SECS-II Message Types

SECS-II Automated Code Generation Tool

by
Hume Integration Software
(C)Copyright 2020

By presenting SECS message type information from a database description we are able to programmatically generate cross-referenced viewing material. Further, we are able to generate program code that can be used with the Hume Datahub SDK to parse or send particular message types.

This presentation differs from SEMI standard documents in the following ways:

The hyperlinked information is far easier to use than PDF documents since message types and data items are seen in one page.
There are no references to octal type codes. Type information is presented using the Hume TSN notation system.
The message format notation also uses the Hume developed TSN system.
Data item definitions have been revised to clarify scalar or array value use. In some cases new data items have been introduced to make this distinction.
Abort message types are not shown since they are processed as a special case of reply messages.
No attempt has been made to include the copious description and explanation material in the SEMI standards. Users of this material will certainly need access to the SEMI E5 Standard document and other standards such as E30. Our aim is only to provide an organizational framework for program code generation.
The distinction between multiblock and single block messages is deemphasized.
The data item cross-references to message types is complete and accurate for the included messages because it is created programmatically. Several documentation errors in E5 are corrected.
Comments have been added which express the personal opinions of the author, and they should not be mistaken as being part of any relevant standard. Further, included descriptions are not the same, or even equivalent to material in the standards.

As far as using the generated code, first ask yourself if it is needed. Be sure to read the reference material for the Datahub library that you are using. The libraries already build-in handling of the core messages of SECS-II including the management of dynamic event reports. There are the handful of cases where you may need to manually edit the generated parsing code:

Items passed as zero length when a scalar value is expected. A scalar value is a non-string, non-list value of length one such as type U4:1 or B:1. The SECS-II standard is loaded with unnecessarily complex special cases. In some situations the standard specifies that a zero length value can be passed to indicate a special meaning. The generated parsing code properly handles zero length arrays, strings, and lists, but you will need to add a test for a zero length value when a scalar value is expected.
Items passed as numeric arrays when a string value or scalar value is expected. Scalar values and strings are parsed with a single assignment statement. For most programming languages, a numeric array requires a loop to set each value element, so if a string was expected, the single assignment statement has to be changed to a loop. The Tcl parser is able to handle a list of values in one statement using lreplace 0 0 instead of lindex 1.
Items passed as a list when a non-list was expected. There are situations such as event reports where the value of a variable can be a list of values instead of a single value. In these situations, the parsing code needs to test the received item format to see if it is a list, and if so, parse the values contained in the list. We have kept the generated parsing code simpler by not adding this logic since it is a less common case.

These documents were last updated 2024-09-10 13:36:39 (UTC). It takes less than a second to regenerate all of the documents from Datahub tables on a 2.8GHz Celeron Linux system!

S1F1R Are You Online? Sent by Host and Equipment

header only