def member(self, dataset, dimension, name, format="html"):
self._get_member(dataset, dimension, name)
handle_request(request, c, c.member, c.dimension.name)
member = [member_apply_links(dataset, dimension, c.member)]
if format == 'json':
return write_json(member, response)
elif format == 'csv':
return write_csv(member, response)
else:
# If there are no views set up, then go direct to the entries
# search page
if c.view is None:
return redirect(
url_for(controller='dimension',
action='entries',
dataset=c.dataset.name,
dimension=dimension,
name=name))
if 'embed' in request.params:
return redirect(
url_for(controller='view',
action='embed',
dataset=c.dataset.name,
widget=c.view.vis_widget.get('name'),
state=json.dumps(c.view.vis_state)))
return templating.render('dimension/member.html')
def index(self, format='html'):
for item in self.extensions:
item.index(c, request, response, c.results)
if format == 'json':
return to_jsonp(map(lambda d: d.as_dict(),
c.datasets))
elif format == 'csv':
results = map(lambda d: d.as_dict(), c.datasets)
return write_csv(results, response)
else:
c.query = request.params.items()
c.add_filter = lambda f, v: '?' + urlencode(c.query +
[(f, v)] if (f, v) not in c.query else c.query)
c.del_filter = lambda f, v: '?' + urlencode([(k,x) for k, x in
c.query if (k,x) != (f,v)])
c.results = c.datasets
for language in request.params.getall('languages'):
l = db.aliased(DatasetLanguage)
c.results = c.results.join(l, Dataset._languages)
c.results = c.results.filter(l.code==language)
for territory in request.params.getall('territories'):
t = db.aliased(DatasetTerritory)
c.results = c.results.join(t, Dataset._territories)
c.results = c.results.filter(t.code==territory)
c.results = list(c.results)
c.territory_options = DatasetTerritory.dataset_counts(c.results)
c.language_options = DatasetLanguage.dataset_counts(c.results)
return render('dataset/index.html')
def member(self, dataset, dimension, name, format="html"):
self._get_member(dataset, dimension, name)
handle_request(request, c, c.member, c.dimension.name)
member = [member_apply_links(dataset, dimension, c.member)]
if format == "json":
return write_json(member, response)
elif format == "csv":
return write_csv(member, response)
else:
# If there are no views set up, then go direct to the entries
# search page
if c.view is None:
return redirect(
url_for(
controller="dimension", action="entries", dataset=c.dataset.name, dimension=dimension, name=name
)
)
if "embed" in request.params:
return redirect(
url_for(
controller="view",
action="embed",
dataset=c.dataset.name,
widget=c.view.vis_widget.get("name"),
state=json.dumps(c.view.vis_state),
)
)
return templating.render("dimension/member.html")
def view(self, dataset, id, format='html'):
self._get_dataset(dataset)
entries = list(c.dataset.entries(c.dataset.alias.c.id == id))
if not len(entries) == 1:
abort(404, _('Sorry, there is no entry %r') % id)
c.entry = entry_apply_links(dataset, entries.pop())
c.id = c.entry.get('id')
c.from_ = c.entry.get('from')
c.to = c.entry.get('to')
c.currency = c.entry.get('currency', c.dataset.currency).upper()
c.amount = c.entry.get('amount')
c.time = c.entry.get('time')
c.custom_html = h.render_entry_custom_html(c.dataset, c.entry)
excluded_keys = ('time', 'amount', 'currency', 'from', 'to', 'dataset',
'id', 'name', 'description')
c.extras = {}
if c.dataset:
c.desc = dict([(d.name, d) for d in c.dataset.dimensions])
for key in c.entry:
if key in c.desc and \
not key in excluded_keys:
c.extras[key] = c.entry[key]
if format == 'json':
return to_jsonp(c.entry)
elif format == 'csv':
return write_csv([c.entry], response)
else:
return templating.render('entry/view.html')
def index(self, format='html'):
for item in self.extensions:
item.index(c, request, response, c.results)
if format == 'json':
return to_jsonp(map(lambda d: d.as_dict(), c.datasets))
elif format == 'csv':
results = map(lambda d: d.as_dict(), c.datasets)
return write_csv(results, response)
else:
c.query = request.params.items()
c.add_filter = lambda f, v: '?' + urlencode(c.query + [(f, v)] if (
f, v) not in c.query else c.query)
c.del_filter = lambda f, v: '?' + urlencode([(k, x)
for k, x in c.query
if (k, x) != (f, v)])
c.results = c.datasets
for language in request.params.getall('languages'):
l = db.aliased(DatasetLanguage)
c.results = c.results.join(l, Dataset._languages)
c.results = c.results.filter(l.code == language)
for territory in request.params.getall('territories'):
t = db.aliased(DatasetTerritory)
c.results = c.results.join(t, Dataset._territories)
c.results = c.results.filter(t.code == territory)
c.results = list(c.results)
c.territory_options = DatasetTerritory.dataset_counts(c.results)
c.language_options = DatasetLanguage.dataset_counts(c.results)
return render('dataset/index.html')
def view(self, dataset, id, format="html"):
self._get_dataset(dataset)
entries = list(c.dataset.entries(c.dataset.alias.c.id == id))
if not len(entries) == 1:
abort(404, _("Sorry, there is no entry %r") % id)
c.entry = entry_apply_links(dataset, entries.pop())
c.id = c.entry.get("id")
c.from_ = c.entry.get("from")
c.to = c.entry.get("to")
c.currency = c.entry.get("currency", c.dataset.currency).upper()
c.amount = c.entry.get("amount")
c.time = c.entry.get("time")
c.custom_html = h.render_entry_custom_html(c.dataset, c.entry)
