Closed (fixed)
Project:
Drupal core
Version:
7.x-dev
Component:
rdf.module
Priority:
Normal
Category:
Task
Assigned:
Unassigned
Issue tags:
Reporter:
Created:
4 Nov 2009 at 22:19 UTC
Updated:
3 Jan 2014 at 00:29 UTC
Jump to comment: Most recent, Most recent file
Comments
Comment #1
Anonymous (not verified) commentedThere was some confusion when I was working on the drupal.org documentation about how to alter mappings that are already defined.
This patch clarifies the comment for rdf_modules_installed and adds a comment to hook_entity_info_alter() to discourage using it to alter the mapping.
Comment #2
Anonymous (not verified) commentedThe current rdf_mapping_save has:
Parameters
$mapping The RDF mapping to save, as an array.
But should havesomething like:
$mapping Associative array containing 'mapping', 'type', 'bundle
Comment #3
christefano commentedRerolling the patch. This should apply cleanly.
I like the new text. Code comments!
Comment #4
scor commentedmarked #621032: Change api.drupal.org title from 'RDFa API' to 'RDF Mapping API' as duplicate. attached patch from #621032: Change api.drupal.org title from 'RDFa API' to 'RDF Mapping API' should be merged with the running patch #3.
Comment #5
scor commentedreopening since first alpha is out, it's time to start cleaning up ;)
Comment #6
scor commenteddocumentation on hook_rdf_namespaces() is missing. though this hook is invoked in common.inc, it should be documented in rdf.api.php
Comment #7
Anonymous (not verified) commentedMerged patches #3 and #4.
Comment #8
Anonymous (not verified) commentedHere is some stuff that I think makes the documentation a little clearer, but someone else should review and make sure that it actually does make things clearer.
I'll add the hook_rdf_namespaces() doc in the next patch.
Comment #9
Anonymous (not verified) commentedThis includes the documentation for hook_rdf_namespaces() in rdf.api.php.
Comment #10
Anonymous (not verified) commentedPer discussions in the Documentation sprint, I made some changes to the Doxygen groups. In the discussion, jhodgden said that RDF shouldn't include its internal functions but should include its hooks on the RDF topic page.
This adds the hooks to the RDF topic page, in addition to the Hooks topic page where they were already listed. It also removes the internal functions by splitting the group and using @addtogroup. It adds the theme functions to the RDF topic page as well.
Comment #11
scor commentedcoded? isn't 'designed' more appropriate here?
should be "Sets the default bundle name...". How about clarifying as follows: Implementation of hook_rdf_mapping() should use this constant for the 'bundle' key when defining a generic set of RDF mapping for an entity type, which will be used as default values for all bundles of this entity type.
'Sets the bundle name' should be 'Sets the default bundle name'
third person
returns
I find the order in which the sentence is constructed confusing. What we do here is merge in default mappings for type or fields which don't have a specified mapping.
this is what typically happens, but the way it's worded sounds like it happens automatically. Adding 'typically' to the sentence should solve this confusion.
rewording suggestion: "..., default entity mappings are not stored in the database. Only overridden mappings are stored in the database."
This is not teaser specific, it is also used in full node mode. (same remarks for all 4 hunks below this one).
this is kind of true but misleading (see above). we don't put it only for teaser here, it's just that in full node the title is not displayed properly. people might wonder what makes this piece of code 'teaser only', I would just not mention teaser here.
".. as the default subject of the predicates embedded as RDFa in the children elements..."
end of line whitespace
It took me some time to remember why we wrote this comment... :) in fact it should be moved to the last line where we use array_merge_recursive(), and clarify the fact that there might be other attributes in $variables['attributes_array'] which we do not want to override, hence the merge.
Powered by Dreditor.
Comment #12
scor commentedalso I just noticed an indentation issue in
Comment #13
Anonymous (not verified) commentedThanks for the feedback!
I think that coded is appropriate, since people unfamiliar with themes might confuse designed for graphically designed (since themes are also visually designed)... but I don't really feel strongly about it, so whatever you think.
I took the first part of yours and put in a second part... let me know if it does/doesn't make sense.
I changed the order, let me know if it still doesn't make sense.
Whoops. And yes, that is much clearer/more accurate.
Cool, moved and slightly reworded
And I also fixed the indentation.
Comment #15
scor commented#13: rdf_code-doc_623684_13.patch queued for re-testing.
Comment #16
scor commentedIt all look good, except one last remark:
This if() block is not teaser specific, we deal with the full node mode a few lines below, so this hunk is not necessary.
Comment #17
scor commentedwe're also ready to go here. #16 is an easy fix, so marking this as novice.
Comment #18
scor commented#720610: Move the core RDF namespace definitions in the rdf module has been committed so this section will need to be updated:
Comment #19
andrewmacpherson commentedFollowing #730420: Make html.tpl.php less dependant on rdf namespaces, we can give
drupal_get_rdf_namespaces()a less HTML-centric explanation...Comment #20
Anonymous (not verified) commentedRerolled to incorporate the last three comments.
Comment #21
scor commentedparfait! thanks Lin!
Comment #22
dries commentedNice work! Committed to CVS HEAD.
Comment #23
Anonymous (not verified) commentedHmmm, I just noticed this @232 in the standard install profile.
We're not inserting a user-defined RDF mapping into the database, are we?
Comment #24
scor commentedgood catch, we should drop the user-defined part.
Comment #25
andrewmacpherson commentedI've posted a patch to fix #23 in a follow up issue:
#738468: Remove "user-defined" comments from standard install profile
"user-defined" appears in more than one comment on the standard install profile, so it's not purely an RDF documentation issue.
Setting status back to fixed here; please follow up in the new issue.