XML DTD Conditional Sections: INCLUDE and IGNORE

Conditional sections in an XML Document Type Definition (DTD) provide a mechanism to selectively enable or disable specific markup declarations. By using the keywords INCLUDE and IGNORE, schema designers can control which parts of a DTD are processed by an XML parser. This article explains the purpose of conditional sections, how INCLUDE and IGNORE function, and how they are commonly implemented with parameter entities for modular schema management.

Understanding Conditional Sections

A conditional section is a construct within an external DTD subset that allows declarations to be parsed or skipped based on a defined keyword. They cannot be used in internal DTD subsets.

The basic syntax for a conditional section is:

<![ KEYWORD [
   <!-- Markup declarations here -->
]]>

The KEYWORD is replaced by either INCLUDE or IGNORE.

The Purpose of `INCLUDE`

When a conditional section uses the INCLUDE keyword, the XML processor treats the declarations within the section normally.

<![ INCLUDE [
   <!ELEMENT notes (#PCDATA)>
]]>

In this example, the notes element declaration is active and fully evaluated by the XML parser as part of the document validation process.

The Purpose of `IGNORE`

When a conditional section uses the IGNORE keyword, the XML parser ignores everything between the opening <![ IGNORE [ and the closing ]]>.

<![ IGNORE [
   <!ELEMENT draftComments (#PCDATA)>
]]>

In this case, the draftComments declaration is ignored. It acts similarly to an XML comment, but it allows you to retain alternative schemas or deprecated rules without permanently deleting them or manually commenting out complex declaration blocks.

Dynamic Control Using Parameter Entities

The primary utility of conditional sections comes from pairing them with parameter entities rather than hardcoding INCLUDE or IGNORE. By referencing a parameter entity as the keyword, developers can toggle entire sections of a DTD from a single location.

Example: Switching Between Draft and Production Modes

<!-- Define the switch -->
<!ENTITY % draft "IGNORE">
<!ENTITY % final "INCLUDE">

<!-- Draft-only elements -->
<![ %draft; [
   <!ELEMENT reviewStatus (#PCDATA)>
   <!ELEMENT internalNotes (#PCDATA)>
]]>

<!-- Production elements -->
<![ %final; [
   <!ELEMENT publicationDate (#PCDATA)>
]]>

Changing the value of %draft; from "IGNORE" to "INCLUDE" instantly activates the draft elements across the entire document without modifying the underlying declarations.

Key Rules and Behaviors

External Subsets Only: Conditional sections are only recognized in external DTD subsets or external parameter entities.
Precedence in Nesting: If a conditional section marked IGNORE contains a nested conditional section marked INCLUDE, the outer IGNORE takes precedence, and the entire block—including the nested INCLUDE—is ignored.
Character Matching: Within an IGNORE section, the parser only scans for nested conditional markers (<![ and ]]>) to ensure proper closure; normal markup errors inside the ignored block will not halt parser execution.