Document Object Model Range

8.1. Introduction

A Range identifies a range of content in a Document, DocumentFragment or Attr. It is contiguous in the sense that it can be characterized as selecting all of the content between a pair of boundary-points.

Note: In a text editor or a word processor, a user can make a selection by pressing down the mouse at one point in a document, moving the mouse to another point, and releasing the mouse. The resulting selection is contiguous and consists of the content between the two points.

The term 'selecting' does not mean that every Range corresponds to a selection made by a GUI user; however, such a selection can be returned to a DOM user as a Range.

Note: In bidirectional writing (Arabic, Hebrew), a range may correspond to a logical selection that is not necessarily contiguous when displayed. A visually contiguous selection, also used in some cases, may not correspond to a single logical selection, and may therefore have to be represented by more than one range.

The Range interface provides methods for accessing and manipulating the document tree at a higher level than similar methods in the Node interface. The expectation is that each of the methods provided by the Range interface for the insertion, deletion and copying of content can be directly mapped to a series of Node editing operations enabled by DOM Core. In this sense, the Range operations can be viewed as convenience methods that also enable the implementation to optimize common editing patterns.

This chapter describes the Range interface, including methods for creating and moving a Range and methods for manipulating content with Ranges. The feature string for the interfaces listed in this section is "Range".

8.2. Definitions and Notation

8.2.1. Position

This chapter refers to two different representations of a document: the text or source form that includes the document markup and the tree representation similar to the one described in the What is the Document Object Model? section.

A Range consists of two boundary-points corresponding to the start and the end of the Range. A boundary-point's position in a document or document fragment tree can be characterized by a node and an offset. The node is called the container of the boundary-point and of its position. The container and its ancestors are the ancestor containers of the boundary-point and of its position. The offset within the node is called the offset of the boundary-point and its position. If the container is an Attr, Document, Document Fragment, Element or EntityReference node, the offset is between its child nodes. If the container is a CharacterData, Comment or ProcessingInstruction node, the offset is between the 16-bit units of the UTF-16 encoded string contained by it.

The boundary-points of a Range must have a common ancestor container which is either a Document, DocumentFragment or Attr node. That is, the content of a Range must be entirely within the subtree rooted by a single Document, DocumentFragment or Attr Node. This common ancestor container is known as the root container of the Range. The tree rooted by the root container is known as the Range's context tree.

The container of an boundary-point of a Range must be an Element, Comment, ProcessingInstruction, EntityReference, CDATASection, Document, DocumentFragment, Attr, or Text node. None of the ancestor containers of the boundary-point of a Range can be a DocumentType, Entity or Notation node.

In terms of the text representation of a document, the boundary-points of a Range can only be on token boundaries. That is, the boundary-point of the text range cannot be in the middle of a start- or end-tag of an element or within the name of an entity or character reference. A Range locates a contiguous portion of the content of the structure model.

The relationship between locations in a text representation of the document and in the Node tree interface of the DOM is illustrated in the following diagram:

Range Example

In this diagram, four different Ranges are illustrated. The boundary-points of each range are labelled with s# (the start of the range) and e# (the end of the range), where # is the number of the Range. For Range 2, the start is in the BODY element and is immediately after the H1 element and immediately before the P element, so its position is between the H1 and P children of BODY. The offset of a boundary-point whose container is not a CharacterData node is 0 if it is before the first child, 1 if between the first and second child, and so on. So, for the start of the Range 2, the container is BODY and the offset is 1. The offset of a boundary-point whose container is a CharacterData node is obtained similarly but using 16-bit unit positions instead. For example, the boundary-point labelled s1 of the Range 1 has a Text node (the one containing "Title") as its container and an offset of 2 since it is between the second and third 16-bit unit.

Notice that the boundary-points of Ranges 3 and 4 correspond to the same location in the text representation. An important feature of the Range is that a boundary-point of a Range can unambiguously represent every position within the document tree.

