def prepare_data_cache(self):
"""
"Index" submission names
"""
from invenio.websubmit_dblayer import get_categories_of_doctype
res = run_sql("SELECT sdocname, ldocname FROM sbmDOCTYPE")
# TODO: only consider submissions that are attached to the tree
if CFG_CERN_SITE:
for submission_name, submission_config in CERN_GENSBM_SUBMISSIONS_CONFIG.iteritems(
):
if not submission_config.has_key('redirect'):
res += (('GENSBM#' +
nmtoken_from_string(cgi.escape(submission_name)),
submission_name), )
cache = {}
for doctype, submission_name in res:
## categories_and_submission_name = ' '.join(get_categories_of_doctype(doctype)) + \
## ' ' + submission_name
# Add submission name info
if CFG_CERN_SITE and doctype in ('ALIPH', 'BULIS', 'CMSREL',
'BULBN', 'BSA'):
# These submissions are not interesting here
continue
for word in clean_and_split_words_and_stem(submission_name):
if not word.strip():
continue
if not cache.has_key(word):
cache[word] = []
item = (doctype, submission_name, '*')
if not item in cache[word]:
cache[word].append(item)
# Add submission categories info
if CFG_CERN_SITE and doctype in ('CMSPUB', 'CMSCOM', 'CMSCMC',
'ATLPUB', 'ATLCOM', 'ATLCMC',
'LHCBPB', 'LHCPCM', 'LHCBCC'):
# These categories are not interesting here
continue
categories = get_categories_of_doctype(doctype)
for dummy, category, dummy in categories:
for word in clean_and_split_words_and_stem(submission_name +
' ' + category):
if not word.strip():
continue
if not cache.has_key(word):
cache[word] = []
item = (doctype, "%s (%s)" % (category, submission_name),
category)
if not item in cache[word]:
cache[word].append(item)
return cache
def test_convert_string_to_nmtoken(self):
"""htmlutils - converting string to Nmtoken"""
# TODO: possibly extend this test to include 'extenders' and
# 'combining characters' as defined in
# http://www.w3.org/TR/2000/REC-xml-20001006#NT-Nmtoken
ascii_str = "".join([chr(i) for i in range(0, 256)])
nmtoken = nmtoken_from_string(ascii_str)
for char in nmtoken:
self.assert_(char in ['.', '-', '_', ':'] or char.isalnum())
def test_convert_string_to_nmtoken(self):
"""htmlutils - converting string to Nmtoken"""
# TODO: possibly extend this test to include 'extenders' and
# 'combining characters' as defined in
# http://www.w3.org/TR/2000/REC-xml-20001006#NT-Nmtoken
ascii_str = "".join([chr(i) for i in range(0, 256)])
nmtoken = nmtoken_from_string(ascii_str)
for char in nmtoken:
self.assert_(char in ['.', '-', '_', ':'] or char.isalnum())
def tmpl_draw_subtitle(self, ln, title, subtitle, guideurl):
"""Draws an html title bar
- 'title' *string* - The name of the titlebar
- 'subtitle' *string* - The header name of the subtitle
- 'guideurl' *string* - The relative url of the guide relative to this section
"""
_ = gettext_set_language(ln)
guidetitle = _("See Guide")
titlebar = """<a name="%s">""" % nmtoken_from_string(title)
titlebar += """ </a>%s <small>""" % subtitle
titlebar += """ [<a title="%s" href="%s/%s">?</a>]</small>""" % (guidetitle, CFG_SITE_URL, guideurl)
return titlebar
def tmpl_draw_subtitle(self, ln, title, subtitle, guideurl):
"""Draws an html title bar
- 'title' *string* - The name of the titlebar
- 'subtitle' *string* - The header name of the subtitle
- 'guideurl' *string* - The relative url of the guide relative to this section
"""
_ = gettext_set_language(ln)
guidetitle = _("See Guide")
titlebar = """<a name="%s">""" % nmtoken_from_string(title)
titlebar += """ </a>%s <small>""" % subtitle
titlebar += """ [<a title="%s" href="%s/%s">?</a>]</small>""" % (guidetitle, CFG_SITE_URL, guideurl)
return titlebar
def prepare_data_cache(self):
"""
"Index" submission names
"""
from invenio.websubmit_dblayer import get_categories_of_doctype
res = run_sql("SELECT sdocname, ldocname FROM sbmDOCTYPE")
# TODO: only consider submissions that are attached to the tree
if CFG_CERN_SITE:
for submission_name, submission_config in CERN_GENSBM_SUBMISSIONS_CONFIG.iteritems():
if not submission_config.has_key('redirect'):
res += (('GENSBM#' + nmtoken_from_string(cgi.escape(submission_name)), submission_name),)
cache = {}
for doctype, submission_name in res:
## categories_and_submission_name = ' '.join(get_categories_of_doctype(doctype)) + \
## ' ' + submission_name
# Add submission name info
if CFG_CERN_SITE and doctype in ('ALIPH', 'BULIS', 'CMSREL', 'BULBN', 'BSA'):
# These submissions are not interesting here
continue
for word in clean_and_split_words_and_stem(submission_name):
if not word.strip():
continue
if not cache.has_key(word):
cache[word] = []
item = (doctype, submission_name, '*')
if not item in cache[word]:
cache[word].append(item)
# Add submission categories info
if CFG_CERN_SITE and doctype in ('CMSPUB', 'CMSCOM', 'CMSCMC',
'ATLPUB', 'ATLCOM', 'ATLCMC',
'LHCBPB', 'LHCPCM', 'LHCBCC'):
# These categories are not interesting here
continue
categories = get_categories_of_doctype(doctype)
for dummy, category, dummy in categories:
for word in clean_and_split_words_and_stem(submission_name + ' ' + category):
if not word.strip():
continue
if not cache.has_key(word):
cache[word] = []
item = (doctype, "%s (%s)" % (category, submission_name), category)
if not item in cache[word]:
cache[word].append(item)
return cache
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 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)