Mike, discriminators and asserts with testKind
'expression' are always evaluated after the component's format annotation,
the timing does not depend on the component kind. See your original
write up.
Regards
Steve Hanson
Architect, Data Format Description Language (DFDL)
Co-Chair, OGF
DFDL Working Group
IBM SWG, Hursley, UK
smh@uk.ibm.com
tel:+44-1962-815848
From:
Mike Beckerle <mbeckerle.dfdl@gmail.com>
To:
Steve Hanson/UK/IBM@IBMGB,
Cc:
dfdl-wg@ogf.org
Date:
19/11/2012 17:43
Subject:
Re: [DFDL-WG]
Action 186: Re: DFDL Statement Evaluation Timing (Assert, Discriminator,
SetVariable, NewVariableInstance)
Easily fixed. Global replace done.
On Mon, Nov 19, 2012 at 12:36 PM, Steve Hanson <smh@uk.ibm.com>
wrote:
Mike, I think we agreed on 'resolved
set of annotations' instead of 'combined annotations' ?
I had the action to create the final errata wording for this item (Action
186).
Last action issue was to change wording to insure it was all dealing with
combined annotations uniformly, and change "annotated construct"
to "schema component".
I've done that, but also reorganized this as errata on specific sections
of the spec consistent with the rest of our errata, and turned the 'ordering'
section into numbered lists for clarity.
-------------------------------------------------------
Errata section 6.2. Clarification: At any single annotation point of
the schema, there can be only one format annotation (dfdl:format, dfdl:element,
dfdl:sequence, dfdl:choice, dfdl:group, dfdl:simpleType).
Errata section 3. Glossary entries
Glossary: Annotation point: A location within a DFDL schema where DFDL
annotation elements are allowed to appear.
Note that annotation point is defined in section 6.2, but it belongs also
in the glossary.
Glossary: DFDL Statement annotations, or just DFDL Statements, are
the annotation elements dfdl:assert, dfdl:discriminator, dfdl:setVariable,
and dfdl:newVariableInstance.
Glossary: DFDL Defining Annotations are the annotation elements dfdl:defineFormat,
dfdl:defineVariable, and dfdl:defineEscapeScheme
Glossary: DFDL Format Annotations are the annotation elements dfdl:format,
dfdl:element, dfdl:simpleType, dfdl:group, dfdl:sequence, and dfdl:choice.
Glossary: Remove existing definition for "Format Annotations"
Glossary: Physical layer - revised to: "Physical Layer - A DFDL
Schema adds DFDL annotations onto an XSDL language schema. The annotations
describe the physical representation or physical layer of the data."
Glossary: Resolved set of annotations: When DFDL annotations
appear on a group reference and the sequence or choice of the referenced
global group, or appear among an element reference, an element declaration,
and its type definition, then they are combined together and the resulting
set of annotations is referred to as the resolved set of annotations
for the schema component.
Errata Section 7.3.1:
Remove "DFDL asserts can be placed on components within a DFDL model."
as this is made more specifically subsequently.
Replace "More than one dfdl:assert may be used at an annotation point.
The dfdl:asserts will be evaluated in the order defined in the schema."
with "If the resolved set of annotations for a schema component contain
multiple dfdl:assert statements, then those with testKind='pattern' are
executed before those with testKind='expression' (the default). However,
within each group the order of execution among them is not specified. Schema
authors can insert sequences to control the timing of evaluation of statements
more precisely. See Section REF: Evaluation Order for Statement Annotations."
Property testPattern: Add this to the description:
Note that the pattern is used to match against the entire representation
of the component; hence, the framing (including initiators) is visible
to the pattern.
It is a schema definition error if dfdl:alignment is not 1 .
It is a schema definition error if dfdl:leadingSkip is not 0.
It is a schema definition error if dfdl:encoding is not defined.
Errata Section 7.3.1, and 7.4.1: Assert and Discriminator Placement
Correct description to be consistent with: dfdl:assert and dfdl:discriminator
can be placed as annotations on sequence, choice, group references, local
and global element declarations, element references, and simple type definitions.
Errata Section 7.4.1:
Replace: "Any one annotation point can contain only a single dfd:discriminator
or one or more dfdl:asserts,
but not both. It is a schema definition error otherwise." with:
"The resolved set of annotations for a schema component can contain
only a single dfd:discriminator or one or more dfdl:asserts, but not both.
