def extra_request_info(request):
"""
Add some extra useful stuff into the RequestContext.
"""
date_boundaries = fulltext_range()
return {
'BASE_URL': settings.BASE_URL,
'city_form': CityForm(),
'fulltext_enddate': date_boundaries[1],
'fulltext_startdate': date_boundaries[0],
'omniture_url': settings.OMNITURE_SCRIPT if "OMNITURE_SCRIPT" in dir(settings) else None,
'project_name': settings.PROJECT_NAME if "PROJECT_NAME" in dir(settings) else None,
'sharetool_url': settings.SHARETOOL_URL if "SHARETOOL_URL" in dir(settings) else None,
'site_title': settings.SITE_TITLE if "SITE_TITLE" in dir(settings) else None,
}
def __init__(self, *args, **kwargs):
super(SearchPagesFormBase, self).__init__(*args, **kwargs)
self.titles, self.states = _titles_states()
fulltextStartYear, fulltextEndYear = fulltext_range()
self.years = [(year, year) for year in range(fulltextStartYear, fulltextEndYear + 1)]
self.fulltextStartYear = fulltextStartYear
self.fulltextEndYear = fulltextEndYear
self.fields["state"].choices = self.states
self.fields["date1"].choices = self.years
self.fields["date1"].initial = fulltextStartYear
self.fields["date2"].choices = self.years
self.fields["date2"].initial = fulltextEndYear
def _solrize_date(date, date_type=''):
"""
Takes a date string like 2018/01/01 and returns an
integer suitable for querying the date field in a solr document.
"""
solr_date = "*"
if date:
date = date.strip()
start_year, end_year = fulltext_range()
if date_type == 'start' and date == str(start_year) + '-01-01':
return '*'
elif date_type == 'end' and date == str(end_year) + '-12-31':
return '*'
# 1900-01-01 -> 19000101
match = re.match(r'(\d{4})-(\d{2})-(\d{2})', date)
if match:
y, m, d = match.groups()
if y and m and d:
solr_date = y + m + d
return solr_date
def page_search(d):
"""
Pass in form data for a given page search, and get back
a corresponding solr query.
"""
q = ['+type:page']
simple_fields = ['city', 'county', 'frequency', 'state', 'lccn']
for field in simple_fields:
if d.get(field, None):
q.append(query_join(d.getlist(field), field))
ocrs = ['ocr_%s' % l for l in settings.SOLR_LANGUAGES]
lang_req = d.get('language', None)
language = models.Language.objects.get(name=lang_req) if lang_req else None
lang = language.code if language else None
if language:
q.append('+language:%s' % language.name)
ocr_lang = 'ocr_' + lang if lang else 'ocr'
if d.get('ortext', None):
q.append('+((' +
query_join(_solr_escape(d['ortext']).split(' '), "ocr"))
if lang:
q.append(
' AND ' +
query_join(_solr_escape(d['ortext']).split(' '), ocr_lang))
q.append(
') OR ' +
query_join(_solr_escape(d['ortext']).split(' '), ocr_lang))
else:
q.append(')')
for ocr in ocrs:
q.append('OR ' +
query_join(_solr_escape(d['ortext']).split(' '), ocr))
q.append(')')
if d.get('andtext', None):
q.append('+((' + query_join(
_solr_escape(d['andtext']).split(' '), "ocr", and_clause=True))
if lang:
q.append('AND ' + query_join(_solr_escape(d['andtext']).split(' '),
ocr_lang,
and_clause=True))
q.append(') OR ' +
query_join(_solr_escape(d['andtext']).split(' '),
ocr_lang,
and_clause=True))
else:
q.append(')')
for ocr in ocrs:
q.append('OR ' +
query_join(_solr_escape(d['andtext']).split(' '),
ocr,
and_clause=True))
q.append(')')
if d.get('phrasetext', None):
phrase = _solr_escape(d['phrasetext'])
q.append('+((' + 'ocr' + ':"%s"^10000' % (phrase))
if lang:
q.append('AND ocr_' + lang + ':"%s"' % (phrase))
q.append(') OR ocr_' + lang + ':"%s"' % (phrase))
else:
q.append(')')
for ocr in ocrs:
q.append('OR ' + ocr + ':"%s"' % (phrase))
q.append(')')
if d.get('proxtext', None):
distance = d.get('proxdistance', PROX_DISTANCE_DEFAULT)
prox = _solr_escape(d['proxtext'])
q.append('+((' + 'ocr' + ':("%s"~%s)^10000' % (prox, distance))
if lang:
q.append('AND ocr_' + lang + ':"%s"~%s' % (prox, distance))
q.append(') OR ocr_' + lang + ':"%s"~%s' % (prox, distance))
else:
q.append(')')
for ocr in ocrs:
q.append('OR ' + ocr + ':"%s"~%s' % (prox, distance))
q.append(')')
if d.get('sequence', None):
q.append('+sequence:"%s"' % d['sequence'])
if d.get('issue_date', None):
q.append('+month:%d +day:%d' %
(int(d['date_month']), int(d['date_day'])))
# yearRange supercedes date1 and date2
year1, year2 = None, None
year_range = d.get('yearRange', None)
if year_range:
split = year_range.split("-")
if len(split) == 2:
year1 = int(split[0])
year2 = int(split[1])
else:
year1 = int(split[0])
year2 = int(split[0])
q.append('+year:[%d TO %d]' % (year1, year2))
else:
date_boundaries = fulltext_range()
date1 = d.get('date1', None)
date2 = d.get('date2', None)
if date1 or date2:
# do NOT apply year min / max to solr query
# do apply it to facets since they require a specific begin / end year
d1 = _solrize_date(str(date1), 'start')
d2 = _solrize_date(str(date2), 'end')
q.append('+date:[%s TO %s]' % (d1, d2))
year1 = date_boundaries[0] if d1 == "*" else int(str(d1)[:4])
year2 = date_boundaries[1] if d2 == "*" else int(str(d2)[:4])
else:
# do not pass any query parameters to solr if no date requested
# but do set the year range for faceting purposes
year1 = date_boundaries[0]
year2 = date_boundaries[1]