excluded_keys = ("time", "amount", "currency", "from", "to", "dataset", "id", "name", "description")
c.extras = {}
if c.dataset:
c.desc = dict([(d.name, d) for d in c.dataset.dimensions])
for key in c.entry:
if key in c.desc and not key in excluded_keys:
c.extras[key] = c.entry[key]
if format == "json":
return to_jsonp(c.entry)
elif format == "csv":
return write_csv([c.entry], response)
else:
return render("entry/view.html")
def aggregate(self):
parser = AggregateParamParser(request.params)
params, errors = parser.parse()
if errors:
response.status = 400
return {'errors': errors}
params['cuts'] = params.pop('cut')
params['drilldowns'] = params.pop('drilldown')
dataset = params.pop('dataset')
format = params.pop('format')
require.dataset.read(dataset)
try:
cache = AggregationCache(dataset)
result = cache.aggregate(**params)
if 'drilldown' in result:
result['drilldown'] = drilldowns_apply_links(
dataset.name, result['drilldown'])
response.last_modified = dataset.updated_at
if cache.cache_enabled and 'cache_key' in result['summary']:
etag_cache(result['summary']['cache_key'])
except (KeyError, ValueError) as ve:
log.exception(ve)
response.status = 400
return {'errors': ['Invalid aggregation query: %r' % ve]}
if format == 'csv':
return write_csv(result['drilldown'],
response,
filename=dataset.name + '.csv')
return to_jsonp(result)
def view(self, dataset, id, format='html'):
self._get_dataset(dataset)
entries = list(c.dataset.entries(c.dataset.alias.c.id == id))
if not len(entries) == 1:
abort(404, _('Sorry, there is no entry %r') % id)
c.entry = entry_apply_links(dataset, entries.pop())
c.id = c.entry.get('id')
c.from_ = c.entry.get('from')
c.to = c.entry.get('to')
c.currency = c.entry.get('currency', c.dataset.currency).upper()
c.amount = c.entry.get('amount')
c.time = c.entry.get('time')
c.custom_html = h.render_entry_custom_html(c.dataset,
c.entry)
excluded_keys = ('time', 'amount', 'currency', 'from',
'to', 'dataset', 'id', 'name', 'description')
c.extras = {}
if c.dataset:
c.desc = dict([(d.name, d) for d in c.dataset.dimensions])
for key in c.entry:
if key in c.desc and \
not key in excluded_keys:
c.extras[key] = c.entry[key]
if format == 'json':
return to_jsonp(c.entry)
elif format == 'csv':
return write_csv([c.entry], response)
else:
return render('entry/view.html')
def aggregate(self):
parser = AggregateParamParser(request.params)
params, errors = parser.parse()
if errors:
response.status = 400
return {'errors': errors}
params['cuts'] = params.pop('cut')
params['drilldowns'] = params.pop('drilldown')
dataset = params.pop('dataset')
format = params.pop('format')
require.dataset.read(dataset)
try:
cache = AggregationCache(dataset)
result = cache.aggregate(**params)
if 'drilldown' in result:
result['drilldown'] = drilldowns_apply_links(dataset.name,
result['drilldown'])
response.last_modified = dataset.updated_at
if cache.cache_enabled and 'cache_key' in result['summary']:
etag_cache(result['summary']['cache_key'])
except (KeyError, ValueError) as ve:
log.exception(ve)
response.status = 400
return {'errors': ['Invalid aggregation query: %r' % ve]}
if format == 'csv':
return write_csv(result['drilldown'], response,
filename=dataset.name + '.csv')
return to_jsonp(result)
def index_export(self, dataset, format):
self._get_dataset(dataset)
# TODO include html_urls in dumps.
#processor = lambda e: entry_apply_links(c.dataset.name, e)
if format == 'json':
return write_json(c.dataset.entries(), response,
filename=c.dataset.name + '.json')
if format == 'csv':
return write_csv(c.dataset.entries(), response,
filename=c.dataset.name + '.csv')
else:
return redirect(h.url_for(controller='entry', action='index'))
def entries(self, dataset, dimension, name, format="html"):
self._get_member(dataset, dimension, name)
handle_request(request, c, c.member, c.dimension.name)
entries = c.dataset.entries(c.dimension.alias.c.name == c.member["name"])
entries = (entry_apply_links(dataset, e) for e in entries)
attachment_name = "__".join([dataset, dimension, name])
if format == "json":
return write_json(entries, response, filename=attachment_name + ".json")
elif format == "csv":
return write_csv(entries, response, filename=attachment_name + ".csv")
else:
return render("dimension/entries.html")
def entries(self, dataset, dimension, name, format='html'):
self._get_member(dataset, dimension, name)
handle_request(request, c, c.member, c.dimension.name)
entries = c.dataset.entries(c.dimension.alias.c.name == c.member['name'])
entries = (entry_apply_links(dataset, e) for e in entries)
attachment_name = '__'.join([dataset, dimension, name])
if format == 'json':
return write_json(entries, response, filename=attachment_name + '.json')
elif format == 'csv':
return write_csv(entries, response, filename=attachment_name + '.csv')
else:
return render('dimension/entries.html')
def member(self, dataset, dimension, name, format="html"):
self._get_member(dataset, dimension, name)
handle_request(request, c, c.member, c.dimension)
if c.view is None:
self._make_browser()
for item in self.extensions:
item.read(c, request, response, c.member)
if format == 'json':
return to_jsonp(c.member)
elif format == 'csv':
return write_csv([c.member], response)
else:
return render('dimension/member.html')
def index_export(self, dataset, format):
self._get_dataset(dataset)
