{23} Trac comments (3729 matches)

Results (2601 - 2700 of 3729)

Ticket Posixtime Author Newvalue
#2639 1345146289000000 rgrp Port of fix for recline issue 196 in https://github.com/okfn/ckan/commit/fd1abd9d5701b4addf7e709c1766d6ffed1bb8bf Fix of embed issue in https://github.com/okfn/ckan/commit/fd1abd9d5701b4addf7e709c1766d6ffed1bb8bf
#2764 1343320440000000 rgrp Duplicate of #2732
#2783 1343814401000000 rgrp My 2c: this is very low priority atm (it will also be hard to do). What I'd vote for is some decent default patterned background or similar (what we have originally wasn't *that* bad ...)
#2881 1345626056000000 rgrp @seanh - this was the material from the wiki moved to the docs. I agree it is somewhat incomplete but we certainly shouldn't remove it without something better in place. Also, IMO this is currently very low priority ...
#2890 1346175867000000 rgrp I know this is tricky. The issue is that Datastore does not exist *just* to provided data views - it also provides a bunch of other stuff like a Data API and, well, a Datastore. I also think there are different audiences. Data view info is about publishers and (viewing) users, while the Data API is for (non-CKAN) devs and CKAN devs and the Datastore sysadmin section is for deployers. My feeling is therefore keeping things as is but with more stuff in the Data View seciton about it all works: * Data viewer: explanation for publisher, links to next 2 sections * Datastore page (as we have it) * Includes datastorer * Using Data API page (as we have it) So main change is a better data viewer page ...
#460 1282330016000000 [email protected] Note, I'm not a sysadmin but I ''can'' see the Status dropdown for this package, with values "active" and "deleted". Not knowing much about the permission system, I was assuming that's because I created the package. I am however certain that I did not modify (or even notice) this dropdown in the mysterious edit that led to the state change from "active" to "None".
#460 1285443859000000 [email protected] I believe the initial report is incorrect. It states that the status was changed from "active" to "deleted". I believe that it was actually changed from "active" to "None". This might indicate a bug in the code: The value of the status field is lost.
#234 1338206486000000 ross 18 months, no activity.
#282 1338206417000000 ross This was discussed with Toby who has a ticket for this same feature because of disqus requirements.
#377 1338206330000000 ross We have another ticket for fixing the API
#863 1338206455000000 ross UI has changed rather a lot in last 18 months, so I killing this bug.
#909 1346669602000000 ross Backlog until funded.
#1035 1338206251000000 ross Removing until we can come up with a case for implementing
#1171 1347358705000000 ross I have implemented this on the datahub, but this change likely needs to be made to CKAN core. Probably as part of 2.0. The format on tdh is currently: ${TITLE}. ${AUTHOR}. Retrieved ${UTC time}. ${DATASET LINK} Best discuss and agree on what format should be supported. Mark, I've assigned this to you as it is tagged academia ;)
#1279 1338212028000000 ross Duplicate of #2366
#1328 1340191065000000 ross Let's see if this is still an issue in 1.8
#1345 1340190737000000 ross Moved out of milestone so that it gets looked at again soon
#1432 1324290164000000 ross PhantomJS ( http://www.phantomjs.org/ ) looks like the perfect way to run a JS scraper offline but still within the context of a browser. Would need to consider how the script would be executed over multi-page datasets, but I'm fairly sure it would be possible to come up with a workable solution.
#1444 1340190852000000 ross 7 months, no activity, dead bug.
#1452 1340190587000000 ross I believe this has been addressed by Toby.
