def late(req):
req.content_type = "text/html"
print >> req, pageheaderonly("Late journals", req=req)
for journal in CFG_JOURNALS:
print >> req, "<h2>%s</h2>" % escape(get_coll_i18nname(journal))
results = get_collection_reclist(journal)
print >> req, "<table>"
print >> req, "<tr><th>DOI</th><th>Title</th><th>DOI registration</th><th>Arrival in SCOAP3</th></tr>"
for recid in results:
creation_date = run_sql("SELECT creation_date FROM bibrec WHERE id=%s", (recid, ))[0][0]
record = get_record(recid)
doi = record_get_field_value(record, '024', '7', code='a')
title = record_get_field_value(record, '245', code='a')
doi_date = run_sql("SELECT creation_date FROM doi WHERE doi=%s", (doi, ))
background = "#eee"
if doi_date:
doi_date = doi_date[0][0]
if (creation_date - doi_date).days < 0:
background = "#66FF00"
elif (creation_date - doi_date).days < 1:
background = "#FF6600"
else:
background = "#FF0000"
else:
doi_date = ''
print >> req, '<tr style="background-color: %s;"><td><a href="http://dx.doi.org/%s" target="_blank">%s</td><td>%s</td><td>%s</td><td>%s</td></tr>' % (
background,
escape(doi, True),
escape(doi),
title,
doi_date,
creation_date)
print >> req, "</table>"
def csv(req):
req.content_type = 'text/csv; charset=utf-8'
req.headers_out['content-disposition'] = 'attachment; filename=scoap3.csv'
header = ','.join(['Nation'] + [get_coll_i18nname(journal) for journal in CFG_JOURNALS])
print >> req, header
for nation_tuple in _CFG_NATION_MAP:
query = _build_query(nation_tuple)
line = ','.join([nation_tuple[0]] + [str(len(perform_request_search(p=query, cc=journal, of='intbitset'))) for journal in CFG_JOURNALS])
print >> req, line
def articles(req, i, mode='html'):
try:
i = int(i)
assert 0 <= i < len(_AFFILIATIONS)
except:
raise SERVER_RETURN(HTTP_BAD_REQUEST)
nation = _AFFILIATIONS[i]
ret = []
page_title = "SCOAP3 Articles by authors from %s" % nation
if mode == 'text':
req.content_type = "text/plain; charset=utf8"
req.headers_out['content-disposition'] = ('attachment; filename=%s.txt'
% nation)
else:
req.content_type = "text/html"
if mode == 'text':
print >> req, page_title
print >> req, "-" * len(page_title)
query = _build_query(nation)
for journal in CFG_JOURNALS:
results = perform_request_search(p=query, cc=journal, of='intbitset')
if not results:
continue
ret.append("<h2>%s (%s)</h2" % (escape(get_coll_i18nname(journal)),
len(results)))
ret.append("<p><ul>")
if mode == 'text':
print >> req, ""
print >> req, get_coll_i18nname(journal)
for recid in results:
record = get_record(recid)
title = record_get_field_value(record, '245', code='a')
doi = record_get_field_value(record, '024', '7', code='a')
if mode == 'text':
print >> req, "http://dx.doi.org/%s" % doi
li = ("<li><a href='http://dx.doi.org/{0}' "
"target='_blank'>{1}</a>: {2}</li>")
ret.append(li.format(escape(doi, True), escape(doi), title))
ret.append("</ul></p>")
body = '\n'.join(ret)
if mode == 'text':
return ""
return page(req=req, title=page_title, body=body)
def articles(req, i, mode='html'):
try:
i = int(i)
assert 0 <= i < len(_AFFILIATIONS)
except:
raise SERVER_RETURN(HTTP_BAD_REQUEST)
nation = _AFFILIATIONS[i]
ret = []
page_title = "SCOAP3 Articles by authors from %s" % nation
if mode == 'text':
req.content_type = "text/plain; charset=utf8"
req.headers_out['content-disposition'] = ('attachment; filename=%s.txt'
% nation)
else:
req.content_type = "text/html"
if mode == 'text':
print >> req, page_title
print >> req, "-" * len(page_title)
query = _build_query(nation)
for journal in CFG_JOURNALS:
results = perform_request_search(p=query, cc=journal, of='intbitset')
if not results:
continue
ret.append("<h2>%s (%s)</h2" % (escape(get_coll_i18nname(journal)),
len(results)))
ret.append("<p><ul>")
if mode == 'text':
print >> req, ""
print >> req, get_coll_i18nname(journal)
for recid in results:
record = get_record(recid)
title = record_get_field_value(record, '245', code='a')
doi = record_get_field_value(record, '024', '7', code='a')