# TODO include html_urls in dumps.
#processor = lambda e: entry_apply_links(c.dataset.name, e)
if format == 'json':
return write_json(c.dataset.entries(),
response,
filename=c.dataset.name + '.json')
if format == 'csv':
return write_csv(c.dataset.entries(),
response,
filename=c.dataset.name + '.csv')
else:
return redirect(h.url_for(controller='entry', action='index'))
def member(self, dataset, dimension, name, format="html"):
self._get_member(dataset, dimension, name)
handle_request(request, c, c.member, c.dimension)
if c.view is None:
self._make_browser()
for item in self.extensions:
item.read(c, request, response, c.member)
if format == 'json':
return to_jsonp(c.member)
elif format == 'csv':
return write_csv([c.member], response)
else:
return render('dimension/member.html')
def member(self, dataset, dimension, name, format="html"):
self._get_member(dataset, dimension, name)
handle_request(request, c, c.member, c.dimension.name)
# If there are no views set up, then go direct to the entries search page
if c.view is None and format is "html":
return redirect(url_for(controller='dimension', action='entries',
dataset=c.dataset.name, dimension=dimension, name=name))
member = [member_apply_links(dataset, dimension, c.member)]
if format == 'json':
return write_json(member, response)
elif format == 'csv':
return write_csv(member, response)
else:
return render('dimension/member.html')
def view(self, dataset, format='html'):
self._get_dataset(dataset)
c.num_entries = len(c.dataset)
handle_request(request, c, c.dataset)
if c.view is None and format == 'html':
return EntryController().index(dataset, format)
for item in self.extensions:
item.read(c, request, response, c.dataset)
if format == 'json':
return to_jsonp(c.dataset.as_dict())
elif format == 'csv':
return write_csv([c.dataset.as_dict()], response)
else:
return render('dataset/view.html')
def view(self, dataset, format='html'):
self._get_dataset(dataset)
c.num_entries = len(c.dataset)
handle_request(request, c, c.dataset)
if c.view is None and format == 'html':
return EntryController().index(dataset, format)
for item in self.extensions:
item.read(c, request, response, c.dataset)
if format == 'json':
return to_jsonp(c.dataset.as_dict())
elif format == 'csv':
return write_csv([c.dataset.as_dict()], response)
else:
return render('dataset/view.html')
def entries(self, dataset, dimension, name, format='html'):
self._get_member(dataset, dimension, name)
handle_request(request, c, c.member, c.dimension.name)
entries = c.dataset.entries(
c.dimension.alias.c.name == c.member['name'])
entries = (entry_apply_links(dataset, e) for e in entries)
attachment_name = '__'.join([dataset, dimension, name])
if format == 'json':
return write_json(entries,
response,
filename=attachment_name + '.json')
elif format == 'csv':
return write_csv(entries,
response,
filename=attachment_name + '.csv')
else:
return render('dimension/entries.html')
def member(self, dataset, dimension, name, format="html"):
self._get_member(dataset, dimension, name)
handle_request(request, c, c.member, c.dimension.name)
member = [member_apply_links(dataset, dimension, c.member)]
if format == 'json':
return write_json(member, response)
elif format == 'csv':
return write_csv(member, response)
else:
# If there are no views set up, then go direct to the entries
# search page
if c.view is None:
return redirect(url_for(controller='dimension', action='entries',
dataset=c.dataset.name, dimension=dimension, name=name))
if 'embed' in request.params:
return redirect(url_for(controller='view',
action='embed', dataset=c.dataset.name,
widget=c.view.vis_widget.get('name'),
state=json.dumps(c.view.vis_state)))
return render('dimension/member.html')
def index(self, format='html'):
c.query = request.params.items()
c.add_filter = lambda f, v: '?' + urlencode(c.query +
[(f, v)] if (f, v) not in c.query else c.query)
c.del_filter = lambda f, v: '?' + urlencode([(k, x) for k, x in
c.query if (k, x) != (f, v)])
c.results = c.datasets
for language in request.params.getall('languages'):
l = db.aliased(DatasetLanguage)
c.results = c.results.join(l, Dataset._languages)
c.results = c.results.filter(l.code == language)
for territory in request.params.getall('territories'):
t = db.aliased(DatasetTerritory)
c.results = c.results.join(t, Dataset._territories)
c.results = c.results.filter(t.code == territory)
c.results = list(c.results)
c.territory_options = [{'code': code,
'count': count,
'url': h.url_for(controller='dataset',
action='index', territories=code),
'label': COUNTRIES.get(code, code)} \
for (code, count) in DatasetTerritory.dataset_counts(c.results)]
c.language_options = [{'code': code,
'count': count,
'url': h.url_for(controller='dataset',
action='index', languages=code),
'label': LANGUAGES.get(code, code)} \
for (code, count) in DatasetLanguage.dataset_counts(c.results)]
if format == 'json':
results = map(lambda d: d.as_dict(), c.results)
results = [dataset_apply_links(r) for r in results]
return to_jsonp({
'datasets': results,
'territories': c.territory_options,
'languages': c.language_options
})
elif format == 'csv':
results = map(lambda d: d.as_dict(), c.results)
return write_csv(results, response)
return render('dataset/index.html')
def _view_csv(self, result):
if not isinstance(result, list):
result = [result]
write_csv(result, response)
return
def to_csv(self):
return write_csv(self.all_entries, response)
def to_csv(self):
from pylons import response
return write_csv(self.entities, response)
def aggregate(self):
"""
Aggregation of a dataset based on URL parameters. It serves the
aggregation from a cache if possible, and if not it computes it (it's
performed in the aggregation cache for some reason).