#1475 1326283573000000 ross Initial draft written at http://wiki.ckan.org/Writing_asynchronous_tasks
#1485 1323186842000000 ross Implementing an initial IPackageController in ckanext-example
#1508 1323184952000000 ross Custom form is referenced in #1485
#1533 1323359638000000 ross Investigating: 1. The XPATH that the middleware is currently using will definitely not match the navigation on 1.5.1, we'll need to change the xpath and also extract the anchors from the LI tags that are retrieved from the wordpress site. 2. Getting the main wordpress content just doesn't seem to work at all. Lots of class/id confusion. Will fix for 1.5.1 in branch
#1533 1338202636000000 ross This has been superceded by #2234
#1549 1324385952000000 ross Comment from John Erickson (on ckan-discuss): FYI, such ability to maintain an identifier was an original motivation/requirement of the DOI (which is why the Handle System was the obvious choice for implementation). Note that providing the ability to maintain the short URL implies some kind of authority --- either based on the originating user or a ckan administrator. The idea of analytics as a motivator is very good; this is a great feature of bit.ly Another feature to consider might be short URL lookup based on target URL at short URL creation time, to avoid multiple short URL for the same target.
#1551 1346662048000000 ross Webstore appears to hav been deprecated in favour of datastore.
#1553 1326288657000000 ross I think this is caused by the javascript xhr call getting redirected (quietly) to the login page, so the original javascript can't complete as it has a login form rather than the auth form it is looking for. Maybe we should change the JS to realise that it is at the login page and do something different than pretend to still be uploading?
#1553 1340628027000000 ross Aron, can I assign this to you for when you do the JS based uploading on demo site/new UI?
#1574 1325582298000000 ross Related: #1608 Merge storage into Core #1609 Celery task for ckanext-archiver to write to webstore.
#1574 1325590211000000 ross Added #1611 as relevant to webstore integration
#1574 1325847164000000 ross Updated description.
#1574 1325847230000000 ross __Some issues to be addressed__ * What about existing content stored on google/s3? * Metadata for OFS should use username/userid? If user changes username we could just change the pairtree metadata json file, it should not be automated or easy for users to change username though. * OFS Metadata should contain the form data from CKAN upload * Remove _owner from OFS metadata (unnecessary) __Some issues to still consider__ * Allow .sqlite upload? * Allow .sql upload?
#1574 1326904965000000 ross Added related ticket #1681 for documentation/announce to be done asap.
#1579 1338212496000000 ross I think the new datastore solves this.
#1608 1325586861000000 ross We should consider some admin dashboard hooks for data such as total file storage usage+count
#1608 1325681756000000 ross The current storage API might need cleaning up and merging into the normal API. Although this may require changes to the current API Controller. Not sure, so adding a note as a reminder.
#1609 1325586917000000 ross from the pad: What happens when I upload a CSV? * End up with file in permanent storage * IF file is of type ... csv,xls,xlsx,sqlite,.sql * End up with new db in webstore * Where? {username}/{resource-id}/... * If a single table: name it after the file name (appropriately slugified) * A resource *always* corresponds to a 'database' in webstore ... * In Data Explorer have "Sheets" tab ... * Resource url = /dataset/{x}/resource/{y}/link -> cached_url ... * Move uploaded file into permanent storage
#1610 1325852472000000 ross Duplicate of https://github.com/okfn/webstore/issues/41
#1611 1325846987000000 ross See #1550
#1649 1331547577000000 ross After looking at some options via http://trac.ckan.org/ticket/2209 I think a simple RDF template or triple in an internal ckanext-semweb might indeed be the most appropriate way forward with this. n3 sounds useful, but given we could allow people to specify an RDF template we may as well do that as well as it would then suffice for ecportal who appear to want rdf as well. Will check out json-ld as an option too.
#1650 1326730780000000 ross Some translations are currently done with full locales (i.e. pt_BR) but we don't also have pt_PT and so where I would normally default to the language if there are only one translation I am unsure what to do in this case. Currently supports the short language code or the locale code, but this will need to be resolved before we address #1653
#1650 1338206684000000 ross Branch exists for #1650
#1650 1343126718000000 ross Probably not going to do this for ECP as John found a cleaner way of doing it.
#1650 1346670010000000 ross Closed as unnecessary now.
#1652 1340187535000000 ross Is this still required?
#1656 1326711659000000 ross Dupe of #1657
#1658 1338206622000000 ross Being done elsewhere
#1661 1338206091000000 ross We have another ticket for upgrading dependencies.