# choose a facet range gap such that the number of date ranges returned
# is <= 10. These would be used to populate a select dropdown on search
# results page.
gap = max(1, int(math.ceil((year2 - year1) // 10)))
# increment year range end by 1 to be inclusive
facet_params = {
'facet': 'true',
'facet_field': [
'city',
'county',
'frequency',
'language',
'state',
],
'facet_range': 'year',
'f_year_facet_range_start': year1,
'f_year_facet_range_end': year2 + 1,
'f_year_facet_range_gap': gap,
'facet_mincount': 1
}
return ' '.join(q), facet_params
def search_pages_results(request, view_type='gallery'):
page_title = "Search Results"
paginator = search_pages_paginator(request)
q = paginator.query
try:
page = paginator.page(paginator._cur_page)
except InvalidPage:
url = urls.reverse('openoni_search_pages_results')
# Set the page to the first page
q['page'] = 1
return HttpResponseRedirect('%s?%s' % (url, q.urlencode()))
start = page.start_index()
end = page.end_index()
# figure out the next page number
query = request.GET.copy()
if page.has_next():
query['page'] = paginator._cur_page + 1
next_url = '?' + query.urlencode()
# and the previous page number
if page.has_previous():
query['page'] = paginator._cur_page - 1
previous_url = '?' + query.urlencode()
rows = query.get("rows", "20")
sort = query.get("sort", default="relevance")
seq_check = "checked" if query.get("sequence", "0") == "1" else ""
crumbs = list(settings.BASE_CRUMBS)
host = request.get_host()
format = request.GET.get('format', None)
if format == 'atom':
feed_url = settings.BASE_URL + request.get_full_path()
updated = rfc3339(timezone.now())
return render(request,
'search/search_pages_results.xml',
locals(),
content_type='application/atom+xml')
elif format == 'json':
results = {
'startIndex': start,
'endIndex': end,
'totalItems': paginator.count,
'itemsPerPage': rows,
'items': [p.solr_doc for p in page.object_list],
}
for i in results['items']:
i['url'] = settings.BASE_URL + i['id'].rstrip('/') + '.json'
json_text = json.dumps(results, indent=2)
# jsonp?
if request.GET.get('callback') is not None:
json_text = "%s(%s);" % (request.GET.get('callback'), json_text)
return HttpResponse(json_text, content_type='application/json')
page_range_short = list(_page_range_short(paginator, page))
# copy the current request query without the page and sort
# query params so we can construct links with it in the template
q = request.GET.copy()
for i in ('page', 'sort'):
if i in q:
q.pop(i)
q = q.urlencode()
# get an pseudo english version of the query
english_search = paginator.englishify()
form = forms.SearchResultsForm({"rows": rows, "sort": sort})
if view_type == "list":
template = "search/search_pages_results_list.html"
else:
template = "search/search_pages_results.html"
page_list = []
lccns = query.getlist("lccn")
titles = []
for lccn in lccns:
name = str(models.Title.objects.get(lccn=lccn))
titles.append({
'abbrev': name[:24] + '...' if len(name) > 24 else name,
'lccn': lccn,
'name': name,
})
for count in range(len(page.object_list)):
page_list.append((count + start, page.object_list[count]))
start_year, end_year = fulltext_range()
searching_all_dates = False
if request.GET.get('date1') and request.GET.get('date2'):
if request.GET.get('date1') == str(start_year) + '-01-01':
if request.GET.get('date2') == str(end_year) + '-12-31':
searching_all_dates = True
return render(request, template, locals())