"""
# Parse the aggregation parameters to get them into the right format
parser = AggregateParamParser(request.params)
params, errors = parser.parse()
# If there were parsing errors we return them with status code 400
# as jsonp, irrespective of what format was asked for.
if errors:
response.status = 400
return to_jsonp({'errors': errors})
# URL parameters are always singular nouns but we work with some
# as plural nouns so we pop them into the plural version
params['cuts'] = params.pop('cut')
params['drilldowns'] = params.pop('drilldown')
params['measures'] = params.pop('measure')
# Get the dataset and the format and remove from the parameters
dataset = params.pop('dataset')
format = params.pop('format')
# User must have the right to read the dataset to perform aggregation
require.dataset.read(dataset)
# Create response headers from the parameters
self._response_params(params)
try:
# Create an aggregation cache for the dataset and aggregate its
# results. The cache will perform the aggreagation if it doesn't
# have a cached result
cache = AggregationCache(dataset)
result = cache.aggregate(**params)
# If the result has drilldown we create html_url values for its
# dimensions (linked data).
if 'drilldown' in result:
result['drilldown'] = drilldowns_apply_links(dataset.name,
result['drilldown'])
# Do the ETag caching based on the cache_key in the summary
# this is a weird place to do it since the heavy lifting has
# already been performed above. TODO: Needs rethinking.
response.last_modified = dataset.updated_at
if cache.cache_enabled and 'cache_key' in result['summary']:
etag_cache(result['summary']['cache_key'])
except (KeyError, ValueError) as ve:
# We log possible errors and return them with status code 400
log.exception(ve)
response.status = 400
return to_jsonp({'errors': [unicode(ve)]})
# If the requested format is csv we write the drilldown results into
# a csv file and return it, if not we return a jsonp result (default)
if format == 'csv':
return write_csv(result['drilldown'], response,
filename=dataset.name + '.csv')
return to_jsonp(result)
def index(self, format='html'):
"""
Get a list of all datasets along with territory, language, and
category counts (amount of datasets for each).
"""
# Create facet filters (so we can look at a single country,
# language etc.)
c.query = request.params.items()
c.add_filter = lambda f, v: \
'?' + urlencode(c.query +
[(f, v)] if (f, v) not in c.query else c.query)
c.del_filter = lambda f, v: \
'?' + urlencode([(k, x) for k, x in
c.query if (k, x) != (f, v)])
# Parse the request parameters to get them into the right format
parser = DatasetIndexParamParser(request.params)
params, errors = parser.parse()
if errors:
concatenated_errors = ', '.join(errors)
abort(400,
_('Parameter values not supported: %s') %
concatenated_errors)
# We need to pop the page and pagesize parameters since they're not
# used for the cache (we have to get all of the datasets to do the
# language, territory, and category counts (these are then only used
# for the html response)
params.pop('page')
pagesize = params.pop('pagesize')
# Get cached indices (this will also generate them if there are no
# cached results (the cache is invalidated when a dataset is published
# or retracted
cache = DatasetIndexCache()
results = cache.index(**params)
# Generate the ETag from the last modified timestamp of the first
# dataset (since they are ordered in descending order by last
# modified). It doesn't matter that this happens if it has (possibly)
# generated the index (if not cached) since if it isn't cached then
# the ETag is definitely modified. We wrap it in a try clause since
# if there are no public datasets we'll get an index error.
# We also don't set c._must_revalidate to True since we don't care
# if the index needs a hard refresh
try:
etag_cache_keygen(
results['datasets'][0]['timestamps']['last_modified'])
except IndexError:
etag_cache_keygen(None)
# Assign the results to template context variables
c.language_options = results['languages']
c.territory_options = results['territories']
c.category_options = results['categories']
if format == 'json':
# Apply links to the dataset lists before returning the json
results['datasets'] = [dataset_apply_links(r)
for r in results['datasets']]
return to_jsonp(results)
elif format == 'csv':
# The CSV response only shows datasets, not languages,
# territories, etc.
return write_csv(results['datasets'], response)
# If we're here then it's an html format so we show rss, do the
# pagination and render the template
c.show_rss = True
# The page parameter we popped earlier is part of request.params but
# we now know it was parsed. We have to send in request.params to
# retain any parameters already supplied (filters)
c.page = templating.Page(results['datasets'], items_per_page=pagesize,
item_count=len(results['datasets']),
**request.params)
return templating.render('dataset/index.html')
def index(self, format="html"):
c.query = request.params.items()
c.add_filter = lambda f, v: "?" + urlencode(c.query + [(f, v)] if (f, v) not in c.query else c.query)
c.del_filter = lambda f, v: "?" + urlencode([(k, x) for k, x in c.query if (k, x) != (f, v)])
c.results = c.datasets
for language in request.params.getall("languages"):
l = db.aliased(DatasetLanguage)
c.results = c.results.join(l, Dataset._languages)
c.results = c.results.filter(l.code == language)
for territory in request.params.getall("territories"):
t = db.aliased(DatasetTerritory)
c.results = c.results.join(t, Dataset._territories)
c.results = c.results.filter(t.code == territory)
category = request.params.get("category")
if category:
c.results = c.results.filter(Dataset.category == category)
c.results = list(c.results)
c.territory_options = [
{
"code": code,
"count": count,
"url": h.url_for(controller="dataset", action="index", territories=code),
"label": COUNTRIES.get(code, code),
}
for (code, count) in DatasetTerritory.dataset_counts(c.results)
]
c.language_options = [
{
"code": code,
"count": count,
"url": h.url_for(controller="dataset", action="index", languages=code),
"label": LANGUAGES.get(code, code),
}
for (code, count) in DatasetLanguage.dataset_counts(c.results)
]
# TODO: figure out where to put this:
ds_ids = [d.id for d in c.results]
if len(ds_ids):
q = db.select(
[Dataset.category, db.func.count(Dataset.id)],
Dataset.id.in_(ds_ids),
group_by=Dataset.category,
order_by=db.func.count(Dataset.id).desc(),
)
c.category_options = [
{
"category": category,
"count": count,
"url": h.url_for(controller="dataset", action="index", category=category),
"label": CATEGORIES.get(category, category),
}
for (category, count) in db.session.bind.execute(q).fetchall()
if category is not None
]
else:
c.category_options = []
c._must_revalidate = True
if len(c.results):
dt = max([r.updated_at for r in c.results])
etag_cache_keygen(dt)
if format == "json":
results = map(lambda d: d.as_dict(), c.results)
results = [dataset_apply_links(r) for r in results]
return to_jsonp(
{
"datasets": results,
"categories": c.category_options,
"territories": c.territory_options,
"languages": c.language_options,
}
)
elif format == "csv":
results = map(lambda d: d.as_dict(), c.results)
return write_csv(results, response)
return render("dataset/index.html")
def view(self, dataset, id, format='html'):
"""
Get a specific entry in the dataset, identified by the id. Entry
can be return as html (default), json or csv.