#1669 1326803055000000 ross Subtickets are at: * Publisher profile analysis #1670 [2d] * Publisher profile sysadmin authz #1671 [1d] * Publisher profile user authz #1672 [1d] * Test auth logic hooks for publisher profile #1674 [3d] * Publisher admin authz #1675 [3d] * Publisher editor authz #1676 [2d] * Add approval status field to groups #1673 [1d]
#1670 1328519319000000 ross Changed to reflect the need to generate documentation
#1670 1330603657000000 ross Explanation of how to work with the auth profile now at http://wiki.ckan.org/Working_with_the_publisher_auth_profile Needs reading/testing. Started admin guide documentation in branch feature-1670-publisher-profile-docs
#1673 1327941049000000 ross Implemented in 31199375a34e04af8158e60335816236029ff96e as part of the feature-1669-publisher-profile branch.
#1675 1328519608000000 ross ckanext publisher_form ui needs reflecting to ensure only admins can admin the group.
#1675 1328709915000000 ross Fixed in #1758
#1680 1326900832000000 ross David suggested that we could implement this with a group extra instead of a new attribute.
#1684 1340187582000000 ross Left in backlog, but may need re-assessing depending on the data services plan.
#1685 1340188083000000 ross May be subsumed into a service, will keep ticket as a reminder to cleanup
#1685 1346670530000000 ross 8 months and no movement, looks like a wontfix.
#1693 1338206601000000 ross Toby fixed this up
#1742 1328116635000000 ross Potentially valid diff, needs test + patch when I have time. --- a/ckan/controllers/api.py +++ b/ckan/controllers/api.py @@ -592,6 +592,8 @@ class ApiController(BaseController): def is_slug_valid(self): slug = request.params.get('slug') or '' slugtype = request.params.get('type') or '' + if slug in ['new', 'edit', 'search']: + return self._finish_ok( dict(valid=False) ) if slugtype==u'package': response_data = dict(valid=not bool(package_exists(slug))) return self._finish_ok(response_data)
#1755 1328630657000000 ross Implemented in v1.0-dev of ckanext-dgu and feature-1645-apply-simple-theme of ckanext-dgutheme
#1760 1328709521000000 ross See src/ckanext-dgu/ckanext/dgu/bin/import_publishers.py which will pull publishers from the provided CSV file and build the hierarchy in the database. This file describes how to obtain the publisher list (with hierarchy detail).
#1761 1328703769000000 ross Have overridden the history function in the DGU package_gov3.py controller to make sure the viewer is in at least one group.
#1796 1339586499000000 ross LXML removed, update requirements on the extensions
#1831 1330956631000000 ross Already have a branch for this.
#1831 1331549284000000 ross Email is not currently 'unique' in the user model. Find out if this is intentional or not.
#1831 1331655458000000 ross Decided to delay this until later but code is in feature-1831-login-by-email
#2204 1334218735000000 ross I'd be happier adding this to another ticket rather than moving the target for this ticket half way through so have created #2284 Should only be a short ticket in the next sprint.
#2204 1335795665000000 ross Giving this ticket to Aron to fix the front end
#2204 1337594723000000 ross Was waiting for documentation, can close now.
#2209 1331547639000000 ross Option 4 is the most appealing at this point, although you probably should look at #2209
#2209 1331547674000000 ross Previous comment should read: Option 4 is the most appealing at this point, although you probably should look at #1649
#2209 1331551170000000 ross Current thinking is that option 4 is a default (as per ckanext-rdf) rdf output that is generated not in code (as currently) but using a genshi xml template to read the package into an RDF format (as if it were HTML). This would then be overrideable so that for ecportal where the format of the RDF is different (change of vocabs etc) we can just point the config to a new template. Pros: Easy to implement Easy to use Not hard-coded as currently Fast execution Cons: Requires knowledge of required RDF output if default is not useful RDF and not any of the other formats yet. Only works with package/resource/tags unless more work is done
#2209 1331551393000000 ross Can't edit comments .. so Current thinking is that option 4 is a default (as per ckanext-rdf) rdf output that is generated not in code (as currently) but using a genshi xml template to read the package into an RDF format (as if it were HTML). This would then be overrideable so that for ecportal where the format of the RDF is different (change of vocabs etc) we can just point the config to a new template. Pros: Easy to implement Easy to use Not hard-coded as currently Fast execution Cons: Requires knowledge of required RDF output if default is not useful RDF and not any of the other formats yet. Only works with package/resource/tags unless more work is done
#2210 1331549350000000 ross This is now done through the logic layer member_create/member_list/member_delete
#2221 1331561095000000 ross Fixed in https://github.com/okfn/ckan/compare/1eeda11...ad2358a
#2224 1338205358000000 ross Being done by Aron
#2234 1337239588000000 ross Yes, definitely need a replacement for wordpresser and sooner rather than later. Doing some experimental noodling to work out an estimate.