The containers and offsets of the boundary-points can be obtained through the following read-only Range attributes:

  readonly attribute Node startContainer; 
  readonly attribute long startOffset;
  readonly attribute Node endContainer; 
  readonly attribute long endOffset;

If the boundary-points of a Range have the same containers and offsets, the Range is said to be a collapsed Range. (This is often referred to as an insertion point in a user agent.)

8.2.2. Selection and Partial Selection

A node or 16-bit unit is said to be selected by a Range if it is between the two boundary-points of the Range, that is, if the position immediately before the node or 16-bit unit is before the end of the Range and the position immediately after the node or 16-bit unit is after the start of the range. For example, in terms of a text representation of the document, an element would be selected by a Range if its corresponding start-tag was located after the start of the Range and its end-tag was located before the end of the Range. In the examples in the above diagram, the Range 2 selects the P node and the Range 3 selects the text node containing the text "Blah xyz."

A node is said to be partially selected by a Range if it is an ancestor container of exactly one boundary-point of the Range. For example, consider Range 1 in the above diagram. The element H1 is partially selected by that Range since the start of the Range is within one of its children.

8.2.3. Notation

Many of the examples in this chapter are illustrated using a text representation of a document. The boundary-points of a range are indicated by displaying the characters (be they markup or data characters) between the two boundary-points in bold, as in

    <FOO>ABC<BAR>DEF</BAR></FOO>

When both boundary-points are at the same position, they are indicated with a bold caret ('^'), as in

    <FOO>A^BC<BAR>DEF</BAR></FOO>

And when referring to a single boundary-point, it will be shown as a bold asterisk ('*') as in

    <FOO>A*BC<BAR>DEF</BAR></FOO>

8.3. Creating a Range

A range is created by calling the createRange() method on the DocumentRange interface. This interface can be obtained from the object implementing the Document interface using binding-specific casting methods.

  interface DocumentRange {
    Range createRange();
  }

The initial state of the range returned from this method is such that both of its boundary-points are positioned at the beginning of the corresponding Document, before any content. In other words, the container of each boundary-point is the Document node and the offset within that node is 0.

Like some objects created using methods in the Document interface (such as Nodes and DocumentFragments), Ranges created via a particular document instance can select only content associated with that Document, or with DocumentFragments and Attrs for which that Document is the ownerDocument. Such Ranges, then, can not be used with other Document instances.

8.4. Changing a Range's Position

A Range's position can be specified by setting the container and offset of each boundary-point with the setStart and setEnd methods.

  void setStart(in Node parent, in long offset)
                        raises(RangeException);
  void setEnd(in Node parent, in long offset)
                raises(RangeException);

If one boundary-point of a Range is set to have a root container other than the current one for the range, the range is collapsed to the new position. This enforces the restriction that both boundary-points of a Range must have the same root container.

The start position of a Range is guaranteed to never be after the end position. To enforce this restriction, if the start is set to be at a position after the end, the range is collapsed to that position. Similarly, if the end is set to be at a position before the start, the range is collapsed to that position.

It is also possible to set a Range's position relative to nodes in the tree:

  void setStartBefore(in Node node);
                              raises(RangeException);
  void setStartAfter(in Node node);
                       raises(RangeException);
  void setEndBefore(in Node node);
                      raises(RangeException);
  void setEndAfter(in Node node);
                     raises(RangeException);

The parent of the node becomes the container of the boundary-point and the Range is subject to the same restrictions as given above in the description of setStart() and setEnd().

A Range can be collapsed to either boundary-point:

  void collapse(in boolean toStart);

Passing TRUE as the parameter toStart will collapse the Range to its start , FALSE to its end.

