Implement proxy caching for performance gains

• Analysis of best solution and what we can deliver in a reasonable time [2d]

• Checks for presence of cookie 'auth_tkt' and appropriate setup for media [2d]

• Remove session requirements for user who isn't logged in [3d]

• Document the configuration and setup so that it can be easily modified and replicated [1d]

New form should allow three steps.

1. Add basic fields to create a dataset
2. Add a data resource to the dataset (this can be repeated multiple times using "Save and add another"
3. Add final metadata to the dataset.

It should be possible to jump backward through the steps by clicking the arrows at the top. Errors in fields should be validated for each step when the next button is clicked.

Implementation:

​http://s031.okserver.org:2375/dataset/new

This is happening whenever people try to run paster commands. polib should only be needed for the check-po-files command don't import otherwise.

Review what chunks of code we have and how to move forward with requirements.

Requirements:

• Bulk parsing is done away from the CKAN server

Possible future requirements:

• /importer for importing spreadsheets in WUI
• validation-light - 'scoring' a package

Current importers:

• /importer controller uses: PackageImporter?, form, PackageSaver?
• loader uses: package_importer -> API
• API uses: form, PackageSaver?

Modules

• package_importer:
  • xls/csv --SpreadsheetData?--> table
  • table --DataRecords?--> records
  • records --PackageImporter?--> pkg_dicts

• pkg_dict --form--> fs & validation
• fs --PackageSaver?--> package

Currently stages of the same colour blur together, these could do with a nice divider.

If the notes field could use a WYSIWYG editor with word cleanup this would really help users who may well be pasting in text from Word (and I guess may have been approved or written by someone else)

having auto-complete on tags would both make the system easier to use - and reduce the risk of synonymous tags being created inadvertently.

Does the department drop down options list interact with user permissions - so that users who only have rights to one department only see that option ?

Similarly for licenses - can this reduce to the set of allowed options for this user/

What level of validation is there on the fields?

This needs to be abstracted and we still need to be able to remove items before saving.

... probably by adapting aidtransparency.net

Improve documentation for plugins and add examples. Could use ckanext-example as a base.

Problem: When creating a related item (e.g. a Visualisation), if you don't put in a URL it succeeds, but on the related items and apps pages it renders it as a link to the same page. Expected: Always require a URL and it should only submit if one is added

• Add help text, drawing from:
  • ca.ckan.net
  • ​http://wiki.okfn.org/ckan/doc/package
  • ​http://wiki.okfn.org/ckan/doc/faq
• Make longer text revealed on click

Issue 1. Help command without arguments fails:

$paster help --plugin ckan --config development.ini

gives:

EITHER

Command help not recognized

OR

paster help: error: no such option: --plugin

Issue 2: Harvester command with arguments fails.

$ paster harvester Traceback (most recent call last): ...

cmd = self.args[0]

IndexError?: list index out of range

Placeholder for thinking about this ...

• Allow users to configure info for CSV import (e.g. separators, quotes etc)

Should list resources in the form of search results rather than as a table.

Storage is now working but there are

• Integrate with Resources (e.g. create a resource for each file upload and give option to associate with a package)
  • Should we introduce rule that files *not* associated with a Resource are periodically deleted?
• Allow setting of a file name/path before upload
• Allow for file overwriting/deleting etc (how should this work -- do we want to allow this sort of thing)
• Integrate local file upload stuff in api/auth/*

Different Backend Issues

Local file store is rather different from 'remote' storage in various ways:

• For remote you don't want to use many buckets as there are bucket limits while for local you want to. Should we there have a single path that users provide which we then partition differently for different backends.

• Decent powered by icon for the Embed for the DataHub?? (make this text customizable and allow html ...)
  • Current icon and text looks really bad
• Embed link uses uuid rather than dataset name (this goes to perma-url question ...)
• Data viewer embed url includes all the resource info (not needed -- just url and backend info ...)
• Embed extension urls do not respect original domain but instead use site_url (I'm guessing). E.g. datahub.io domain ends up with links to embed on thedatahub.org

Several things:

• Default package search (used e.g. on front page) should deal well with multiple items: "abc xyz" means search for both abc and xyz (as separate searches)
• Should automatically display a list of matching tags in a line at top with number of associated packages
• have an option to only search for fully open material (or openly licensed material ...)