if mode == 'text':
print >> req, "http://dx.doi.org/%s" % doi
li = ("<li><a href='http://dx.doi.org/{0}' "
"target='_blank'>{1}</a>: {2}</li>")
ret.append(li.format(escape(doi, True), escape(doi), title))
ret.append("</ul></p>")
body = '\n'.join(ret)
if mode == 'text':
return ""
return page(req=req, title=page_title, body=body)
def csv(req):
req.content_type = 'text/csv; charset=utf-8'
req.headers_out['content-disposition'] = 'attachment; filename=scoap3.csv'
header = (','.join(['Nation']
+ [get_coll_i18nname(journal) for journal in CFG_JOURNALS]))
print >> req, header
for nation in _AFFILIATIONS:
query = _build_query(nation)
line = (','.join([nation]
+ [str(len(perform_request_search(p=query,
cc=journal,
of='intbitset')))
for journal in CFG_JOURNALS]))
print >> req, line
def us_affiliations_csv(req):
req.content_type = 'text/csv; charset=utf-8'
req.headers_out['content-disposition'] = 'attachment; filename=us_aff.csv'
header = ';'.join(['University'] + [get_coll_i18nname(journal) for journal in CFG_JOURNALS] + ['sum'])
print >> req, header
for university in sorted(CFG_SELECTED_AFF):
line = university
count = 0
search = create_search_from_affiliation(university)
for collection in CFG_JOURNALS:
res = perform_request_search(p='/%s/' % (search,), c=collection)
line = line + ";" + str(len(res))
count = count + len(res)
print >> req, line+";"+str(count)
def main():
for journal in CFG_JOURNALS:
name = get_coll_i18nname(journal)
reclist = get_collection_reclist(journal)
print "<h2>%s</h2>" % escape(name)
if not reclist:
print "<p>None yet.</p>"
continue
print "<p><ul>"
for recid in reclist:
record = get_record(recid)
title = remove_html_markup(record_get_field_value(record, '245', code='a'), remove_escaped_chars_p=False).strip()
doi = record_get_field_value(record, '024', '7', code='a')
print '<li><a href="http://dx.doi.org/%s" target="_blank">%s</a>: %s</li>' % (escape(doi, True), escape(doi), title)
print "</ul></p>"
def us_affiliations_csv(req):
req.content_type = 'text/csv; charset=utf-8'
req.headers_out['content-disposition'] = 'attachment; filename=us_aff.csv'
header = (';'.join(['University']
+ [get_coll_i18nname(journal) for journal in CFG_JOURNALS]
+ ['sum']))
print >> req, header
for university in sorted(CFG_SELECTED_AFF):
line = university
count = 0
search = create_search_from_affiliation(university)
for collection in CFG_JOURNALS:
res = perform_request_search(p='/%s/' % (search,), c=collection)
line = line + ";" + str(len(res))
count = count + len(res)
print >> req, line+";"+str(count)
def late(req):
req.content_type = "text/html"
print >> req, pageheaderonly("Late journals", req=req)
th = ("<tr><th>DOI</th><th>Title</th><th>DOI registration</th>"
"<th>Arrival in SCOAP3</th></tr>")
tr = ("<tr style='background-color: {0};'><td>"
"<a href='http://dx.doi.org/{1}' target='_blank'>{2}</td>"
"<td>{3}</td><td>{4}</td><td>{5}</td></tr>")
sql_bibrec = "SELECT creation_date FROM bibrec WHERE id=%s"
sql_doi = "SELECT creation_date FROM doi WHERE doi=%s"
for journal in CFG_JOURNALS:
print >> req, "<h2>%s</h2>" % escape(get_coll_i18nname(journal))
results = get_collection_reclist(journal)
print >> req, "<table>"
print >> req, th
for recid in results:
creation_date = run_sql(sql_bibrec, (recid, ))[0][0]
record = get_record(recid)
doi = record_get_field_value(record, '024', '7', code='a')
title = record_get_field_value(record, '245', code='a')
doi_date = run_sql(sql_doi, (doi, ))
background = "#eee"
if doi_date:
doi_date = doi_date[0][0]
if (creation_date - doi_date).days < 0:
background = "#66FF00"
elif (creation_date - doi_date).days < 1:
background = "#FF6600"
else:
background = "#FF0000"
else:
doi_date = ''
print >> req, tr.format(background,
escape(doi, True),
escape(doi),
title,
doi_date,
creation_date)
print >> req, "</table>"
def display_collection(req, c, aas, verbose, ln, em=""):
"""Display search interface page for collection c by looking
in the collection cache."""