Testing whether a Range is collapsed can be done by examining the collapsed attribute:

  readonly attribute boolean collapsed;

The following methods can be used to make a range select the contents of a node or the node itself.

  void selectNode(in Node n);
  void selectNodeContents(in Node n);

The following examples demonstrate the operation of the methods selectNode and selectNodeContents:

Before:
  ^<BAR><FOO>A<MOO>B</MOO>C</FOO></BAR>
After range.selectNodeContents(FOO):
  <BAR><FOO>A<MOO>B</MOO>C</FOO></BAR>
(In this case, FOO is the parent of both boundary-points)
After range.selectNode(FOO):

<BAR><FOO>A<MOO>B</MOO>C</FOO></BAR>

8.5. Comparing Range Boundary-Points

It is possible to compare two Ranges by comparing their boundary-points:

  int compareBoundaryPoints(in CompareHow how, in Range sourceRange) raises(RangeException);

where CompareHow is one of four values: START_TO_START, START_TO_END, END_TO_END and END_TO_START. The return value is -1, 0 or 1 depending on whether the corresponding boundary-point of the Range is before, equal to, or after the corresponding boundary-point of sourceRange. An exception is thrown if the two Ranges have different root containers.

The result of comparing two boundary-points (or positions) is specified below. An informal but not always correct specification is that an boundary-point is before, equal to, or after another if it corresponds to a location in a text representation before, equal to, or after the other's corresponding location.

Let A and B be two boundary-points or positions. Then one of the following holds: A is before B, A is equal to B, or A is after B. Which one holds is specified in the following by examining four cases:

In the first case the boundary-points have the same container. A is before B if its offset is less than the offset of B, A is equal to B if its offset is equal to the offset of B, and A if after B if its offset is greater than the offset of B.

In the second case a child C of the container of A is an ancestor container of B. In this case, A is before B if the offset of A is less than or equal to the index of the child C and A is after B otherwise.

In the third case a child C of the container of B is an ancestor container of A. In this case, A is before B if the index of the child C is less than the offset of B and A is after B otherwise.

In the fourth case, none of three other cases hold: the containers of A and B are siblings or descendants of sibling nodes. In this case, A is before B if the container of A is before the container of B in a pre-order traversal of the Ranges' context tree and A is after B otherwise.

Note that because the same location in a text representation of the document can correspond to two different positions in the DOM tree, it is possible for two boundary-points to not compare equal even though they would be equal in the text representation. For this reason, the informal definition above can sometimes be incorrect.

8.6. Deleting Content with a Range

One can delete the contents selected by a Range with:

  void deleteContents();

deleteContents() deletes all nodes and characters selected by the Range. All other nodes and characters remain in the context tree of the Range. Some examples of this deletion operation are:

(1) <FOO>AB<MOO>CD</MOO>CD</FOO>  -->
<FOO>A^CD</FOO>

(2) <FOO>A<MOO>BC</MOO>DE</FOO>  -->
<FOO>A<MOO>B</MOO>^E</FOO>

(3) <FOO>XY<BAR>ZW</BAR>Q</FOO>  -->
<FOO>X^<BAR>W</BAR>Q</FOO>

(4)
<FOO><BAR1>AB</BAR1><BAR2/><BAR3>CD</BAR3></FOO>
-->  <FOO><BAR1>A</BAR1>^<BAR3>D</BAR3>

After deleteContents() is invoked on a Range, the Range is collapsed. If no node was partially selected by the Range, then it is collapsed to its original start point, as in example (1). If a node was partially selected by the Range and was an ancestor container of the start of the Range and no ancestor of the node satisfies these two conditions, then the Range is collapsed to the position immediately after the node, as in examples (2) and (4). If a node waspartially selected by the Range and was an ancestor container of the end of the Range and no ancestor of the node satisfies these two conditions, then the Range is collapsed to the position immediately before the node, as in examples (3) and (4).

8.7. Extracting Content

If the contents of a range need to be extracted rather than deleted, the following method may be used:

  DocumentFragment extractContents();

The extractContents() method removes nodes from the Range's context tree similarly to the deleteContents() method. In addition, it places the deleted contents in a new DocumentFragment. The following examples illustrate the contents of the returned document fragment:

(1) <FOO>AB<MOO>CD</MOO>CD</FOO>  -->
B<MOO>CD</MOO>

(2) <FOO>A<MOO>BC</MOO>DE</FOO>  -->
<MOO>C<MOO>D

(3) <FOO>XY<BAR>ZW</BAR>Q</FOO>  -->
Y<BAR>Z</BAR>

(4)
<FOO><BAR1>AB</BAR1><BAR2/><BAR3>CD</BAR3></FOO>
-->  <BAR1>B</BAR1><BAR2/><BAR3>C</BAR3>

It is important to note that nodes that are partially selected by the range are cloned. Since part of such a node's contents must remain in the Range's context tree and part of the contents must be moved to the new fragment, a clone of the partially selected node is included in the new fragment. Note that cloning does not take place for selected elements; these nodes are moved to the new fragment.