Since we should be moving to a more sophisticated search solution anyway (to allow prioritisation etc) maybe we should cautious as to what we do now.

• Forgot password (email a new password)
• Confirm email
• Do not show register page if you are logged in (redirect to home page)
• ticket:1010 - listing of users.
  • Do not use /user for general user account home page (either user normal user page /user/{id} or /user/myaccount)

As A

Visitor

I Want To

View lists of items but not have too many on a page (as this leads to slow page loads and difficulty navigating the list). Where there are more items in the results of my action that can be fit on a single page the results should be 'paged' in some manner.

Details

Main place this issue arises:

• List of tags
• List of packages
• Repository history (/revision)

Should probably have no more than 50 (or so) items on a page. Obvious solution is for actions that return lists to have some kind of pageno parameter and then use this in controller to select what to display.

Perhaps using deliverance?

• Move About to the right of the nav tabs and make it an about IATI page
• Make a Help / FAQ section linked of main menu
• Front page more dynamic - latest data registered and news (microblog thing)
• make the front page blurb noticable / shorter etc.
• link to iatistandard.org from the front page blurb

1. One of the datasets here: ​http://s031.okserver.org:2375/en/user/shevski has [DRAFT] after it. It's not clear why this has been added -> user confusion
2. The draft dataset is not clickable in the dataset list
3. The ff dataset (same as the ff [DRAFT] dataset) in the Activity Steam section a) is clickable and b) does not have [DRAFT] after it

When you create a package then the 'location' header gets set. This doesn't happen for any other domain objects. I think this should be consistent - either none or all.

I've removed the info about the header in the docs in the meantime.

Install and document (step-by-step) CKAN (and dependencies, and extensions) onto VM on s031.okserver.org which has CentOS 6.2 installed. Paying particular attention to the fact that we won't have root access on the staging nor production servers, and so a) anything that requires root access must be documented carefully; and b) we want to be able to perform maintenance without being root.

Maintenance

We want to have non-root access to:

• CKAN logs; (relevant) apache logs; qa process' logs; rdf cron logs
• restarting CKAN
• restarting qa processes
• changing/disabling rdf cronjob
• our database (through sql)
• restarting solr

CKAN and Dependencies

• CKAN installed from source
• Required plugins: synchronous_search ecportal ecportal_form organizations multilingual_dataset multilingual_group multilingual_tag qa

Extra configuration

• uploads
• analytics (toby)
• qa
• rdf cron
• caching
• datastore

Things to remember

• Use ckanext/multilingual/solr/schema.xml and stopword files
• Use nginx as proxy.

The CKAN instance on data.gov.uk serves invalid URIs out of its API.

For example the following can be found,

​http://uk.sitestat.com/homeoffice/rds/s?rds.hosb0509tabsxls&ns_type=pdf&ns_url=[http://www.homeoffice.gov.uk/rds/pdfs09/hosb0509tabs.xls]

In this URI, the : and / characters after the ? in the query part are invalid according to section 3.4 of RFC2396

Also URIs are not stripped of whitespace at the end.

This causes problems when other software with a more correct interpretation of what a valid URI is attempts to consume data from CKAN. In this instance the Talis triplestore complains about such URIs.

"Be liberal in what you accept and conservative in what you send" would seem apt.

Actions

• Validation of urls as part of form entry or data loading
  • Need to consider situation where this should happen out-of-band (i.e. we allow load even with invalid data and then flag bad dates in separate validation process). In general doubtful that we should do this here because url invalidity is such a big deal
• This code should support analysis of existing data so we can go through existing database and find invalid urls
  • Also useful to have this so we can do out of band validation

Super ticket: #1602

Steps:

• Include (or git submodule) data explorer into vendor
• Update existing js integration code
• Fix css and any bugs

Est: 1d

Actual: 1d

Create/edit a package -> add a resource -> Oh, i haven't uploaded yet, 'upload here' -> /upload -> standard upload form

Do we need to create a resource out of the file after upload? Do we sync file metadata with local Resource metadata?

