'position' property is set to 'static'.)
When a user agent is to set up the position of an element subject using an anchor anchor, it must run the following steps:
If anchor is a MouseEvent
object, then run these
substeps:
If anchor's target element does not have a rendered box, or is in a different document than subject, then let subject be mundanely aligned, set up the default static position of subject without an anchor, and abort the set up the position steps.
Let anchor element be an anonymous element rendered as a box with zero height and width (so its margin and border boxes both just form a point), positioned so that its top and left are at the coordinate identified by the event, and whose properties all compute to their initial values.
Otherwise, let anchor element be anchor.
Let subject be magically aligned to anchor element.
While an element A is magically aligned to an element B, the following requirements apply:
If at any time either A or B cease having rendered
boxes, A and B cease being in the same
Document
, or B ceases being earlier than A
in tree order, let subject be mundanely aligned,
and set up the default static position of subject without an
anchor.
A's 'position' property must compute to the keyword 'absolute-anchored' rather than whatever it would otherwise compute to (i.e. the 'position' property's specified value is ignored).
The 'absolute-anchored' keyword's requirements are described below.
The anchor points for A and B are defined as per the appropriate entry in the following list:
The anchor points of A and B are the center points of their respective first boxes' border boxes.
The anchor point of B is the point given by its 'anchor-point' property.
If the anchor point of B is the center point of B's first box's border box, then A's anchor point is the center point of its first box's margin box.
Otherwise, A's anchor point is on one of its margin edges. Consider four hypothetical half-infinite lines L1, L2, L3, and L4 that each start in the center of B's first box's border box, and that extend respectively through the top left corner, top right corner, bottom right corner, and bottom left corner of B's first box's border box. A's anchor point is determined by the location of B's anchor point relative to these four hypothetical lines, as follows:
If the anchor point of B lies on L1 or L2, or inside the area bounded by L1 and L2 that also contains the points above B's first box's border box, then let A's anchor point be the horizontal center of A's bottom margin edge.
Otherwise, if the anchor point of B lies on L3 or L4, or inside the area bounded by L4 and L4 that also contains the points below B's first box's border box, then let A's anchor point be the horizontal center of A's top margin edge.
Otherwise, if the anchor point of B lies inside the area bounded by L4 and L1 that also contains the points to the left of B's first box's border box, then let A's anchor point be the vertical center of A's right margin edge.
Otherwise, the anchor point of B lies inside the area bounded by L2 and L3 that also contains the points to the right of B's first box's border box; let A's anchor point be the vertical center of A's left margin edge.
The anchor point of A is the point given by its 'anchor-point' property.
If the anchor point of A is the center point of A's first box's margin box, then B's anchor point is the center point of its first box's border box.
Otherwise, B's anchor point is on one of its border edges. Consider four hypothetical half-infinite lines L1, L2, L3, and L4 that each start in the center of A's first box's margin box, and that extend respectively through the top left corner, top right corner, bottom right corner, and bottom left corner of A's first box's margin box. B's anchor point is determined by the location of A's anchor point relative to these four hypothetical lines, as follows:
If the anchor point of A lies on L1 or L2, or inside the area bounded by L1 and L2 that also contains the points above A's first box's margin box, then let B's anchor point be the horizontal center of B's bottom border edge.
Otherwise, if the anchor point of A lies on L3 or L4, or inside the area bounded by L4 and L4 that also contains the points below A's first box's margin box, then let B's anchor point be the horizontal center of B's top border edge.
Otherwise, if the anchor point of A lies inside the area bounded by L4 and L1 that also contains the points to the left of A's first box's margin box, then let B's anchor point be the vertical center of B's right border edge.
Otherwise, the anchor point of A lies inside the area bounded by L2 and L3 that also contains the points to the right of A's first box's margin box; let B's anchor point be the vertical center of B's left border edge.
The anchor points of A and B are the points given by their respective 'anchor-point' properties.
The rules above generally use A's margin box, but B's border box. This is because while A always has a margin box, and using the margin box allows for the dialog to be positioned offset from the box it is annotating, B sometimes does not have a margin box (e.g. if it is a table-cell), or has a margin box whose position may be not entirely clear (e.g. in the face of margin collapsing and 'clear' handling of in-flow blocks).
In cases where B does not have a border box but its border box is used by the algorithm above, user agents must use its first box's content area instead. (This is in particular an issue with boxes in tables that have 'border-collapse' set to 'collapse'.)
When an element's 'position' property computes to 'absolute-anchored', the 'float' property does not apply and must compute to 'none', the 'display' property must compute to a value as described by the table in the section of CSS 2.1 describing the relationships between 'display', 'position', and 'float', and the element's box must be positioned using the rules for absolute positioning but with its static position set such that if the box is positioned in its static position, its anchor point is exactly aligned over the anchor point of the element to which it is magically aligned. Elements aligned in this way are absolutely positioned. For the purposes of determining the containing block of other elements, the 'absolute-anchored' keyword must be treated like the 'absolute' keyword.
The trivial example of an element that does not have a rendered box is one whose 'display' property computes to 'none'. However, there are many other cases; e.g. table columns do not have boxes (their properties merely affect other boxes).
If an element to which another element is anchored changes rendering, the anchored element will be be repositioned accordingly. (In other words, the requirements above are live, they are not just calculated once per anchored element.)
The 'absolute-anchored'
keyword is not a keyword that can be specified in CSS; the 'position' property can only compute to
this value if the dialog
element is positioned via the APIs described above.
Elements positioned in this way are not clipped by the 'overflow' property of ancestors (nor moved by the resulting scrolling mechanisms), since the containing block is the initial containing block. Anchoring to an element that is so clipped (and shifted) can therefore result in unexpected effects (where the anchored element moves along with the clipped element, but isn't itself clipped).
User agents in visual interactive media should allow the user to pan the viewport to access all
parts of a dialog
element's border box, even if the element is larger than the
viewport and the viewport would otherwise not have a scroll mechanism (e.g. because the viewport's
'overflow' property is set to 'hidden').
The open
IDL attribute must
reflect the open
content attribute.
This section will eventually be moved to a CSS specification; it is specified here only on an interim basis until an editor can be found to own this.
Value: | none | <position> |
---|---|
Initial: | none |
Applies to: | all elements |
Inherited: | no |
Percentages: | refer to width or height of box; see prose |
Media: | visual |
Computed value: | The specified value, but with any lengths replaced by their corresponding absolute length |
Animatable: | no |
Canonical order: | per grammar |
The 'anchor-point' property specifies a point to which dialog boxes are to be aligned.
If the value is a <position>, the anchor point is the point given by the value, which must be interpreted relative to the element's first rendered box's margin box. Percentages must be calculated relative to the element's first rendered box's margin box (specifically, its width for the horizontal position and its height for the vertical position). [CSSVALUES] [CSS]
If the value is the keyword 'none', then no explicit anchor point is defined. The user agent
will pick an anchor point automatically if necessary (as described in the definition of the
open()
method above).