It is a schema definition error otherwise."
Property testPattern: Add this to the description:
Note that the pattern is used to match against the entire representation
of the component; hence, the framing (including initiators) is visible
to the pattern.
It is a schema definition error if dfdl:alignment is not 1 .
It is a schema definition error if dfdl:leadingSkip is not 0.
It is a schema definition error if dfdl:encoding is not defined.
Errata Section 7.8 newVariableInstance
dfdl:newVariableInstance can be placed as an annotation on sequence, choice,
and group references.
Replace: "It is a schema definition error to have more than one newVariableInstances
for the same variable at any given point in the document." with "The
resolved set of annotations for any schema component can contain multiple
dfdl:newVariableInstance annotations, but they must each refer to a different
variable. It is a schema definition error otherwise."
Errata Section 7.9: setVariable
Correct language to be consistent with: "dfdl:setVariable may be placed
as an annotation on sequence, choice, group references, local and global
element declarations for elements of simple type, element references to
elements of simple type, and simple type definitions. "
Replace "It is a schema definition error to have more than one dfdl:setVariable
for the same variable at any given point in the document." with "The
resolved set of annotations for any schema component can contain multiple
dfdl:setVariable annotations, but they must each refer to a different variable.
It is a schema definition error otherwise."
Errata: New Section 7.10: Evaluation Order for Statement Annotations
Of the resolved set of annotations for a schema component, some are statement
annotations and the order of their evaluation relative to the actual processing
of the schema component itself (parsing or unparsing per its format annotations)
is as given in the ordered lists below.
Implementations are free to optimize by recognizing and executing discriminators
or assertions with testKind='expression' earlier so long as the resulting
behavior is consistent with what results from the this description:
For elements:
1. dfdl:discriminator
with testKind='pattern' (parsing only)
2. dfdl:assert(s) with testKind='pattern' (parsing
only)
3. Processing of the element schema component
itself (as per format annotations)
4. dfdl:setVariable(s) - See note below.
5. dfdl:discriminator with testKind='expression'
(parsing only) - Evaluated even on failure of steps 3, or 4. See note below.
6. dfdl:assert(s) with testKind='expression'
(parsing only)
For sequences and choices:
1. dfdl:discriminator with testKind='pattern'
(parsing only)
2. dfdl:assert(s) with testKind='pattern'
(parsing only)
3. dfdl:newVariableInstance(s)
4. dfdl:setVariable(s)
5. dfdl:discriminator with testKind='expression'
(parsing only) - Evaluated even on failure of steps 3 or 4. See note below.
6. dfdl:assert(s) with testKind='expression'
(parsing only)
7. Processing of the sequence/choice component
itself (as per format annotations)
Note on Discriminators with testKind='expression'
When parsing, an attempt to evaluate a discriminator must be made even
if preceding statements or the parse of the schema component ended in a
parse error.
This is because a discriminator's expression could evaluate to true thereby
resolving a point of uncertainty even if the complete parsing of the construct
ultimately caused a parse error.
Such discriminator evaluation has access to the DFDL Infoset of the attempted
parse as it existed immediately before detecting the parse failure. Attempts
to reference parts of the DFDL Infoset that do not exist are parse errors.
It is also a parse error to read any variables defined by any dfdl:newVariableInstance
statement evaluations that failed, or to read any variables set by any
dfdl:setVariable statements that failed.
Note on the dfdl:setVariable Statement for Elements
Of the resolved set of annotations, the dfdl:setVariable statements for
an element schema component (combined from an element reference, element
declaration, and base and derived simpleType definitions) are executed
after the parsing of the element.
This implies that these variables are set after the evaluation of expressions
corresponding to any computed DFDL format properties for the same schema
component, and so the variables may not be referenced from expressions
that compute DFDL properties.
That is, if an expression is used to provide the value of a format property
(such as dfdl:terminator, or dfdl:byteOrder), the evaluation of that format
property expression occurs before any dfdl:setVariable annotation from
the resolved set of annotations for that schema component are executed;
hence, the expression providing the value of the format property may not
reference the variable. Schema authors can insert sequences to provide
more precise control over when variables are set.
Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number
741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6
3AU