"""
# Generate the dataset
self._get_dataset(dataset)
# Get the entry that matches the given id. c.dataset.entries is
# a generator so we create a list from it's responses based on the
# given constraint
entries = list(c.dataset.entries(c.dataset.alias.c.id == id))
# Since we're trying to get a single entry the list should only
# contain one entry, if not then we return an error
if not len(entries) == 1:
abort(404, _('Sorry, there is no entry %r') % id)
# Add urls to the dataset and assign assign it as a context variable
c.entry = entry_apply_links(dataset, entries.pop())
# Get and set some context variables from the entry
# This shouldn't really be necessary but it's here so nothing gets
# broken
c.id = c.entry.get('id')
c.from_ = c.entry.get('from')
c.to = c.entry.get('to')
c.currency = c.entry.get('currency', c.dataset.currency).upper()
c.time = c.entry.get('time')
# Get the amount for the entry
amount = c.entry.get('amount')
# We adjust for inflation if the user as asked for this to be inflated
if request.params.has_key('inflate'):
try:
# Inflate the amount. Target date is provided in request.params
# as value for inflate and reference date is the date of the
# entry. We also provide a list of the territories to extract
# a single country for which to do the inflation
c.inflation = h.inflate(amount, request.params['inflate'],
c.time, c.dataset.territories)
# The amount to show should be the inflated amount
# and overwrite the entry's amount as well
c.amount = c.inflation['inflated']
c.entry['amount'] = c.inflation['inflated']
# We include the inflation response in the entry's dict
# HTML description assumes every dict value for the entry
# includes a label so we include a default "Inflation
# adjustment" for it to work.
c.inflation['label'] = 'Inflation adjustment'
c.entry['inflation_adjustment'] = c.inflation
except:
# If anything goes wrong in the try clause (and there's a lot
# that can go wrong). We just say that we can't adjust for
# inflation and set the context amount as the original amount
h.flash_notice(_('Unable to adjust for inflation'))
c.amount = amount
else:
# If we haven't been asked to inflate then we just use the
# original amount
c.amount = amount
# Add custom html for the dataset entry if the dataset has some
# custom html
# 2013-11-17 disabled this as part of removal of genshi as depended on
# a genshi specific helper.
# TODO: reinstate if important
# c.custom_html = h.render_entry_custom_html(c.dataset, c.entry)
# Add the rest of the dimensions relating to this entry into a
# extras dictionary. We first need to exclude all dimensions that
# are already shown and then we can loop through the dimensions
excluded_keys = ('time', 'amount', 'currency', 'from',
'to', 'dataset', 'id', 'name', 'description')
c.extras = {}
if c.dataset:
# Create a dictionary of the dataset dimensions
c.desc = dict([(d.name, d) for d in c.dataset.dimensions])
# Loop through dimensions of the entry
for key in c.entry:
# Entry dimension must be a dataset dimension and not in
# the predefined excluded keys
if key in c.desc and \
not key in excluded_keys:
c.extras[key] = c.entry[key]
# Return entry based on
if format == 'json':
return to_jsonp(c.entry)
elif format == 'csv':
return write_csv([c.entry], response)
else:
return templating.render('entry/view.html')
def search(self):
parser = SearchParamParser(request.params)
params, errors = parser.parse()
if errors:
response.status = 400
return to_jsonp({'errors': errors})
expand_facets = params.pop('expand_facet_dimensions')
format = params.pop('format')
if format == 'csv':
params['stats'] = False
params['facet_field'] = None
datasets = params.pop('dataset', None)
if datasets is None or not datasets:
q = Dataset.all_by_account(c.account)
if params.get('category'):
q = q.filter_by(category=params.pop('category'))
datasets = q.all()
expand_facets = False
if not datasets:
return {'errors': ["No dataset available."]}
params['filter']['dataset'] = []
for dataset in datasets:
require.dataset.read(dataset)
params['filter']['dataset'].append(dataset.name)
response.last_modified = max([d.updated_at for d in datasets])
etag_cache_keygen(parser.key(), response.last_modified)
if params['pagesize'] > parser.defaults['pagesize']:
# http://wiki.nginx.org/X-accel#X-Accel-Buffering
response.headers['X-Accel-Buffering'] = 'no'
if format == 'csv':
csv_headers(response, 'entries.csv')
streamer = CSVStreamingResponse(
datasets,
params,
pagesize=parser.defaults['pagesize']
)
return streamer.response()
else:
json_headers(filename='entries.json')
streamer = JSONStreamingResponse(
datasets,
params,
pagesize=parser.defaults['pagesize'],
expand_facets=util.expand_facets
if expand_facets else None,
callback=request.params.get('callback')
)
return streamer.response()
solr_browser = Browser(**params)
try:
solr_browser.execute()
except SolrException as e:
return {'errors': [unicode(e)]}
entries = []
for dataset, entry in solr_browser.get_entries():
entry = entry_apply_links(dataset.name, entry)
