Submitter: Henderson

Addressed to: JTC1/SC 24/WG 6 Rapporteur Group on ISO/IEC 8632, CGM

WG secretariat: NNI

Date Circulated by WG secretariat: 1 July 1994

Deadline on response from editor: : 1 October 1994

Defect Report concerning IS 8632:1992 Computer Graphics: Metafile for the storage and transfer of picture description information (CGM) Part 1, Functional

specification.

Qualifier (e.g. error, omission, clarification required): Clarification.

References in document (e.g. page, clause, figure and/or table numbers):

p 21, clause 4.4.3

Nature of defect (complete, concise explanation of the perceived problem):

Obviously the LineWidth element must be in the mode prevailing at its point of occurrence in the metafile. But does changing the mode without respecifying the line

width leave the width alone, or does it change it to the value last specified in that mode?

For example:

VDCExtent 0,0 10000,10000;

LineWidthMode `scaled’;

LineWidth 100.0; %Really wide line% LineWidthMode `absolute’;

LineWidth 1; %Really narrow line% LineWidthMode `scaled’;

Polyline ...;

Do we have a really wide line, or a really narrow line? The treatment of default line width in Version 1 implies that it is a wide line, i.e., at any given point there is really a ScaledLineWidth and an AbsoluteLineWidth, and the mode selects one or the other. While it might seem attractive to say "narrow" for the above case, it would mean defining different behavior for default values and explicit values. This would be horrendous.

This issue also pertains to Colour Selection Mode.

ColourMode `direct’;

LineColour 255 0 0; %red lines%

ColourMode `indexed’

ColourTable 7 0 255 0;

LineColour 7; %green lines%

ColourMode `direct’;

Polyline ....; %red or green?%

Additional related information. When aspects are defined in representation tables, conceptually they are stored with their modes. I think when aspects are defined in segments, they are also conceptually stored with their modes, so that you get the expected results upon COPY segment.

It seems to me that the only clean model, and the only one without the potential for many special cases and strange effects, is that there are not only two forms of specification of attributes in these cases, but conceptually two distinct attributes (or in V3 metafiles, we have 4 different specification mode values).

Add to last para of 4.7.6 on p89:

"Conceptually, for each colour selection mode there is an associated current colour in that mode. This is set by, either the direct colour specified, for ‘direct’, or that defined by the index into the colour table, for ‘indexed’. When the colour selection mode is changed the current colour takes the current colour value for the new colour selection mode."

Add new para at end of 4.4.3:

"For each of the four values -- scaled, absolute, mm, fractional -- of the line width specification mode, there is an associated current value of the line width attribute in that mode. This current value is defined either by the last occurrence of a LINE WIDTH attribute element in that mode, or by the default line width for that mode. The same is true for markers, edges, and filled-area interiors -- there are conceptually four current values of each of these attributes, one associated with the each of the four values of the appropriate specification mode element."