OPAC search RSS and ATOM feeds now have the correct
Content-type sent - "application/rss+xml" and "application/atom+xml",
respectively.
As part of this patch, added an optional fourth parameter
to C4::Output::output_html_with_http_headers to specify
the content type. If that parameter is now supplied, or if
the value of the parameter does not contain at least a "/",
the default type of "text/html" is returned.
No documentation changes.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Fix several validation errors in RSS and Atom feeds
generated from the OPAC:
- add missing guid to RSS elements
- add missing feed ID and element ID to Atom elements
- reflect OPACBaseURL changes
- fix atom:link self links
- add HTML escaping to fields comming from bib record
- set default timestamp for Atom updated elements
Issues identified but not solved in this patch:
- setting Atom updated element for each bib (presumably
from biblio.timestamp)
- possible problems performing paging of Atom feed
Based on successful validation of feeds by feedvalidatory.org,
it is expected that feeds should now work in IE7 and IE6.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Calculate OPACBaseURL from the
CGI environment variables instead of using the
system preference. As a result of this patch,
it should be possible to change the hostname,
protocol, or port of the OPAC without having to
reset a system preference.
Also added a FIXME to opac/unapi - the URL of
the SRU/W server has no necessary relationship
to the URL of the OPAC.
Once this patch is confirmed, the syspref can
be removed.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
If a patron's fine balance is over the limit set by the
noissuecharges syspref, checkouts are blocked. However,
this was not made very clear in the interface.
Improved the display by doing the following:
[1] If the fine balance is over the noissuecharges limit, use the "blocker"
styling (i.e., make all of the text red).
[2] Add wording to explicitly signal whether or not the fine balance is
blocking charges.
Documentation changes: new screenshots of blocked checkouts.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
But that is a separate bug from whether they display or not.
Also added basic error handling if biblionumber is not received, or
if biblionumber is bad, since that otherwise crashes the page.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Note that when the user is not logged in, the Tags input box/button will not display.
The user will see a message telling them to log in to add tags. However, in order to
keep the results display clean, the message will display only once, utilizing the loop
context variable "__first__". Useful stuff!
I also threw OPACShelfBrowser sysprefs.sql, but it still needs added to updatedatabase.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
[1] Display the "No records have been staged" message only
if no import batches exist.
[2] When displaying a batch that has no bibs in it, display
a "There are no records in this batch to import". Also,
do not display the "import record into catalogue" button
for an empty batch.
[3] Add CSS class "problem" to both messages.
Documentation changes: minor; only required if screenshots
exist of the manage staged MARC records page that has no
import batches on it.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
I agree that opac-suggestions.tmpl modification is Not required any longer.
But still, the modification in opac-suggestions.pl is really necessary.
If you donot do that, ppl are presented with ALL suggestions and not
their suggestions by default, which is not what My Suggestions suggests.
Major FIXME's still remain, like the use of GET instead of POST.
The code is also a bit too INCLUDE-happy to net good performance.
The entire mechanism of adding to a batch should probably be proper
AJAX instead of the GET-centric opener.location approach.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
For details: see http://search.cpan.org/~timb/DBI-1.14/DBI.pm
"There's no need to call finish if you're about to destroy or re-execute the statement handle."
In all these cases the sth is falling out of scope, about to be destroyed.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
removing "link" field which has been removed from table auth_subfield_structure
Removing a $dbh->disconnect
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
When editing a child or professional patron, display
the drop-down of relationship types only if at least
one relationship type is defined by the borrowerRelationship
syspref.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
Also, prevent duplicate tags (same user, biblio and term). Strip
leading/trailing whitespace from term. Block whitespace terms.
The CSS for details could be enhanced to emphasize tagstatus more.
Signed-off-by: Joshua Ferraro <jmf@liblime.com>
