Adding documents to the help system

Nodes metadata file

Information specific to individual HTML files (``nodes'') is supplied in one or more .node metadata files. Each distribution and update contains a node file for each topic directory. An update node file contains information only about nodes in the update. These node files are merged by the ODSSI tools into the topicID.titles database file.

A node metadata file should have a filename extension of .node, and typically resides in the topic directory with the HTML files.

The file contains a record for each HTML file in a topic or book. The node file uses the stanza file data format, with the following syntax for each HTML file (optional fields are marked with square brackets):

   filename.html:
      Title = <TITLE> element of node
      [Version = arbitrary version number]
      [ChapterTOCWeight = number]
      [Keywords = comma-separated list]
      [Status = delete]
      [PrintSequence = print sequence URL]


NOTE: The stanza field names are not case sensitive; however, the field values are.

The fields are:

filename.html:
Identifies the file for which the following fields apply.

Title
Title of the HTML node, usually the same as the <TITLE> element in the file. This title is added to the search database and is displayed in search results.

Version
Arbitrary version number. This field is only used to determine the latest record if more than one is supplied for a given node. An update ships a set of updated node records and a higher version number takes precedence over previous records. The version strings support multiple levels of point releases, and letter modifiers after each.

ChapterTOCWeight
If a record contains a ChapterTOCWeight field, then the given node is a chapter table of contents. It will be included as a single item in the topic or book's main table of contents, sorted according to the weight value. See the list of TOCTypes for more information on how this information is used. A higher number indicates that the item is placed further down the list (similar to chapter numbers). It is expected that the initial set of chapters will use integer weights. Fractional values permit inserting a new chapter between existing chapters. It is recommended that chapters not display chapter numbers so they can be reordered in an update.

Keywords
Not currently implemented, but will be used for a keyword index in the future.

Status
The ``Status=delete'' field permits a node to be made obsolete and removed from the generated files when they are reproduced. When all the records for a particular node are processed, if the one with the latest version number indicates ``delete'', the node is not included in generated files.

The update should still include an HTML file that has the same name as the deleted file, but whose content indicates how the user should proceed. That ensures that links to that file from other files will still work, and provides the user with a way to recover.

PrintSequence
The printing interface in SCOhelp permits the user to select a larger context for printing than a single HTML file. This context is predetermined by the author and contains a sequence of files that are related to the topic. This field specifies a file that lists a sequence of HTML files for that print sequence. If this field is not specified, the user can only print the single node.
The following example shows a single record:
   UsingMouse.html:
      Title = Using a mouse
      Version = 5.0.2
      Keywords = mouse, pointing device
      Status = delete
      PrintSequence = BOOKCHAPTER-1.html

Next topic: How help topic material is processed
Previous topic: Processing topic tables of contents

© 1999 The Santa Cruz Operation, Inc. All rights reserved.
UnixWare 7 Release 7.1.1 - 5 November 1999