8.8. Cloning Content

The contents of a range may be duplicated using the following method:

  DocumentFragment cloneContents();

This method returns a DocumentFragment that is similar to the one returned by the method extractContents(). However, in this case, the original nodes and character data in the Range are not removed from the Range's context tree. Instead, all of the nodes and text content within the returned DocumentFragment are cloned.

8.9. Inserting Content

A node may be inserted into a range using the following method:

  void insertNode(in Node n) raises(RangeException);

The insertNode() method inserts the specified node into the Range's context tree. For this method, the end of the range is ignored and the node is inserted at the start of the range.

The Node passed into this method can be a DocumentFragment. In that case, the contents of the fragment are inserted at the start position of the range, but the fragment itself is not. Note that if the Node represents the root of a sub-tree, the entire sub-tree is inserted.

The same rules that apply to the insertBefore() method on the Node interface apply here. Specifically, the Node passed in, if it already has a parent, will be removed from its existing position.

8.10. Surrounding Content

The insertion of a single node to subsume the content selected by a Range can be performed with:

  void surroundContents(in Node n);

The surroundContents() method causes all of the content selected by the range to be rooted by the specified node. Calling surroundContents() with the node FOO in the following examples yields:

     Before:
       <BAR>AB<MOO>C</MOO>DE</BAR>
     After surroundContents(FOO):

<BAR>A<FOO>B<MOO>C</MOO>D</FOO>E</BAR>

Another way of describing the effect of this method on the Range's context tree is to decompose it in terms of other operations:

Remove the contents selected by the range with a call to extractContents().
Insert node where the range is now collapsed (after the extraction) with insertNode()
Insert the entire contents of the extracted fragment into node. Specifically, invoke the appendChild() on node passing in the DocumentFragment returned as a result of the call to extractContents()
Select node and all of its contents with selectNode().

The surroundContents() method raises an exception if the range partially selects a non-Text node. An example of a range for which surroundContents() raises an exception is:

     <FOO>AB<BAR>CD</BAR>E</FOO>

If node has any children, those children are removed before its insertion. Also, if node already has a parent, it is removed from the original parent's childNodes list.

8.11. Miscellaneous Members

One can clone a Range:

  Range cloneRange();

This creates a new Range which selects exactly the same content as that selected by the Range on which the method cloneRange was invoked. No content is affected by this operation.

Because the boundary-points of a range do not necessarily have the same containers, use:

  readonly attribute Node commonAncestorContainer;

to get the ancestor container of both boundary-points that is furthest down from the Range's root container

One can get a copy of all the character data selected or partially selected by a range with:

  DOMString toString();

This does nothing more than simply concatenate all the character data selected by the range. This includes character data in both Text and CDATASection nodes.

8.12. Range modification under document mutation

As a document is modified, the Ranges within the document need to be updated. For example, if one boundary-point of a Range is within a node and that node is removed from the document, then the Range would be invalid unless it is fixed up in some way. This section describes how Ranges are modified under document mutations so that they remain valid.

There are two general principles which apply to Ranges under document mutation: The first is that all Ranges in a document will remain valid after any mutation operation and the second is that, as much as possible, all Ranges will select the same portion of the document after any mutation operation.

Any mutation of the document tree which affect Ranges can be considered to be a combination of basic delete and insertion operations. In fact, it can be convenient to think of those operations as being accomplished using the deleteContents() and insertNode() Range methods.