If yes suggests (need to ticket):

• 'ResourceExtras?' for extra metadata on resources
• Makes 'Resource' into first class object (and make available in WUI)?

Depends

• #877 (Upload in WUI) - should be part of that extension

1. Package needs a groups property (backref Group.packages).
2. Put groups in the search api.
3. Display them on the package, but not editable. (Only editable from the group page.)

Follow up to ticket:105 (Groups).

Cost: 1d

We have an all new and much improved data preview system developed as part of QA in July. Need to integrate this.

• Is pure javascript
• Will have this in CKAN core but considered to be a plugin (so need datapreview in ckan.plugins list to work -- will enable this by default)
• Will depend on the ​http://wiki.ckan.org/DataExplorer

Is this ready for release? What needs to be done?

There are still some intermittent errors due to objects not having a revision_id, if you save an object at the same time as the revision. We add the uuid earlier to fix this. ​https://bitbucket.org/kindly/vdm/changeset/0050d023ca4e.

From David Raznick

A summary/user story doc of how organizations and groups are expected to work.

• Review process of releases and updating strings (pot) and translations (po) files
  • c.f. ​http://lists.okfn.org/pipermail/ckan-dev/2011-May/000718.html
• Get language changing thing to get home page to mostly work - push to key strings translated (or don't advertise the translation?)
• Review long list of strings for improvement
  • Consolidate any?
  • Get rid of i18n for strings in api
  • Try and remove javascript
  • Add in extensions
• #1374 Fix switching to en from other default language
• #1417 Fix browser language detection

This is a resource that is *just* in the datastore. At the moment there are hacky ways to do this (create resource with meaningless URL etc) but we should do this properly:

• New Option: DataStore? Table in add a resource form
• URL field is disabled (or points at /api/data or something rather pointless)
• Download button is disabled
  • Better would be to find a way to serialize an ES table or stream it ...

The ckan queuing system should provide the option to subscribe to timed re-submissions of specific resources. This could look as follows:

• routing_key: Package
• operation: daily
• payload: pkg.as_dict()

Where operation is one of daily, weekly, monthly or any other interval.

See how possible it is to run CKAN on a Windows server.

Plan:

• Get CKAN code and dependencies installed on Windows
  • ideally as similar to standard package install as possible
  • virtualenv/pip should work
  • try Cygwin to help
  • SOLR and Postgres could potentially run on other servers if necessary.
• Try to run on Apache
• See if there are any issues

Creation of duplicate packages named with trailing underscore

ww 22/9/10:

Need to ... merge some packages (the "find packages by attributes" is not 100% reliable because of inconsistent data already in there).

The "find by tags" algorithm uses (title, department). Unfortunately some of the data already there has no department. This causes it to make a new package (with trailing underscore).

dread 24/9/10:

I've done a test for the missing department issue and I can't reproduce the problem. This may be due to my general tidy up and getting tests passing using the latest ckanclient. Anyway, can you check over my test to see if this is what you mean?: dgu/ckanext/dgu/tests/ons/test_ons_loader.py:TestOnsLoadMissingDept

ww 20/10/10:

you should uncover it if you run against (a copy of) the dgu database.