_ = gettext_set_language(ln)
req.argd = drop_default_urlargd({'aas': aas, 'verbose': verbose, 'ln': ln, 'em' : em},
search_interface_default_urlargd)
if em != "":
em = em.split(",")
# get user ID:
try:
uid = getUid(req)
user_preferences = {}
if uid == -1:
return page_not_authorized(req, "../",
text="You are not authorized to view this collection",
navmenuid='search')
elif uid > 0:
user_preferences = get_user_preferences(uid)
except Error:
register_exception(req=req, alert_admin=True)
return page(title=_("Internal Error"),
body=create_error_box(req, verbose=verbose, ln=ln),
description="%s - Internal Error" % CFG_SITE_NAME,
keywords="%s, Internal Error" % CFG_SITE_NAME,
language=ln,
req=req,
navmenuid='search')
# deduce collection id:
normalised_name = get_coll_normalised_name(c)
colID = get_colID(normalised_name)
if type(colID) is not int:
page_body = '<p>' + (_("Sorry, collection %s does not seem to exist.") % ('<strong>' + str(c) + '</strong>')) + '</p>'
page_body = '<p>' + (_("You may want to start browsing from %s.") % ('<a href="' + CFG_SITE_URL + '?ln=' + ln + '">' + get_coll_i18nname(CFG_SITE_NAME, ln) + '</a>')) + '</p>'
if req.header_only:
raise apache.SERVER_RETURN, apache.HTTP_NOT_FOUND
return page(title=_("Collection %s Not Found") % cgi.escape(c),
body=page_body,
description=(CFG_SITE_NAME + ' - ' + _("Not found") + ': ' + cgi.escape(str(c))),
keywords="%s" % CFG_SITE_NAME,
uid=uid,
language=ln,
req=req,
navmenuid='search')
if normalised_name != c:
redirect_to_url(req, normalised_name, apache.HTTP_MOVED_PERMANENTLY)
# start display:
req.content_type = "text/html"
req.send_http_header()
c_body, c_navtrail, c_portalbox_lt, c_portalbox_rt, c_portalbox_tp, c_portalbox_te, \
c_last_updated = perform_display_collection(colID, c, aas, ln, em,
user_preferences.get('websearch_helpbox', 1))
if em == "" or EM_REPOSITORY["body"] in em:
try:
title = get_coll_i18nname(c, ln)
except:
title = ""
else:
title = ""
show_title_p = True
body_css_classes = []
if c == CFG_SITE_NAME:
# Do not display title on home collection
show_title_p = False
body_css_classes.append('home')
if len(collection_reclist_cache.cache.keys()) == 1:
# if there is only one collection defined, do not print its
# title on the page as it would be displayed repetitively.
show_title_p = False
if aas == -1:
show_title_p = False
if CFG_INSPIRE_SITE == 1:
# INSPIRE should never show title, but instead use css to
# style collections
show_title_p = False
body_css_classes.append(nmtoken_from_string(c))
# RSS:
rssurl = CFG_SITE_URL + '/rss'
rssurl_params = []
if c != CFG_SITE_NAME:
rssurl_params.append('cc=' + quote(c))
if ln != CFG_SITE_LANG and \
c in CFG_WEBSEARCH_RSS_I18N_COLLECTIONS:
rssurl_params.append('ln=' + ln)
if rssurl_params:
rssurl += '?' + '&'.join(rssurl_params)
if 'hb' in CFG_WEBSEARCH_USE_MATHJAX_FOR_FORMATS:
metaheaderadd = get_mathjax_header(req.is_https())
else:
metaheaderadd = ''
return page(title=title,
body=c_body,
navtrail=c_navtrail,
description="%s - %s" % (CFG_SITE_NAME, c),
keywords="%s, %s" % (CFG_SITE_NAME, c),
metaheaderadd=metaheaderadd,
uid=uid,
language=ln,
req=req,
cdspageboxlefttopadd=c_portalbox_lt,
cdspageboxrighttopadd=c_portalbox_rt,
titleprologue=c_portalbox_tp,
titleepilogue=c_portalbox_te,
lastupdated=c_last_updated,
navmenuid='search',
rssurl=rssurl,
body_css_classes=body_css_classes,
show_title_p=show_title_p,
show_header=em == "" or EM_REPOSITORY["header"] in em,
show_footer=em == "" or EM_REPOSITORY["footer"] in em)
def display_collection(req, c, aas, verbose, ln, em=""):
"""Display search interface page for collection c by looking
in the collection cache."""
_ = gettext_set_language(ln)
req.argd = drop_default_urlargd({'aas': aas, 'verbose': verbose, 'ln': ln, 'em' : em},
search_interface_default_urlargd)
if em != "":
em = em.split(",")
# get user ID:
try:
uid = getUid(req)
user_preferences = {}
if uid == -1:
return page_not_authorized(req, "../",
text="You are not authorized to view this collection",
navmenuid='search')
elif uid > 0:
user_preferences = get_user_preferences(uid)
except Error:
register_exception(req=req, alert_admin=True)
return page(title=_("Internal Error"),
body=create_error_box(req, verbose=verbose, ln=ln),
description="%s - Internal Error" % CFG_SITE_NAME,
keywords="%s, Internal Error" % CFG_SITE_NAME,
language=ln,
req=req,
navmenuid='search')
# start display:
req.content_type = "text/html"
req.send_http_header()
# deduce collection id:
colID = get_colID(get_coll_normalised_name(c))
if type(colID) is not int:
page_body = '<p>' + (_("Sorry, collection %s does not seem to exist.") % ('<strong>' + str(c) + '</strong>')) + '</p>'
page_body = '<p>' + (_("You may want to start browsing from %s.") % ('<a href="' + CFG_SITE_URL + '?ln=' + ln + '">' + get_coll_i18nname(CFG_SITE_NAME, ln) + '</a>')) + '</p>'
if req.header_only:
raise apache.SERVER_RETURN, apache.HTTP_NOT_FOUND
return page(title=_("Collection %s Not Found") % cgi.escape(c),
body=page_body,
description=(CFG_SITE_NAME + ' - ' + _("Not found") + ': ' + cgi.escape(str(c))),
keywords="%s" % CFG_SITE_NAME,
uid=uid,
language=ln,
req=req,
navmenuid='search')
c_body, c_navtrail, c_portalbox_lt, c_portalbox_rt, c_portalbox_tp, c_portalbox_te, \
c_last_updated = perform_display_collection(colID, c, aas, ln, em,
user_preferences.get('websearch_helpbox', 1))
if em == "" or EM_REPOSITORY["body"] in em:
try:
title = get_coll_i18nname(c, ln)
except:
title = ""
else:
title = ""
show_title_p = True
body_css_classes = []
if c == CFG_SITE_NAME:
# Do not display title on home collection
show_title_p = False
body_css_classes.append('home')
if len(collection_reclist_cache.cache.keys()) == 1:
# if there is only one collection defined, do not print its
# title on the page as it would be displayed repetitively.
show_title_p = False
if aas == -1:
show_title_p = False
if CFG_INSPIRE_SITE == 1:
# INSPIRE should never show title, but instead use css to
# style collections
show_title_p = False
body_css_classes.append(nmtoken_from_string(c))
# RSS:
rssurl = CFG_SITE_URL + '/rss'
rssurl_params = []
if c != CFG_SITE_NAME:
rssurl_params.append('cc=' + quote(c))
if ln != CFG_SITE_LANG and \
c in CFG_WEBSEARCH_RSS_I18N_COLLECTIONS:
rssurl_params.append('ln=' + ln)
if rssurl_params:
rssurl += '?' + '&'.join(rssurl_params)
if 'hb' in CFG_WEBSEARCH_USE_MATHJAX_FOR_FORMATS:
metaheaderadd = get_mathjax_header(req.is_https())
else:
metaheaderadd = ''
return page(title=title,
body=c_body,
navtrail=c_navtrail,
description="%s - %s" % (CFG_SITE_NAME, c),
keywords="%s, %s" % (CFG_SITE_NAME, c),
metaheaderadd=metaheaderadd,
uid=uid,
language=ln,
req=req,
cdspageboxlefttopadd=c_portalbox_lt,
cdspageboxrighttopadd=c_portalbox_rt,
titleprologue=c_portalbox_tp,
titleepilogue=c_portalbox_te,
lastupdated=c_last_updated,
navmenuid='search',
rssurl=rssurl,
body_css_classes=body_css_classes,
show_title_p=show_title_p,
show_header=em == "" or EM_REPOSITORY["header"] in em,
show_footer=em == "" or EM_REPOSITORY["footer"] in em)
def name_ln(self):
from invenio.search_engine import get_coll_i18nname
return get_coll_i18nname(self.name, g.ln).decode('utf-8')
def answer(self, req, user_info, of, cc, colls_to_search, p, f, search_units, ln):
"""
Answer question given by context.
Return (relevance, html_string) where relevance is integer
from 0 to 100 indicating how relevant to the question the
answer is (see C{CFG_WEBSEARCH_SERVICE_MAX_SERVICE_ANSWER_RELEVANCE} for details) ,
and html_string being a formatted answer.
"""
from invenio.search_engine import \
get_permitted_restricted_collections, \
get_coll_i18nname, \
collection_i18nname_cache, \
collection_restricted_p
_ = gettext_set_language(ln)
# stem search units. remove those with field
# TODO: search in hosted collection names too
# TODO: ignore unattached trees
# TODO: use synonyms
if f or (CFG_WEBSEARCH_COLLECTION_NAMES_SEARCH < 0) or \
(CFG_WEBSEARCH_COLLECTION_NAMES_SEARCH == 0 and cc != CFG_SITE_NAME):
return (0, '')
words = [stem(unit[1], ln) for unit in search_units if unit[2] in ('', 'collection')] # Stemming
if not words:
return (0, '')
permitted_restricted_collections = get_permitted_restricted_collections(user_info)
cache = self.get_data_cache()
matching_collections = {}
for word in words:
if CFG_CERN_SITE and word == 'cern':
# This keyword is useless here...
continue
colls = cache.get(word.lower(), [])
for coll in colls:
if collection_restricted_p(coll) and \
not coll in permitted_restricted_collections:
# Skip restricted collection user do not have access
continue
if not matching_collections.has_key(coll):
matching_collections[coll] = 0
matching_collections[coll] += 1
matching_collections_sorted = sorted(matching_collections.iteritems(), key=lambda (k, v): (v, k), reverse=True)
if not matching_collections_sorted:
return (0, '')
matching_collections_names = [(get_coll_i18nname(coll, ln, False), CFG_SITE_URL + '/collection/' + urllib.quote(coll, safe='') + '?ln=en') \
for coll, score in matching_collections_sorted]
best_score = matching_collections_sorted[0][1]
best_coll_words = whitespace_re.split(matching_collections_sorted[0][0])
relevance = min(100, max(0, (100 * float(2 * best_score) / float(len(best_coll_words) + len(words)) - 10)))
if (('submit' in p.lower()) or (_('submit') in p.lower())) and \
not (('submit' in best_coll_words) or (_('submit') in best_coll_words)):
# User is probably looking for a submission. Decrease relevance
relevance = max(0, relevance - 30)
return (relevance, self.display_answer_helper(matching_collections_names, ln))
def display_collection(req, c, aas, verbose, ln):
"""Display search interface page for collection c by looking
in the collection cache."""