8.12.1. Insertions

An insertion occurs at a single point, the insertion point, in the document. For any Range in the document tree, consider each boundary-point. The only case in which the boundary-point will be changed after the insertion is when the boundary-point and the insertion point have the same container and the offset of the insertion point is strictly less than the offset of the Range's boundary-point. In that case the offset of the Range's boundary-point will be increased so that it is between the same nodes or characters as it was before the insertion.

Note that when content is inserted at a boundary-point, it is ambiguous as to where the boundary-point should be repositioned if its relative position is to be maintained. There are two possibilities: at the start or at the end of the newly inserted content. We have chosen that in this case neither the container nor offset of the boundary-point is changed. As a result, it will be positioned at the start of the newly inserted content.

Examples:

Suppose the Range selects the following:

<P>Abcd efgh XY blah ijkl</P>

Consider the insertion of the text "inserted text" at the following positions:

1. Before the 'X':

<P>Abcd efgh inserted textXY blah ijkl</P>

2. After the 'X':

<P>Abcd efgh Xinserted textY blah ijkl</P>

3. After the 'Y':

<P>Abcd efgh XYinserted text blah ijkl</P>

4. After the 'h' in "Y blah":

<P>Abcd efgh XY blahinserted text ijkl</P>

8.12.2. Deletions

Any deletion from the document tree can be considered as a sequence of deleteContents() operations applied to a minimal set of disjoint Ranges. To specify how a Range is modified under deletions we need only to consider what happens to a Range under a single deleteContents() operation of another Range. And, in fact, we need only to consider what happens to a single boundary-point of the Range since both boundary-points are modified using the same algorithm.

If a boundary-point is within the content being deleted, then after the deletion it will be at the same position as the one common to the boundary-points of the Range used to delete the contents.

If a boundary-point is after the content being deleted then it is not affected by the deletion unless its container is also the container of one of the boundary-points of the range being deleted. If there is such a common container, then the index of the boundary-point is modified so that the boundary-point maintains its position relative to the content of the container.

If a boundary-point is before the content being deleted then it is not affected by the deletion at all.

Examples:

In these examples, the Range on which deleteContents() is invoked is indicated by the underline.

Example 1.

Before:

<P>Abcd efgh The Range
ijkl</P>

After:

<P>Abcd Range ijkl</P>

Example 2.

Before:

<p>Abcd efgh The Range ijkl</p>

After:

<p>Abcd ^kl</p>

Example 3.

Before:

<P>ABCD efgh The
<EM>Range</EM> ijkl</P>

After:

<P>ABCD <EM>ange</EM> ijkl</P>

In this example, the container of the start boundary-point after the deletion is the Text node holding the string "ange".

Example 4.

Before:

<P>Abcd efgh The Range ijkl</P>

After:

<P>Abcd he Range ijkl</P>

Example 5.

Before:

<P>Abcd <EM>efgh The Range
ij</EM>kl</P>

After:

<P>Abcd ^kl</P>

8.13. Formal Description of the Range Interface

To summarize, the complete, formal description of the Range interface is given below:

Interface Range (introduced in DOM Level 2)

IDL Definition

// Introduced in DOM Level 2:
interface Range {
  readonly attribute Node             startContainer;
                                        // raises(DOMException) on retrieval

  readonly attribute long             startOffset;
                                        // raises(DOMException) on retrieval

  readonly attribute Node             endContainer;
                                        // raises(DOMException) on retrieval

  readonly attribute long             endOffset;
                                        // raises(DOMException) on retrieval

  readonly attribute boolean          collapsed;
                                        // raises(DOMException) on retrieval

  readonly attribute Node             commonAncestorContainer;
                                        // raises(DOMException) on retrieval