#2239 1334581838000000 ross Changes to schema/rdf output
#2240 1332408769000000 ross Distributed to ckan-coord
#2255 1332772340000000 ross > What about ownership by users? Good question. WIll add to description. > > New datasets can be created from the organization form, in which case the organization will be set in the dataset form dropdown (with privacy set to private - see below) > Does this mean like the setup on groups on the datahub (i.e. forward to new dataset but with some things hard-coded, organization/group?)? Or does it mean actual new dataset form on organization page. The former, although currently we will have a slightly different form in the organizations extension, until such a time as we're ready to replace the existing one. Tried merging the two forms but it ended up with such a mess of logic that it was unmanageable.
#2255 1332934041000000 ross Changed description based on docs of stuff to verify.
#2255 1334582784000000 ross Review + restrict resource access (when private)
#2288 1338204815000000 ross Auth groups are being deprecated
#2293 1339062853000000 ross Intercepting the group name change and forcing a reindex of all of the datasets within that group seems like the obvious approach (rather than more lookups by ID after processing search results). Should take less than a day.
#2302 1335516537000000 ross Will eventually want to allow customisation of different CSS but designs are currently under review/change.
#2302 1342097600000000 ross Have assigned it to Toby. Not sure how high priority this is, but only provides a very minimal level of configurability.
#2302 1343820855000000 ross Just a quick note, if Rufus asks about editing about page, he's talking about this ticket.
#2303 1334654727000000 ross feature-2303-rdf-export-command
#2312 1340188741000000 ross We have a ticket for object ownership (group and package) #2548 which will make it much easier to implement this functionality of showing user owned datasets in a 'virtual group'. Will close this in preference to the other ticket which is needed for other features too.
#2319 1340624083000000 ross Resolved with rework of UI
#2322 1337339587000000 ross +1 on these, although ../download instead of ../raw and ../view instead of ../viewer
#2322 1337864855000000 ross Viewer almost exists with the /embed link and so have just used that with a better default size to provide a prettier url. Can still use the width=&height= that are used in the /embed url.
#2322 1339750135000000 ross No, new new features in point releases, we only put bug features in unless there is an extremely pressing need.
#2322 1339750208000000 ross Going to try that last comment again. No, no new features in point releases, we only put bug fixes in unless there is an extremely pressing need.
#2323 1335516967000000 ross Doesn't make sense in Organizations mode. Adding existing dataset would mean that dataset would have to already be in another organization. Leaving this open though to fix in case we decide that datasets *can* be in more than one organization at a time.
#2323 1340188400000000 ross Can create a new ticket if the requirements change.
#2329 1346664271000000 ross Moved to 2.0 branch for new UI.
#2331 1345192722000000 ross This is the dullest game of ticket tennis ever :D Google don't OR the results, they have an explicit operator for ORing query terms and default to ensuring that all terms are present (in one form or another) in the documents. I understand the argument about cut-offs and relevancy scores, but for a large search engine they have the benefit of scale both in resource and documents. I spent 6 months of my life tweaking weightings for n-dimensional vector space search, I don't think it is likely to be a number we can, or will find by accident. Given that Marcus also got confused by the current behaviour I think we should either: a) Make it configurable b) Just change it to what has been asked for Don't care which.
Note: See TracReports for help on using and creating reports.