def article_json(id=None):
data = karp_query('query', {'q': "extended||and|url|equals|%s" % (id)})
if data['hits']['total'] == 1:
page = jsonify(data['hits']['hits'][0]['_source'])
return set_cache(page)
data = karp_query('query',
{'q': "extended||and|id.search|equals|%s" % (id)})
if data['hits']['total'] == 1:
page = jsonify(data['hits']['hits'][0]['_source'])
return set_cache(page)
def compute_article(lang="", cache=True, url=''):
set_language_switch_link("article_index", lang=lang)
art, lang = getcache('article', lang, cache)
if art is not None:
return art
show = ','.join(
['name', 'url', 'undertitel', 'lifespan', 'undertitel_eng'])
infotext = helpers.get_infotext("article", request.url_rule.rule)
if lang == 'sv':
data = karp_query('minientry', {
'q':
"extended||and|namn|exists",
'show':
show,
'sort':
'sorteringsnamn.sort,sorteringsnamn.init,sorteringsnamn,tilltalsnamn.sort,tilltalsnamn'
},
mode="skbllinks")
else:
data = karp_query('minientry', {
'q':
"extended||and|namn|exists",
'show':
show,
'sort':
'sorteringsnamn.eng_sort,sorteringsnamn.eng_init,sorteringsnamn,tilltalsnamn.sort,tilltalsnamn'
},
mode="skbllinks")
art = render_template('list.html',
hits=data["hits"],
headline=gettext(u'Women A-Z'),
alphabetic=True,
split_letters=True,
infotext=infotext,
title='Articles',
page_url=url)
try:
with mc_pool.reserve() as client:
client.set(cache_name('article', lang),
art,
time=app.config['CACHE_TIME'])
except:
# TODO what to do?
pass
return art
def article(id=None):
rule = request.url_rule
if 'sv' in rule.rule:
lang = "sv"
else:
lang = "en"
pagename = 'article_' + id
page = check_cache(pagename, lang=lang)
if page is not None:
return page
data = karp_query('query', {'q': "extended||and|url|equals|%s" % (id)})
if data['hits']['total'] == 0:
data = karp_query('query',
{'q': "extended||and|id.search|equals|%s" % (id)})
set_language_switch_link("article_index", id)
page = show_article(data, lang)
return set_cache(page, name=pagename, lang=lang, no_hits=1)
def compute_organisation(lang="", infotext="", cache=True, url=''):
set_language_switch_link("organisation_index", lang=lang)
art, lang = getcache('organisation', lang, cache)
if art is not None:
return art
infotext = helpers.get_infotext("organisation", request.url_rule.rule)
if lang == "en":
data = karp_query(
'minientry', {
'q': 'extended||and|anything|regexp|.*',
'show': 'organisationsnamn,organisationstyp_eng'
})
typefield = "type_eng"
else:
data = karp_query(
'minientry', {
'q': 'extended||and|anything|regexp|.*',
'show': 'organisationsnamn,organisationstyp'
})
typefield = "type"
nested_obj = {}
for hit in data['hits']['hits']:
for org in hit['_source'].get('organisation', []):
orgtype = helpers.unescape(org.get(typefield, '-'))
if orgtype not in nested_obj:
nested_obj[orgtype] = defaultdict(set)
nested_obj[orgtype][org.get('name', '-')].add(hit['_id'])
art = render_template('nestedbucketresults.html',
results=nested_obj,
title=gettext("Organisations"),
infotext=infotext,
name='organisation',
page_url=url)
try:
with mc_pool.reserve() as client:
client.set(cache_name('organisation', lang),
art,
time=app.config['CACHE_TIME'])
except:
# TODO what to do?
pass
return art
def compute_artikelforfattare(infotext='',
description='',
lang="",
cache=True,
url=''):
set_language_switch_link("articleauthor_index", lang=lang)
art, lang = getcache('author', lang, cache)
if art is not None:
return art
q_data = {
'buckets':
'artikel_forfattare_fornamn.bucket,artikel_forfattare_efternamn.bucket'
}
data = karp_query('statlist', q_data)
# strip kw0 to get correct sorting
stat_table = [[kw[0].strip()] + kw[1:] for kw in data['stat_table']
if kw[0] != ""]
stat_table = [[kw[1] + ',', kw[0], kw[2]] for kw in stat_table]
# Remove duplicates and some wrong ones (because of backend limitation):
stoplist = {
u"Grevesmühl,Kajsa": True,
u"Ohrlander,Anders": True,
u"Myrberg Burström,Mats": True,
u"Burström,Nanouschka": True,
u"Ljung,Yvonne": True,
u"Lindholm,Barbro": True
}
added = {}
new_stat_table = []
for item in stat_table:
fullname = item[0] + item[1]
if fullname not in added and fullname not in stoplist:
new_stat_table.append(item)
added[fullname] = True
art = render_template('bucketresults.html',
results=new_stat_table,
alphabetical=True,
title=gettext('Article authors'),
name='articleauthor',
infotext=infotext,
description=description,
sortnames=True,
page_url=url)
try:
with mc_pool.reserve() as client:
client.set(cache_name('author', lang),
art,
time=app.config['CACHE_TIME'])
except:
# TODO what to do?
pass
return art
def find_link(searchstring):
# Finds an article based on ISNI or name
if re.search('^[0-9 ]*$', searchstring):
searchstring = searchstring.replace(" ", "")
data = karp_query(
'query',
{'q': "extended||and|swoid.search|equals|%s" % (searchstring)})
else:
parts = searchstring.split(" ")
if "," in searchstring or len(
parts) == 1: # When there is only a first name (a queen or so)
# case 1: "Margareta"
# case 2: "Margareta, drottning"
firstname = parts[0] if len(parts) == 1 else searchstring
data = karp_query('query', {
'q':
"extended||and|fornamn.search|contains|%s" % (firstname)
})
else:
fornamn = " ".join(parts[0:-1])
prefix = ""
last_fornamn = fornamn.split(" ")[-1]
if last_fornamn == "von" or last_fornamn == "af":
fornamn = " ".join(fornamn.split(" ")[0:-1])
prefix = last_fornamn + " "
efternamn = prefix + parts[-1]
data = karp_query(
'query', {
'q':
"extended||and|fornamn.search|contains|%s||and|efternamn.search|contains|%s"
% (fornamn, efternamn)
})
# The expected case: only one hit is found
if data['hits']['total'] == 1:
url = data['hits']['hits'][0]['_source'].get('url')
es_id = data['hits']['hits'][0]['_id']
return data, (url or es_id)
# Otherwise just return the data
else:
return data, ''
def compute_place(lang="", cache=True, url=''):
set_language_switch_link("place_index", lang=lang)
art, lang = getcache('place', lang, cache)
if art is not None:
return art
infotext = helpers.get_infotext("place", request.url_rule.rule)
def parse(kw):
place = kw.get('key')
# May be used to parse names with or without coordinates:
# "Lysekil" or "Lysekil|58.275573|11.435558"
if '|' in place:
name, lat, lon = place.split('|')
else:
name = place.strip()
lat, lon = 0, 0
placename = name if name else '%s, %s' % (lat, lon)
return {
'name': placename,
'lat': lat,
'lon': lon,
'count': kw.get('doc_count')
}
def has_name(kw):
name = kw.get('key').split('|')[0]
if name and u"(osäker uppgift)" not in name:
return name
else:
return None
# To use the coordinates, use 'getplaces' instead of 'getplacenames'
data = karp_query('getplacenames', {})
stat_table = [parse(kw) for kw in data['places'] if has_name(kw)]
art = render_template('places.html',
places=stat_table,
title=gettext("Placenames"),
infotext=infotext,
description=helpers.get_shorttext(infotext),
page_url=url)
try:
with mc_pool.reserve() as client:
client.set(cache_name('place', lang),
art,
time=app.config['CACHE_TIME'])
except:
# TODO what to do?
pass
return art
def search():
set_language_switch_link("search")
search = request.args.get('q', '').encode('utf-8')
pagename = 'search' + urllib.quote(search)
page = check_cache(pagename)
if page is not None:
return page
advanced_search_text = ''
if search:
show = ','.join(
['name', 'url', 'undertitel', 'undertitel_eng', 'lifespan'])
karp_q = {
'highlight': True,
'size': app.config['SEARCH_RESULT_SIZE'],
'show': show
}
if '*' in search:
search = re.sub('(?<!\.)\*', '.*', search)
karp_q['q'] = "extended||and|anything|regexp|%s" % search
else:
karp_q['q'] = "extended||and|anything|contains|%s" % search
data = karp_query('minientry', karp_q, mode='skbl')
with app.open_resource("static/pages/advanced-search/%s.html" %
(g.language)) as f:
advanced_search_text = f.read()
karp_url = "https://spraakbanken.gu.se/karp/#?mode=skbl&advanced=false&hpp=25&extended=and%7Cnamn%7Cequals%7C&searchTab=simple&page=1&search=simple%7C%7C" + search.decode(
"utf-8")
else:
data = {"hits": {"total": 0, "hits": []}}
karp_url = ""
search = u'\u200b'.encode('utf8')
t = render_template(
'list.html',
headline="",
subheadline=gettext('Hits for "%s"') % search.decode("UTF-8"),
hits_name=data["hits"],
hits=data["hits"],
advanced_search_text=advanced_search_text.decode("UTF-8"),
search=search.decode("UTF-8"),
alphabetic=True,
karp_url=karp_url,
more=data["hits"]["total"] > app.config["SEARCH_RESULT_SIZE"],
show_lang_switch=False)
return set_cache(t, name=pagename, no_hits=data["hits"]["total"])
def bucketcall(queryfield='',
name='',
title='',
sortby='',
lastnamefirst=False,
infotext='',
description='',
query='',
alphabetical=False,
insert_entries=None,
page_url=''):
q_data = {'buckets': '%s.bucket' % queryfield}
if query:
q_data['q'] = query
data = karp_query('statlist', q_data)
# Strip kw0 to get correct sorting
stat_table = [[kw[0].strip()] + kw[1:] for kw in data['stat_table']
if kw[0] != ""]
# Insert entries that function as references
if insert_entries:
stat_table.extend(insert_entries)
if sortby:
stat_table.sort(key=sortby)
else:
stat_table.sort()
if lastnamefirst:
stat_table = [[kw[1] + ',', kw[0], kw[2]] for kw in stat_table]
# if showfield:
# stat_table = [[showfield(kw), kw[2]] for kw in stat_table]
return render_template('bucketresults.html',
results=stat_table,
alphabetical=alphabetical,
title=gettext(title),
name=name,
infotext=infotext,
description=description,
page_url=page_url)
def place(place=None):
pagename = urllib.quote('place_' + place.encode('utf8'))
art = check_cache(pagename)
if art is not None:
return art
lat = request.args.get('lat')
lon = request.args.get('lon')
set_language_switch_link("place_index", place)
hits = karp_query('query', {
'q':
"extended||and|plats.searchraw|equals|%s" % (place.encode('utf-8'))
})
no_hits = hits['hits']['total']
if no_hits > 0:
page = render_template('placelist.html',
title=place,
lat=lat,
lon=lon,
headline=place,
hits=hits["hits"])
else:
page = render_template('page.html',
content=gettext('Contents could not be found!'))
return set_cache(page, name=pagename, no_hits=no_hits)
def searchresult(result,
name='',
searchfield='',
imagefolder='',
query='',
searchtype='equals',
title='',
authorinfo=False,
lang='',
show_lang_switch=True,
cache=True):
set_language_switch_link("%s_index" % name, result)
try:
result = result.encode("UTF-8")
pagename = name + '_' + urllib.quote(result)
art = check_cache(pagename, lang)
if art is not None:
return art
show = ','.join(
['name', 'url', 'undertitel', 'lifespan', 'undertitel_eng'])
if query:
hits = karp_query('minientry', {'q': query, 'show': show})
else:
hits = karp_query(
'minientry', {
'q':
"extended||and|%s.search|%s|%s" %
(searchfield, searchtype, result),
'show':
show
})
title = title or result.decode("UTF-8")
no_hits = hits['hits']['total']
if no_hits > 0:
picture = None
if os.path.exists(app.config.root_path +
'/static/images/%s/%s.jpg' %
(imagefolder, result)):
picture = '/static/images/%s/%s.jpg' % (imagefolder, result)
page = render_template('list.html',
picture=picture,
alphabetic=True,
title=title,
headline=title,
hits=hits["hits"],
authorinfo=authorinfo,
show_lang_switch=show_lang_switch)
if no_hits >= app.config['CACHE_HIT_LIMIT']:
try:
with mc_pool.reserve() as client:
client.set(cache_name(pagename, lang),
page,
time=app.config['CACHE_TIME'])
except:
# TODO what to do?
pass
return page
else:
return render_template(
'page.html', content=gettext('Contents could not be found!'))
except Exception as e:
return render_template(
'page.html',
content="%s\n%s: extended||and|%s.search|%s|%s" %
(e, app.config['KARP_BACKEND'], searchfield, searchtype, result))