  void               setStart(in Node refNode, 
                              in long offset)
                                        raises(RangeException, 
                                               DOMException);
  void               setEnd(in Node refNode, 
                            in long offset)
                                        raises(RangeException, 
                                               DOMException);
  void               setStartBefore(in Node refNode)
                                        raises(RangeException, 
                                               DOMException);
  void               setStartAfter(in Node refNode)
                                        raises(RangeException, 
                                               DOMException);
  void               setEndBefore(in Node refNode)
                                        raises(RangeException, 
                                               DOMException);
  void               setEndAfter(in Node refNode)
                                        raises(RangeException, 
                                               DOMException);
  void               collapse(in boolean toStart)
                                        raises(DOMException);
  void               selectNode(in Node refNode)
                                        raises(RangeException, 
                                               DOMException);
  void               selectNodeContents(in Node refNode)
                                        raises(RangeException, 
                                               DOMException);
  // CompareHow
  const unsigned short      START_TO_START                 = 0;
  const unsigned short      START_TO_END                   = 1;
  const unsigned short      END_TO_END                     = 2;
  const unsigned short      END_TO_START                   = 3;

  short              compareBoundaryPoints(in unsigned short how, 
                                           in Range sourceRange)
                                        raises(DOMException);
  void               deleteContents()
                                        raises(DOMException);
  DocumentFragment   extractContents()
                                        raises(DOMException);
  DocumentFragment   cloneContents()
                                        raises(DOMException);
  void               insertNode(in Node newNode)
                                        raises(DOMException, 
                                               RangeException);
  void               surroundContents(in Node newParent)
                                        raises(DOMException, 
                                               RangeException);
  Range              cloneRange()
                                        raises(DOMException);
  DOMString          toString()
                                        raises(DOMException);
  void               detach()
                                        raises(DOMException);
};

Definition group CompareHow

Passed as a parameter to the compareBoundaryPoints method.

Defined Constants

END_TO_END	Compare end boundary-point of `sourceRange` to end boundary-point of Range on which `compareBoundaryPoints` is invoked.
END_TO_START	Compare end boundary-point of `sourceRange` to start boundary-point of Range on which `compareBoundaryPoints` is invoked.
START_TO_END	Compare start boundary-point of `sourceRange` to end boundary-point of Range on which `compareBoundaryPoints` is invoked.
START_TO_START	Compare start boundary-point of `sourceRange` to start boundary-point of Range on which `compareBoundaryPoints` is invoked.

Attributes

collapsed of type boolean, readonly

TRUE if the range is collapsed

Exceptions on retrieval

`RangeException`		INVALID_NODE_TYPE_ERR: Raised if an ancestor of `refNode` is an Entity, Notation or DocumentType node or if `refNode` is a Document, DocumentFragment, Attr, Entity, or Notation node.
`DOMException`		INVALID_STATE_ERR: Raised if `detach()` has already been invoked on this object.

`RangeException`		INVALID_NODE_TYPE_ERR: Raised if `refNode` or an ancestor of `refNode` is an Entity, Notation or DocumentType node.
`DOMException`		INVALID_STATE_ERR: Raised if `detach()` has already been invoked on this object.

`Node`		`refNode`		The `refNode` value. This parameter must be different from `null`.
`long`		`offset`		The `endOffset` value.

`RangeException`		INVALID_NODE_TYPE_ERR: Raised if the root container of `refNode` is not an Attr, Document or DocumentFragment node or if `refNode` is a Document, DocumentFragment, Attr, Entity, or Notation node.
`DOMException`		INVALID_STATE_ERR: Raised if `detach()` has already been invoked on this object.

BAD_BOUNDARYPOINTS_ERR	If the boundary-points of a range do not meet specific requirements.
INVALID_NODE_TYPE_ERR	If the container of an boundary-point of a range is being set to either a node of an invalid type or a node with an ancestor of an invalid type.

`unsigned short`		`how`
`Range`		`sourceRange`

8. Document Object Model Range

Table of contents

8.1. Introduction

8.2. Definitions and Notation

8.2.1. Position

8.2.2. Selection and Partial Selection

8.2.3. Notation

8.3. Creating a Range

8.4. Changing a Range's Position

8.5. Comparing Range Boundary-Points

8.6. Deleting Content with a Range

8.7. Extracting Content

8.8. Cloning Content

8.9. Inserting Content

8.10. Surrounding Content

8.11. Miscellaneous Members

8.12. Range modification under document mutation

8.12.1. Insertions

8.12.2. Deletions

8.13. Formal Description of the Range Interface