Here's the dump from 22:10 last night:

URL: ​http://ckan.net/api/search/resource?all_fields=1&offset=0&limit=20&qjson=%3Cspan%20class= Module weberror.errormiddleware:162 in call << traceback_supplement = Supplement, self, environ

sr_checker = ResponseStartChecker?(start_response) app_iter = self.application(environ, sr_checker) return self.make_catching_iter(app_iter, environ, sr_checker)

except:

app_iter = self.application(environ, sr_checker)

Module beaker.middleware:73 in call << self.cache_manager)

environ[self.environ_key] = self.cache_manager return self.app(environ, start_response)

return self.app(environ, start_response)

Module beaker.middleware:152 in call << headers.append(('Set-cookie', cookie))

return start_response(status, headers, exc_info)

return self.wrap_app(environ, session_start_response)

def _get_session(self):

return self.wrap_app(environ, session_start_response)

Module routes.middleware:130 in call << environSCRIPT_NAME? = environSCRIPT_NAME?[:-1]

response = self.app(environ, start_response)

# Wrapped in try as in rare cases the attribute will be gone already

response = self.app(environ, start_response)

Module pylons.wsgiapp:125 in call <<

controller = self.resolve(environ, start_response) response = self.dispatch(controller, environ, start_response)

if 'paste.testing_variables' in environ and hasattr(response,

response = self.dispatch(controller, environ, start_response)

Module pylons.wsgiapp:324 in dispatch << if log_debug:

log.debug("Calling controller class with WSGI interface")

return controller(environ, start_response)

def load_test_env(self, environ):

return controller(environ, start_response)

Module ckan.lib.base:73 in call << # available in environpylons.routes_dict?

try:

return WSGIController.call(self, environ, start_response)

finally:

model.Session.remove()

return WSGIController.call(self, environ, start_response)

Module pylons.controllers.core:221 in call << return response(environ, self.start_response)

response = self._dispatch_call() if not start_response_called:

self.start_response = start_response

response = self._dispatch_call()

Module pylons.controllers.core:172 in _dispatch_call << req.environpylons.action_method? = func

response = self._inspect_call(func)

else:

if log_debug:

response = self._inspect_call(func)

Module pylons.controllers.core:107 in _inspect_call << func.name, args)

try:

result = self._perform_call(func, args)

except HTTPException, httpe:

if log_debug:

result = self._perform_call(func, args)

Module pylons.controllers.core:60 in _perform_call << """Hide the traceback for everything above this method"""

traceback_hide = 'before_and_this' return func(args)

def _inspect_call(self, func):

return func(args)

Module ckan.controllers.rest:400 in search << response.status_int = 400

return gettext('Blank qjson parameter')

params = json.loads(request.paramsqjson?)

elif request.params.values() and request.params.values() != [u] and request.params.values() != [u'1']:

params = request.params

params = json.loads(request.paramsqjson?)

Module simplejson:384 in loads << parse_constant is None and object_pairs_hook is None

and not use_decimal and not kw):

return _default_decoder.decode(s)

if cls is None:

cls = JSONDecoder

return _default_decoder.decode(s)

Module simplejson.decoder:402 in decode << """

obj, end = self.raw_decode(s, idx=_w(s, 0).end()) end = _w(s, end).end() if end != len(s):

obj, end = self.raw_decode(s, idx=_w(s, 0).end())

Module simplejson.decoder:420 in raw_decode << obj, end = self.scan_once(s, idx)

except StopIteration?:

raise JSONDecodeError("No JSON object could be decoded", s, idx)

return obj, end

raise JSONDecodeError("No JSON object could be decoded", s, idx)