entry['dataset'] = dataset_apply_links(dataset.as_dict())
entries.append(entry)
if format == 'csv':
return write_csv(entries, response,
filename='entries.csv')
if expand_facets and len(datasets) == 1:
facets = solr_browser.get_expanded_facets(datasets[0])
else:
facets = solr_browser.get_facets()
return to_jsonp({
'stats': solr_browser.get_stats(),
'facets': facets,
'results': entries
})
def index(self, format='html'):
c.query = request.params.items()
c.add_filter = lambda f, v: '?' + urlencode(c.query + [(f, v)] if (
f, v) not in c.query else c.query)
c.del_filter = lambda f, v: '?' + urlencode([(k, x) for k, x in c.query
if (k, x) != (f, v)])
c.results = c.datasets
for language in request.params.getall('languages'):
l = db.aliased(DatasetLanguage)
c.results = c.results.join(l, Dataset._languages)
c.results = c.results.filter(l.code == language)
for territory in request.params.getall('territories'):
t = db.aliased(DatasetTerritory)
c.results = c.results.join(t, Dataset._territories)
c.results = c.results.filter(t.code == territory)
category = request.params.get('category')
if category:
c.results = c.results.filter(Dataset.category == category)
c.results = list(c.results)
c.territory_options = [{'code': code,
'count': count,
'url': h.url_for(controller='dataset',
action='index', territories=code),
'label': COUNTRIES.get(code, code)} \
for (code, count) in DatasetTerritory.dataset_counts(c.results)]
c.language_options = [{'code': code,
'count': count,
'url': h.url_for(controller='dataset',
action='index', languages=code),
'label': LANGUAGES.get(code, code)} \
for (code, count) in DatasetLanguage.dataset_counts(c.results)]
# TODO: figure out where to put this:
ds_ids = [d.id for d in c.results]
if len(ds_ids):
q = db.select(
[Dataset.category, db.func.count(Dataset.id)],
Dataset.id.in_(ds_ids),
group_by=Dataset.category,
order_by=db.func.count(Dataset.id).desc())
c.category_options = [{'category': category,
'count': count,
'url': h.url_for(controller='dataset',
action='index', category=category),
'label': CATEGORIES.get(category, category)} \
for (category, count) in db.session.bind.execute(q).fetchall() \
if category is not None]
else:
c.category_options = []
c._must_revalidate = True
if len(c.results):
dt = max([r.updated_at for r in c.results])
etag_cache_keygen(dt)
if format == 'json':
results = map(lambda d: d.as_dict(), c.results)
results = [dataset_apply_links(r) for r in results]
return to_jsonp({
'datasets': results,
'categories': c.category_options,
'territories': c.territory_options,
'languages': c.language_options
})
elif format == 'csv':
results = map(lambda d: d.as_dict(), c.results)
return write_csv(results, response)
c.show_rss = True
return templating.render('dataset/index.html')
def search(self):
parser = SearchParamParser(request.params)
params, errors = parser.parse()
if errors:
response.status = 400
return to_jsonp({'errors': errors})
expand_facets = params.pop('expand_facet_dimensions')
format = params.pop('format')
if format == 'csv':
params['stats'] = False
params['facet_field'] = None
datasets = params.pop('dataset', None)
if datasets is None or not datasets:
q = Dataset.all_by_account(c.account)
if params.get('category'):
q = q.filter_by(category=params.pop('category'))
datasets = q.all()
expand_facets = False
if not datasets:
return {'errors': ["No dataset available."]}
params['filter']['dataset'] = []
for dataset in datasets:
require.dataset.read(dataset)
params['filter']['dataset'].append(dataset.name)
response.last_modified = max([d.updated_at for d in datasets])
etag_cache_keygen(parser.key(), response.last_modified)
if params['pagesize'] > parser.defaults['pagesize']:
# http://wiki.nginx.org/X-accel#X-Accel-Buffering
response.headers['X-Accel-Buffering'] = 'no'
if format == 'csv':
csv_headers(response, 'entries.csv')
streamer = CSVStreamingResponse(
datasets, params, pagesize=parser.defaults['pagesize'])
return streamer.response()
else:
json_headers(filename='entries.json')
streamer = JSONStreamingResponse(
datasets,
params,
pagesize=parser.defaults['pagesize'],
expand_facets=util.expand_facets
if expand_facets else None,
callback=request.params.get('callback'))
return streamer.response()
solr_browser = Browser(**params)
try:
solr_browser.execute()
except SolrException as e:
return {'errors': [unicode(e)]}
entries = []
for dataset, entry in solr_browser.get_entries():
entry = entry_apply_links(dataset.name, entry)
entry['dataset'] = dataset_apply_links(dataset.as_dict())
entries.append(entry)
if format == 'csv':
return write_csv(entries, response, filename='entries.csv')
if expand_facets and len(datasets) == 1:
facets = solr_browser.get_expanded_facets(datasets[0])
else:
facets = solr_browser.get_facets()
return to_jsonp({
'stats': solr_browser.get_stats(),
'facets': facets,
'results': entries
})
def aggregate(self):
"""
Aggregation of a dataset based on URL parameters. It serves the
aggregation from a cache if possible, and if not it computes it (it's
performed in the aggregation cache for some reason).
