T.R | Title | User | Personal Name | Date | Lines |
---|
1185.1 | When your software can't say no ... | IOSG::MAURICE | Ceci n'est pas une note | Wed Aug 05 1992 10:26 | 18 |
| Hi Hong,
The underlying problem is a bug in the CAB$ATTACH data set. The
difference between attach from the MAIN drawer and an attach from the
other is that the first gives the attachment an attribute DFORMAT with
a value of blank, the second does not have an attribute DFORMAT.
Unfortunately CAB$ATTACH, if asked to provide a value for an attribute
that does not exist, cannot bear to disappoint the caller and so makes
up a random 1-character value to return. This has the effect of fooling
the printing logic when determining how to format the attachment.
I wasn't aware of the problem you report, but it has emerged recently
in another guise where restoring documents with attachments leads to
the same printing problem.
Cheers
Stuart
|
1185.2 | how to obtain this attribute ? | GIDDAY::LEH | | Wed Aug 05 1992 11:13 | 16 |
| Stuart
Thanks for the interesting insight
Just curiosity, why do I keep getting null value for the symbol that
equates to:
CAB$ATTACH_ATTRIBUTES.VALUE["DFORMAT"]
for ANY message with attachments ?
For this problem, is there a simple workaround via script changes ?
Thanks
Hong
|
1185.3 | Possible workround | IOSG::MAURICE | Ceci n'est pas une note | Wed Aug 05 1992 15:06 | 26 |
| Hi Hong,
If the document being attached has a blank Handling field (the FORMAT
field) then the resultant attachment will have a blank DFORMAT field.
For an example of getting a non-blank DFORMAT you could use the U
(update) option on a document and change the handling field. Then
attach the document. After the attach do
<for cab$attach do get CAB$ATTACH_ATTRIBUTES.VALUE["DFORMAT"]
This will return a non-blank value.
As to a work-round a possibility would be to find the place in
WPPRINT.SCP that fetches the value of the DFORMAT
attribute and check that the length of fetched value is 1 and if so
blank it, i.e.
.if fn$length(#PRINT_FORMAT) eq 1 then get #PRINT_FORMAT = ""
All the usual caveats apply (use SCM, don't blame me!), and I'm sure
there must be a more elegant work-round.
Cheers
Stuart
|
1185.5 | SRF ref as THR-16679 | GIDDAY::LEH | | Wed Aug 12 1992 06:13 | 1 |
|
|