_ = gettext_set_language(ln)
req.argd = drop_default_urlargd({
'aas': aas,
'verbose': verbose,
'ln': ln
}, search_interface_default_urlargd)
# get user ID:
try:
uid = getUid(req)
user_preferences = {}
if uid == -1:
return page_not_authorized(
req,
"../",
text="You are not authorized to view this collection",
navmenuid='search')
elif uid > 0:
user_preferences = get_user_preferences(uid)
except Error:
register_exception(req=req, alert_admin=True)
return page(title=_("Internal Error"),
body=create_error_box(req, verbose=verbose, ln=ln),
description="%s - Internal Error" % CFG_SITE_NAME,
keywords="%s, Internal Error" % CFG_SITE_NAME,
language=ln,
req=req,
navmenuid='search')
# start display:
req.content_type = "text/html"
req.send_http_header()
# deduce collection id:
colID = get_colID(get_coll_normalised_name(c))
if type(colID) is not int:
page_body = '<p>' + (
_("Sorry, collection %s does not seem to exist.") %
('<strong>' + str(c) + '</strong>')) + '</p>'
page_body = '<p>' + (
_("You may want to start browsing from %s.") %
('<a href="' + CFG_SITE_URL + '?ln=' + ln + '">' +
get_coll_i18nname(CFG_SITE_NAME, ln) + '</a>')) + '</p>'
if req.header_only:
raise apache.SERVER_RETURN, apache.HTTP_NOT_FOUND
return page(title=_("Collection %s Not Found") % cgi.escape(c),
body=page_body,
description=(CFG_SITE_NAME + ' - ' + _("Not found") +
': ' + cgi.escape(str(c))),
keywords="%s" % CFG_SITE_NAME,
uid=uid,
language=ln,
req=req,
navmenuid='search')
# wash `aas' argument:
if not os.path.exists("%s/collections/%d/body-as=%d-ln=%s.html" % \
(CFG_CACHEDIR, colID, aas, ln)):
# nonexistent `aas' asked for, fall back to Simple Search:
aas = 0
# display collection interface page:
try:
filedesc = open("%s/collections/%d/navtrail-as=%d-ln=%s.html" % \
(CFG_CACHEDIR, colID, aas, ln), "r")
c_navtrail = filedesc.read()
filedesc.close()
except:
c_navtrail = ""
try:
filedesc = open("%s/collections/%d/body-as=%d-ln=%s.html" % \
(CFG_CACHEDIR, colID, aas, ln), "r")
c_body = filedesc.read()
filedesc.close()
except:
c_body = ""
try:
filedesc = open("%s/collections/%d/portalbox-tp-ln=%s.html" % \
(CFG_CACHEDIR, colID, ln), "r")
c_portalbox_tp = filedesc.read()
filedesc.close()
except:
c_portalbox_tp = ""
try:
filedesc = open("%s/collections/%d/portalbox-te-ln=%s.html" % \
(CFG_CACHEDIR, colID, ln), "r")
c_portalbox_te = filedesc.read()
filedesc.close()
except:
c_portalbox_te = ""
try:
filedesc = open("%s/collections/%d/portalbox-lt-ln=%s.html" % \
(CFG_CACHEDIR, colID, ln), "r")
c_portalbox_lt = filedesc.read()
filedesc.close()
except:
c_portalbox_lt = ""
try:
# show help boxes (usually located in "tr", "top right")
# if users have not banned them in their preferences:
c_portalbox_rt = ""
if user_preferences.get('websearch_helpbox', 1) > 0:
filedesc = open("%s/collections/%d/portalbox-rt-ln=%s.html" % \
(CFG_CACHEDIR, colID, ln), "r")
c_portalbox_rt = filedesc.read()
filedesc.close()
except:
c_portalbox_rt = ""
try:
filedesc = open("%s/collections/%d/last-updated-ln=%s.html" % \
(CFG_CACHEDIR, colID, ln), "r")
c_last_updated = filedesc.read()
filedesc.close()
except:
c_last_updated = ""
try:
title = get_coll_i18nname(c, ln)
except:
title = ""
show_title_p = True
body_css_classes = []
if c == CFG_SITE_NAME:
# Do not display title on home collection
show_title_p = False
body_css_classes.append('home')
if len(collection_reclist_cache.cache.keys()) == 1:
# if there is only one collection defined, do not print its
# title on the page as it would be displayed repetitively.
show_title_p = False
if aas == -1:
show_title_p = False
if CFG_INSPIRE_SITE == 1:
# INSPIRE should never show title, but instead use css to
# style collections
show_title_p = False
body_css_classes.append(nmtoken_from_string(c))
# RSS:
rssurl = CFG_SITE_URL + '/rss'
rssurl_params = []
if c != CFG_SITE_NAME:
rssurl_params.append('cc=' + quote(c))
if ln != CFG_SITE_LANG and \
c in CFG_WEBSEARCH_RSS_I18N_COLLECTIONS:
rssurl_params.append('ln=' + ln)
if rssurl_params:
rssurl += '?' + '&'.join(rssurl_params)
if 'hb' in CFG_WEBSEARCH_USE_MATHJAX_FOR_FORMATS:
metaheaderadd = get_mathjax_header(req.is_https())
else:
metaheaderadd = ''
return page(title=title,
body=c_body,
navtrail=c_navtrail,
description="%s - %s" % (CFG_SITE_NAME, c),
keywords="%s, %s" % (CFG_SITE_NAME, c),
metaheaderadd=metaheaderadd,
uid=uid,
language=ln,
req=req,
cdspageboxlefttopadd=c_portalbox_lt,
cdspageboxrighttopadd=c_portalbox_rt,
titleprologue=c_portalbox_tp,
titleepilogue=c_portalbox_te,
lastupdated=c_last_updated,
navmenuid='search',
rssurl=rssurl,
body_css_classes=body_css_classes,
show_title_p=show_title_p)
def answer(self, req, user_info, of, cc, colls_to_search, p, f,
search_units, ln):
"""
Answer question given by context.
Return (relevance, html_string) where relevance is integer
from 0 to 100 indicating how relevant to the question the
answer is (see C{CFG_WEBSEARCH_SERVICE_MAX_SERVICE_ANSWER_RELEVANCE} for details) ,
and html_string being a formatted answer.
"""
from invenio.search_engine import \
get_permitted_restricted_collections, \
get_coll_i18nname, \
collection_i18nname_cache, \
collection_restricted_p
_ = gettext_set_language(ln)
# stem search units. remove those with field
# TODO: search in hosted collection names too
# TODO: ignore unattached trees
# TODO: use synonyms
if f or (CFG_WEBSEARCH_COLLECTION_NAMES_SEARCH < 0) or \
(CFG_WEBSEARCH_COLLECTION_NAMES_SEARCH == 0 and cc != CFG_SITE_NAME):
return (0, '')
words = [
stem(unit[1], ln) for unit in search_units
if unit[2] in ('', 'collection')
] # Stemming
if not words:
return (0, '')
permitted_restricted_collections = get_permitted_restricted_collections(
user_info)
cache = self.get_data_cache()
matching_collections = {}
for word in words:
if CFG_CERN_SITE and word == 'cern':
# This keyword is useless here...
continue
colls = cache.get(word.lower(), [])
for coll in colls:
if collection_restricted_p(coll) and \
not coll in permitted_restricted_collections:
# Skip restricted collection user do not have access
continue
if not matching_collections.has_key(coll):
matching_collections[coll] = 0
matching_collections[coll] += 1
matching_collections_sorted = sorted(matching_collections.iteritems(),
key=lambda (k, v): (v, k),
reverse=True)
if not matching_collections_sorted:
return (0, '')
matching_collections_names = [(get_coll_i18nname(coll, ln, False), CFG_SITE_URL + '/collection/' + urllib.quote(coll, safe='') + '?ln=en') \
for coll, score in matching_collections_sorted]
best_score = matching_collections_sorted[0][1]
best_coll_words = whitespace_re.split(
matching_collections_sorted[0][0])
relevance = min(
100,
max(0, (100 * float(2 * best_score) /
float(len(best_coll_words) + len(words)) - 10)))
if (('submit' in p.lower()) or (_('submit') in p.lower())) and \
not (('submit' in best_coll_words) or (_('submit') in best_coll_words)):
# User is probably looking for a submission. Decrease relevance
relevance = max(0, relevance - 30)
return (relevance,
self.display_answer_helper(matching_collections_names, ln))