"""
# Parse the aggregation parameters to get them into the right format
parser = AggregateParamParser(request.params)
params, errors = parser.parse()
# If there were parsing errors we return them with status code 400
# as jsonp, irrespective of what format was asked for.
if errors:
response.status = 400
return to_jsonp({'errors': errors})
# URL parameters are always singular nouns but we work with some
# as plural nouns so we pop them into the plural version
params['cuts'] = params.pop('cut')
params['drilldowns'] = params.pop('drilldown')
params['measures'] = params.pop('measure')
# Get the dataset and the format and remove from the parameters
dataset = params.pop('dataset')
format = params.pop('format')
# User must have the right to read the dataset to perform aggregation
require.dataset.read(dataset)
# Create response headers from the parameters
self._response_params(params)
try:
# Create an aggregation cache for the dataset and aggregate its
# results. The cache will perform the aggreagation if it doesn't
# have a cached result
cache = AggregationCache(dataset)
result = cache.aggregate(**params)
# If the result has drilldown we create html_url values for its
# dimensions (linked data).
if 'drilldown' in result:
result['drilldown'] = drilldowns_apply_links(
dataset.name, result['drilldown'])
# Do the ETag caching based on the cache_key in the summary
# this is a weird place to do it since the heavy lifting has
# already been performed above. TODO: Needs rethinking.
response.last_modified = dataset.updated_at
if cache.cache_enabled and 'cache_key' in result['summary']:
etag_cache(result['summary']['cache_key'])
except (KeyError, ValueError) as ve:
# We log possible errors and return them with status code 400
log.exception(ve)
response.status = 400
return to_jsonp({'errors': [unicode(ve)]})
# If the requested format is csv we write the drilldown results into
# a csv file and return it, if not we return a jsonp result (default)
if format == 'csv':
return write_csv(result['drilldown'],
response,
filename=dataset.name + '.csv')
return to_jsonp(result)
def index(self, format='html'):
c.query = request.params.items()
c.add_filter = lambda f, v: '?' + urlencode(c.query +
[(f, v)] if (f, v) not in c.query else c.query)
c.del_filter = lambda f, v: '?' + urlencode([(k, x) for k, x in
c.query if (k, x) != (f, v)])
c.results = c.datasets
for language in request.params.getall('languages'):
l = db.aliased(DatasetLanguage)
c.results = c.results.join(l, Dataset._languages)
c.results = c.results.filter(l.code == language)
for territory in request.params.getall('territories'):
t = db.aliased(DatasetTerritory)
c.results = c.results.join(t, Dataset._territories)
c.results = c.results.filter(t.code == territory)
category = request.params.get('category')
if category:
c.results = c.results.filter(Dataset.category == category)
c.results = list(c.results)
c.territory_options = [{'code': code,
'count': count,
'url': h.url_for(controller='dataset',
action='index', territories=code),
'label': COUNTRIES.get(code, code)} \
for (code, count) in DatasetTerritory.dataset_counts(c.results)]
c.language_options = [{'code': code,
'count': count,
'url': h.url_for(controller='dataset',
action='index', languages=code),
'label': LANGUAGES.get(code, code)} \
for (code, count) in DatasetLanguage.dataset_counts(c.results)]
# TODO: figure out where to put this:
ds_ids = [d.id for d in c.results]
if len(ds_ids):
q = db.select([Dataset.category, db.func.count(Dataset.id)],
Dataset.id.in_(ds_ids), group_by=Dataset.category,
order_by=db.func.count(Dataset.id).desc())
c.category_options = [{'category': category,
'count': count,
'url': h.url_for(controller='dataset',
action='index', category=category),
'label': CATEGORIES.get(category, category)} \
for (category, count) in db.session.bind.execute(q).fetchall() \
if category is not None]
else:
c.category_options = []
c._must_revalidate = True
if len(c.results):
dt = max([r.updated_at for r in c.results])
etag_cache_keygen(dt)
if format == 'json':
results = map(lambda d: d.as_dict(), c.results)
results = [dataset_apply_links(r) for r in results]
return to_jsonp({
'datasets': results,
'categories': c.category_options,
'territories': c.territory_options,
'languages': c.language_options
})
elif format == 'csv':
results = map(lambda d: d.as_dict(), c.results)
return write_csv(results, response)
c.show_rss = True
return templating.render('dataset/index.html')
class Api2Controller(BaseController):
def _response_params(self, params):
for k, v in params.items():
k = k.replace('_', ' ').replace('-', ' ').split()
k = '-'.join(['X'] + [l.capitalize() for l in k])
response.headers[k] = unicode(v).encode('ascii', 'ignore')
def aggregate(self):
parser = AggregateParamParser(request.params)
params, errors = parser.parse()
if errors:
response.status = 400
return to_jsonp({'errors': errors})
params['cuts'] = params.pop('cut')
params['drilldowns'] = params.pop('drilldown')
dataset = params.pop('dataset')
format = params.pop('format')
require.dataset.read(dataset)
self._response_params(params)
try:
cache = AggregationCache(dataset)
result = cache.aggregate(**params)
if 'drilldown' in result:
result['drilldown'] = drilldowns_apply_links(
dataset.name, result['drilldown'])
response.last_modified = dataset.updated_at
if cache.cache_enabled and 'cache_key' in result['summary']:
etag_cache(result['summary']['cache_key'])
except (KeyError, ValueError) as ve:
log.exception(ve)
response.status = 400
return to_jsonp({'errors': [unicode(ve)]})
if format == 'csv':
return write_csv(result['drilldown'],
response,
filename=dataset.name + '.csv')
