Bug 45070 - Spurious WARNING on span="inherit"
Summary: Spurious WARNING on span="inherit"
Status: NEW
Alias: None
Product: Fop - Now in Jira
Classification: Unclassified
Component: general (show other bugs)
Version: 0.94
Hardware: PC Linux
: P3 normal
Target Milestone: ---
Assignee: fop-dev
URL: http://lists.oasis-open.org/archives/...
Keywords:
Depends on:
Blocks:
 
Reported: 2008-05-23 05:43 UTC by Andrew McFarland
Modified: 2012-04-11 06:17 UTC (History)
1 user (show)



Attachments
simple docbook file that demonstrates the problem (432 bytes, text/plain)
2008-05-23 05:44 UTC, Andrew McFarland
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Andrew McFarland 2008-05-23 05:43:17 UTC
When I transform the attached file to XSL-FO and process the resulting FO with FOP, I get this warning:

WARNING: span="inherit" on fo:block, but no explicit value found on the parent FO.

I'm using the 1.73.2 DocBook XSL.

I get this problem with both FOP 0.94 and FOP 0.95beta.
Comment 1 Andrew McFarland 2008-05-23 05:44:40 UTC
Created attachment 21991 [details]
simple docbook file that demonstrates the problem
Comment 2 Vincent Hennebert 2008-05-23 07:02:19 UTC
(This issue came up on the docbook-apps mailing list.) There are two problems IMO:
- I don't see why FOP should issue a warning while such a construct is perfectly legal. I realize that this may help debugging stylesheets, but this will also annoy most users.
- The code in PropertyMaker.make(PropertyList, String, FObj) looks suspicious. It tries to get the explicit value set on the parent, whereas it should get its /computed/ value. Granted, the end result is probably the same, still.

Leaving it over to property specialists...

Vincent

Comment 3 Andreas L. Delmelle 2008-05-23 08:08:32 UTC
(In reply to comment #2)
> (This issue came up on the docbook-apps mailing list.) 
> There are two problems IMO:
> - I don't see why FOP should issue a warning while such a construct is
> perfectly legal. I realize that this may help debugging stylesheets, but this
> will also annoy most users.

Granted...
Question in this particular case, however, is what the stylesheet authors hope to achieve by adding  'span="inherit"' anyway. The property only has effect for areas returned by a flow-child. Although the property is not inherited, further descendants automatically take on the characteristics of their parent (see: http://www.w3.org/TR/xsl/#span)

> - The code in PropertyMaker.make(PropertyList, String, FObj) looks suspicious.
> It tries to get the explicit value set on the parent, whereas it should get its
> /computed/ value. Granted, the end result is probably the same, still.

I think the reason behind the warning is indeed stylesheet debugging, in the sense that one would expect a value on the parent for a specified value of 'inherit' on the child (?)

The keyword is permitted as a value for inherited properties, but hardly useful there...
If used for non-inherited properties, then it should indeed evaluate to the computed value.

The explicit value obtained in PropertyMaker is used /only/ to check whether there was a value specified on the parent. If this is not the case, then the computed value will evaluate to the initial value, and so could just as well have been omitted...

All the same for me to have FOP shut up in those cases, or reduce it to debug-level.
Comment 4 Glenn Adams 2012-04-07 01:44:40 UTC
resetting P2 open bugs to P3 pending further review
Comment 5 Glenn Adams 2012-04-11 06:17:29 UTC
change status from ASSIGNED to NEW for consistency