Java Doc for OutputDocument.java in  » HTML-Parser » jericho-html » au » id » jericho » lib » html » Java Source Code / Java DocumentationJava Source Code and Java Documentation

An OutputDocument represents an original source document that has been modified by substituting segments of it with other text. Each of these substitutions must be registered in the output document, which is most commonly done using the various replace, remove or insert methods in this class. These methods internally one or more OutputSegment objects to define each substitution. After all of the substitutions have been registered, the modified text can be retrieved using the OutputDocument.writeTo(Writer) or OutputDocument.toString() methods.

The registered may be adjacent, and as of version 2.5 may also overlap. In most cases only output segments that have been or legitimately overlap each other. Registering overlapping output segments that generate output will generally yield unexpected results.

If unexpected results are being generated from an OutputDocument, the OutputDocument.getDebugInfo() method provides information on each , which should provide enough information to determine the cause of the problem. In most cases the problem will be caused by overlapping output segments.

The following example converts all externally referenced style sheets to internal style sheets:

 URL sourceUrl=new URL(sourceUrlString);
 String htmlText=Util.getString(new InputStreamReader(sourceUrl.openStream()));
 Source source=new Source(htmlText);
 OutputDocument outputDocument=new OutputDocument(source);
 StringBuffer sb=new StringBuffer();
 List linkStartTags=source.findAllStartTags(Tag.LINK);
 for (Iterator i=linkStartTags.iterator(); i.hasNext();) {
 StartTag startTag=(StartTag)i.next();
 Attributes attributes=startTag.getAttributes();
 String rel=attributes.getValue("rel");
 if (!"stylesheet".equalsIgnoreCase(rel)) continue;
 String href=attributes.getValue("href");
 if (href==null) continue;
 String styleSheetContent;
 try {
 styleSheetContent=Util.getString(new InputStreamReader(new URL(sourceUrl,href).openStream()));
 } catch (Exception ex) {
 continue; // don't convert if URL is invalid
 }
 sb.setLength(0);
 sb.append("<style");
 Attribute typeAttribute=attributes.get("type");
 if (typeAttribute!=null) sb.append(' ').append(typeAttribute);
 sb.append(">\n").append(styleSheetContent).append("\n</style>");
 outputDocument.replace(startTag,sb);
 }
 String convertedHtmlText=outputDocument.toString();
 

The effect of this method is to zero or more in the output document as required to reflect previous modifications to the control's state. The state of a control includes its submission value, , and whether it has been .

The state of the form control should not be modified after this method is called, as there is no guarantee that subsequent changes either will or will not be reflected in the final output. A second call to this method with the same parameter is not allowed. It is therefore recommended to call this method as the last action before the output is generated.

Although the specifics of the number and nature of the output segments added in any particular circumstance is not defined in the specification, it can generally be assumed that only the minimum changes necessary are made to the original document.

The documentation of the OutputDocument.replace(Attributes,Map) method contains more information about the requirements of the map entries.

Specifying a value of true as an argument to the convertNamesToLowerCase parameter causes all original attribute names to be converted to lower case in the map. This simplifies the process of finding/updating specific attributes since map keys are case sensitive.

Attribute values are automatically before being loaded into the map.

This method is logically equivalent to:
OutputDocument.replace(Attributes,Map) replace (attributes, attributes. Attributes.populateMap(Mapboolean) populateMap(new LinkedHashMap(),convertNamesToLowerCase) )

The use of LinkedHashMap to implement the map ensures (probably unnecessarily) that existing attributes are output in the same order as they appear in the source document, and new attributes are output in the same order as they are added.

Example:

 Source source=new Source(htmlDocument);
 Attributes bodyAttributes
 =source.findNextStartTag(0,Tag.BODY).getAttributes();
 OutputDocument outputDocument=new OutputDocument(source);
 Map attributesMap=outputDocument.replace(bodyAttributes,true);
 attributesMap.put("bgcolor","green");
 String htmlDocumentWithGreenBackground=outputDocument.toString();

The output includes details of all the OutputDocument.getRegisteredOutputSegments() registered output segments . a string representation of this object useful for debugging purposes.

The output segments are sorted in order of their in the document.

The returned list is modifiable and any changes will affect the output generated by this OutputDocument. a list all of the OutputSegment objects in this output document.

Use this method if you want to use a customised OutputSegment class.
Parameters:
  outputSegment - the output segment to register.

This is equivalent to OutputDocument.replace(Segment,CharSequence) replace (segment,null).
Parameters:
  segment - the segment to remove.