JSONDecodeError: No JSON object could be decoded: line 1 column 0 (char 0) CGI Variables DOCUMENT_ROOT '/htdocs' GATEWAY_INTERFACE 'CGI/1.1' HTTP_ACCEPT 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8, application/json' HTTP_ACCEPT_CHARSET 'ISO-8859-1,utf-8;q=0.7,*;q=0.7' HTTP_ACCEPT_ENCODING 'gzip,deflate' HTTP_ACCEPT_LANGUAGE 'en-us,en;q=0.5' HTTP_CONNECTION 'keep-alive' HTTP_COOKIE 'utma=27730403.1245320310.1281386803.1281386803.1282164955.2; utmz=27730403.1282164955.2.2.utmcsr=jira|utmccn=(referral)|utmcmd=referral|utmcct=/browse/PLATFORM-892; utmb=27730403.3.10.1282164955; utmc=27730403' HTTP_HOST 'ckan.net' HTTP_KEEP_ALIVE '300' HTTP_REFERER '​http://jira/browse/PLATFORM-892' HTTP_USER_AGENT 'Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.1.7) Gecko/20091221 Firefox/3.5.7' PATH '/usr/local/bin:/usr/bin:/bin' PATH_INFO '/api/search/resource' PATH_TRANSLATED '/home/okfn/var/srvc/ckan.net/pyenv/bin/ckan.net.py/api/search/resource' QUERY_STRING 'all_fields=1&offset=0&limit=20&qjson=%3Cspan%20class=' REMOTE_ADDR '64.235.97.218' REMOTE_PORT '20720' REQUEST_METHOD 'GET' REQUEST_URI '/api/search/resource?all_fields=1&offset=0&limit=20&qjson=%3Cspan%20class=' SCRIPT_FILENAME '/home/okfn/var/srvc/ckan.net/pyenv/bin/ckan.net.py' SCRIPT_URI '​http://ckan.net/api/search/resource' SCRIPT_URL '/api/search/resource' SERVER_ADDR '10.226.226.118' SERVER_ADMIN '[no address given]' SERVER_NAME 'ckan.net' SERVER_PORT '80' SERVER_PROTOCOL 'HTTP/1.1' SERVER_SIGNATURE '<address>Apache/2.2.9 (Debian) mod_wsgi/2.5 Python/2.5.2 Server at ckan.net Port 80</address>\n' SERVER_SOFTWARE 'Apache/2.2.9 (Debian) mod_wsgi/2.5 Python/2.5.2' WSGI Variables application <beaker.middleware.CacheMiddleware? object at 0xa1c13ec> beaker.cache <beaker.cache.CacheManager? object at 0xa1c142c> beaker.get_session <bound method SessionMiddleware?._get_session of <beaker.middleware.SessionMiddleware? object at 0xa1c12ac>> beaker.session {'_accessed_time': 1282165818.0880959, '_creation_time': 1282165818.0880959} mod_wsgi.application_group 'ckan.net|' mod_wsgi.callable_object 'application' mod_wsgi.listener_host mod_wsgi.listener_port '80' mod_wsgi.process_group mod_wsgi.reload_mechanism '0' mod_wsgi.script_reloading '1' mod_wsgi.version (2, 5) paste.cookies (<SimpleCookie: __utma='27730403.1245320310.1281386803.1281386803.1282164955.2' __utmb='27730403.3.10.1282164955' __utmc='27730403' __utmz='27730403.1282164955.2.2.utmcsr=jira|utmccn=(referral)|utmcmd=referral|utmcct=/browse/PLATFORM-892'>, 'utma=27730403.1245320310.1281386803.1281386803.1282164955.2; utmz=27730403.1282164955.2.2.utmcsr=jira|utmccn=(referral)|utmcmd=referral|utmcct=/browse/PLATFORM-892; utmb=27730403.3.10.1282164955; utmc=27730403') paste.parsed_querystring ([('all_fields', '1'), ('offset', '0'), ('limit', '20'), ('qjson', '<span class=')], 'all_fields=1&offset=0&limit=20&qjson=%3Cspan%20class=') paste.registry <paste.registry.Registry object at 0x130ed84c> paste.throw_errors True pylons.action_method <bound method RestController?.search of <ckan.controllers.rest.RestController? object at 0xe9bbe0c>> pylons.controller <ckan.controllers.rest.RestController? object at 0xe9bbe0c> pylons.environ_config {'session': 'beaker.session', 'cache': 'beaker.cache'} pylons.pylons <pylons.util.PylonsContext? object at 0xe9bbe8c> pylons.routes_dict {'action': u'search', 'controller': u'rest', 'register': u'resource'} repoze.who.logger <logging.Logger instance at 0xa3cb0cc> repoze.who.plugins {'openid': <OpenIdIdentificationPlugin? 170067148>, 'auth_tkt': <AuthTktCookiePlugin? 171739788>} routes.route <routes.route.Route object at 0xa102fac> routes.url <routes.util.URLGenerator object at 0x13a5a3cc> webob._parsed_query_vars (GET([('all_fields', '1'), ('offset', '0'), ('limit', '20'), ('qjson', '<span class=')]), 'all_fields=1&offset=0&limit=20&qjson=%3Cspan%20class=') webob.adhoc_attrs {'language': 'en-us'} wsgi process 'Multi process AND threads (?)' wsgi.file_wrapper <built-in method file_wrapper of mod_wsgi.Adapter object at 0x12f53530> wsgiorg.routing_args (<routes.util.URLGenerator object at 0x13a5a3cc>, {'action': u'search', 'controller': u'rest', 'register': u'resource'})