return to_jsonp(result)
def search(self):
parser = SearchParamParser(request.params)
params, errors = parser.parse()
if errors:
response.status = 400
return to_jsonp({'errors': errors})
expand_facets = params.pop('expand_facet_dimensions')
format = params.pop('format')
if format == 'csv':
params['stats'] = False
params['facet_field'] = None
datasets = params.pop('dataset', None)
if datasets is None or not datasets:
q = model.Dataset.all_by_account(c.account)
if params.get('category'):
q = q.filter_by(category=params.pop('category'))
datasets = q.all()
expand_facets = False
if not datasets:
return {'errors': ["No dataset available."]}
params['filter']['dataset'] = []
for dataset in datasets:
require.dataset.read(dataset)
params['filter']['dataset'].append(dataset.name)
response.last_modified = max([d.updated_at for d in datasets])
etag_cache_keygen(parser.key(), response.last_modified)
self._response_params(params)
if params['pagesize'] > parser.defaults['pagesize']:
# http://wiki.nginx.org/X-accel#X-Accel-Buffering
response.headers['X-Accel-Buffering'] = 'no'
if format == 'csv':
csv_headers(response, 'entries.csv')
streamer = CSVStreamingResponse(
datasets, params, pagesize=parser.defaults['pagesize'])
return streamer.response()
else:
json_headers(filename='entries.json')
streamer = JSONStreamingResponse(
datasets,
params,
pagesize=parser.defaults['pagesize'],
expand_facets=_expand_facets if expand_facets else None,
callback=request.params.get('callback'))
return streamer.response()
b = Browser(**params)
try:
b.execute()
except SolrException, e:
return {'errors': [unicode(e)]}
stats, facets, entries = b.get_stats(), b.get_facets(), b.get_entries()
_entries = []
for dataset, entry in entries:
entry = entry_apply_links(dataset.name, entry)
entry['dataset'] = dataset_apply_links(dataset.as_dict())
_entries.append(entry)
if format == 'csv':
return write_csv(_entries, response, filename='entries.csv')
if expand_facets and len(datasets) == 1:
_expand_facets(facets, datasets[0])
return to_jsonp({
'stats': stats,
'facets': facets,
'results': _entries
})
def view(self, dataset, id, format='html'):
"""
Get a specific entry in the dataset, identified by the id. Entry
can be return as html (default), json or csv.
"""
# Generate the dataset
self._get_dataset(dataset)
# Get the entry that matches the given id. c.dataset.entries is
# a generator so we create a list from it's responses based on the
# given constraint
entries = list(c.dataset.entries(c.dataset.alias.c.id == id))
# Since we're trying to get a single entry the list should only
# contain one entry, if not then we return an error
if not len(entries) == 1:
abort(404, _('Sorry, there is no entry %r') % id)
# Add urls to the dataset and assign assign it as a context variable
c.entry = entry_apply_links(dataset, entries.pop())
# Get and set some context variables from the entry
# This shouldn't really be necessary but it's here so nothing gets
# broken
c.id = c.entry.get('id')
c.from_ = c.entry.get('from')
c.to = c.entry.get('to')
c.currency = c.entry.get('currency', c.dataset.currency).upper()
c.time = c.entry.get('time')
# Get the amount for the entry
amount = c.entry.get('amount')
# We adjust for inflation if the user as asked for this to be inflated
if 'inflate' in request.params:
try:
# Inflate the amount. Target date is provided in request.params
# as value for inflate and reference date is the date of the
# entry. We also provide a list of the territories to extract
# a single country for which to do the inflation
c.inflation = h.inflate(amount, request.params['inflate'],
c.time, c.dataset.territories)
# The amount to show should be the inflated amount
# and overwrite the entry's amount as well
c.amount = c.inflation['inflated']
c.entry['amount'] = c.inflation['inflated']
# We include the inflation response in the entry's dict
# HTML description assumes every dict value for the entry
# includes a label so we include a default "Inflation
# adjustment" for it to work.
c.inflation['label'] = 'Inflation adjustment'
c.entry['inflation_adjustment'] = c.inflation
except:
# If anything goes wrong in the try clause (and there's a lot
# that can go wrong). We just say that we can't adjust for
# inflation and set the context amount as the original amount
h.flash_notice(_('Unable to adjust for inflation'))
c.amount = amount
else:
# If we haven't been asked to inflate then we just use the
# original amount
c.amount = amount
# Add custom html for the dataset entry if the dataset has some
# custom html
# 2013-11-17 disabled this as part of removal of genshi as depended on
# a genshi specific helper.
# TODO: reinstate if important
# c.custom_html = h.render_entry_custom_html(c.dataset, c.entry)
# Add the rest of the dimensions relating to this entry into a
# extras dictionary. We first need to exclude all dimensions that
# are already shown and then we can loop through the dimensions
excluded_keys = ('time', 'amount', 'currency', 'from', 'to', 'dataset',
'id', 'name', 'description')
c.extras = {}
if c.dataset:
# Create a dictionary of the dataset dimensions
c.desc = dict([(d.name, d) for d in c.dataset.dimensions])
# Loop through dimensions of the entry
for key in c.entry:
# Entry dimension must be a dataset dimension and not in
# the predefined excluded keys
if key in c.desc and \
key not in excluded_keys:
c.extras[key] = c.entry[key]
# Return entry based on
if format == 'json':
return to_jsonp(c.entry)
elif format == 'csv':
return write_csv([c.entry], response)
else:
return templating.render('entry/view.html')
def to_csv(self):
return write_csv(self.all_entries, response)