This is equivalent to the following code:

 for (Iterator i=segments.iterator(); i.hasNext();)
OutputDocument.remove(Segment) remove ((Segment)i.next());

Specifying a null argument to the text parameter is exactly equivalent to specifying an empty string, and results in the segment being completely removed from the output document.
Parameters:
  segment - the segment to replace.
Parameters:
  text - the replacement text, or null to remove the segment.

Specifying a null argument to the text parameter is exactly equivalent to specifying an empty string, and results in the segment being completely removed from the output document.
Parameters:
  begin - the character position at which to begin the replacement.
Parameters:
  end - the character position at which to end the replacement.
Parameters:
  text - the replacement text, or null to remove the segment.

The effect of this method is to zero or more in the output document as required to reflect previous modifications to the control's state. The state of a control includes its submission value, , and whether it has been .

The state of the form control should not be modified after this method is called, as there is no guarantee that subsequent changes either will or will not be reflected in the final output. A second call to this method with the same parameter is not allowed. It is therefore recommended to call this method as the last action before the output is generated.

Although the specifics of the number and nature of the output segments added in any particular circumstance is not defined in the specification, it can generally be assumed that only the minimum changes necessary are made to the original document. If the state of the control has not been modified, calling this method has no effect at all.
Parameters:
  formControl - the form control to replace.
See Also:   OutputDocument.replace(FormFields)

This is equivalent to the following code:

for (Iterator i=formFields.
FormFields.getFormControls getFormControls() .iterator(); i.hasNext();)
OutputDocument.replace(FormControl) replace ((FormControl)i.next());

The state of any of the form controls in the specified form fields should not be modified after this method is called, as there is no guarantee that subsequent changes either will or will not be reflected in the final output. A second call to this method with the same parameter is not allowed. It is therefore recommended to call this method as the last action before the output is generated.
Parameters:
  formFields - the form fields to replace.
See Also:   OutputDocument.replace(FormControl)

The documentation of the OutputDocument.replace(Attributes,Map) method contains more information about the requirements of the map entries.

Specifying a value of true as an argument to the convertNamesToLowerCase parameter causes all original attribute names to be converted to lower case in the map. This simplifies the process of finding/updating specific attributes since map keys are case sensitive.

Attribute values are automatically before being loaded into the map.

This method is logically equivalent to:
OutputDocument.replace(Attributes,Map) replace (attributes, attributes. Attributes.populateMap(Mapboolean) populateMap(new LinkedHashMap(),convertNamesToLowerCase) )

The use of LinkedHashMap to implement the map ensures (probably unnecessarily) that existing attributes are output in the same order as they appear in the source document, and new attributes are output in the same order as they are added.

Example:

 Source source=new Source(htmlDocument);
 Attributes bodyAttributes
 =source.findNextStartTag(0,Tag.BODY).getAttributes();
 OutputDocument outputDocument=new OutputDocument(source);
 Map attributesMap=outputDocument.replace(bodyAttributes,true);
 attributesMap.put("bgcolor","green");
 String htmlDocumentWithGreenBackground=outputDocument.toString();

This method might be used if the Map containing the new attribute values should not be preloaded with the same entries as the source attributes, or a map implementation other than LinkedHashMap is required. Otherwise, the #replace(Attributes, boolean convertNamesToLowerCase) method is generally more useful.

Keys in the map must be String objects, and values must implement the CharSequence interface.

An attribute with no value is represented by a map entry with a null value.

Attribute values are stored unencoded in the map, and are automatically if necessary during output.

The use of invalid characters in attribute names results in unspecified behaviour.

Note that methods in the Attributes class treat attribute names as case insensitive, whereas the Map treats them as case sensitive.
Parameters:
  attributes - the Attributes object defining the span of the segment to replace.
Parameters:
  map - the Map containing the name/value entries.
See Also:    #replace(Attributes, boolean convertNamesToLowerCase)

This method is most commonly used to remove segments of the document without affecting the character positions of the remaining elements.

It is used internally to implement the functionality available through the Segment.ignoreWhenParsing method.

To remove a segment from the output document completely, use the OutputDocument.remove(Segment) method instead.
Parameters:
  begin - the character position at which to begin the replacement.
Parameters:
  end - the character position at which to end the replacement.

As of version 2.5, the presence of overlapping output segments no longer results in an OverlappingOutputSegmentsException . It is now up to the developer to detect unintentional overlapping segments.

If the output is required in the form of a Reader, use CharStreamSourceUtil.getReader(CharStreamSource) CharStreamSourceUtil.getReader(this) instead.
Parameters:
  writer - the destination java.io.Writer for the output.
throws:
  IOException - if an I/O exception occurs.
See Also:   OutputDocument.toString()