Edge-side includes or partial page caching are a standard way to deal with caching of pages in which some (usually small) part of the content cannot be cached or should be cached in a different manner (e.g. much more briefly) than the rest of the page.

Edge-side includes have the advantage that they integrate with general 3rd-party caching systems such as varnish.

Introducing either partial page or ESI will require some overhaul work to change the page render processing somewhat.

Tom Kralidis kindly told us on IRC [1] that the latest trunk of pycsw includes support for wsgi and different backends, which sounds like could be really useful for CKAN integration.

[1] ​http://irclogs.okfn.org/ckan/2012-04-25.txt 17:16

Work already here: ​http://knowledgeforge.net/ckan/trac/wiki/SearchEngine

Add knowledge there or to:

​http://wiki.okfn.org/SoftwareTools/Search

Do we want to build on top of the existing debian packaging code? Or throw it away and start fresh?

Revision.packages should not include all changes so why are we getting revisions with no associated changes? See e.g. ​http://ckan.net/revision/read/814e39e1-2540-4449-90d1-ac9a92ab40fb?

Investigate and fix.

NB: if this is bugfix this should go against stable or metastable branch if at all possible and then be merged into default.

Cost: 1.5h (?)

Need to investigate alternative payment gateways

Some extensions import pylons features (such as request) at root.

Have encountered following issues:

1. You need to import ckanext before any ckan stuff in your wsgi script (extension stats and dataapi)

2. Running under wsgi daemon mode (normal mode is fine!) the front page (front page only) will not work (does not occur with disqus or synchronous_search but does with stats and dataapi extensions). You get errors like this in the log:

[Mon Jan 17 09:01:19 2011] [error] Exception KeyError: KeyError(-1218594160,) in <module 'threading' from '/usr/lib/python2.6/threading.pyc'> ignored
[Mon Jan 17 09:01:20 2011] [error] Exception AttributeError: "'NoneType' object has no attribute 'clearing'" in <bound method PluginEnvironment.__del__ of  Services for Environment '<default>'

Consolidate js. Remove cruft.

• Move stuff to vendor
  • Have all vendor libs local (?)
  • jquery, jquery-ui etc
• Consolidate autocomplete to use jquery-ui
  • package autocomplete (groups)
  • tag autocomplete (package)
  • user autocomplete (authz group create)
    • turns out that authz group edit had autocomplete disabled so refactoring meant enabling ...
• Move forms.css into main
• Remove tag cloud from front page (can be example going forward)
• Remove bookmarklet (since in wiki at: ​http://wiki.ckan.net/Managing_Data_Packages#How_Do_I_Install_the_CKAN_Bookmarklet.3F)
• Move name slug generation into standard js and clean up

Possible:

1. JS unit tests for the given pages ...
2. Option to use external site to serve js and css (e.g. CDN)

It is possible to use the CKAN API to insert JSON format data into package extra values, but this data is displayed in the package edit form.

Example: Package ​http://ckan.net/package/hbz_unioncatalog in the API the extra value as a list:

"extras": {"publishingInstitution": ["http://lobid.org/organisation/DE-605", "http://lobid.org/organisation/DE-290"...

yet when you edit the package it loses all the quotes and brackets: ​http://ckan.net/package/edit/hbz_unioncatalog {{{​http://lobid.org/organisation/DE-605http://lobid.org/organisation/DE-290... }}} so when you save the package, the list is mangled into a bad string.