Page MenuHomec4science
Files F90983702

summarizer.py
No OneTemporary
Actions

File Metadata

Created
Wed, Nov 6, 15:54

summarizer.py
View Options

# -*- coding: utf-8 -*-
## This file is part of Invenio.
## Copyright (C) 2008, 2010, 2011, 2012 CERN.
##
## Invenio is free software; you can redistribute it and/or
## modify it under the terms of the GNU General Public License as
## published by the Free Software Foundation; either version 2 of the
## License, or (at your option) any later version.
##
## Invenio is distributed in the hope that it will be useful, but
## WITHOUT ANY WARRANTY; without even the implied warranty of
## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
## General Public License for more details.
##
## You should have received a copy of the GNU General Public License
## along with Invenio; if not, write to the Free Software Foundation, Inc.,
## 59 Temple Place, Suite 330, Boston, MA 02111-1307, USA.
"""
Search Engine Summarizer, producing summary formats such as citesummary.
The main API is summarize_records().
"""
__lastupdated__ = """$Date$"""
__revision__ = "$Id$"
from invenio.config import CFG_INSPIRE_SITE
from invenio.legacy.bibrank.citation_searcher import get_cited_by_list
from invenio.legacy.bibrank.selfcites_indexer import get_self_citations_count
from StringIO import StringIO
from invenio.legacy.search_engine import search_pattern, perform_request_search
from invenio.intbitset import intbitset
import invenio.legacy.template
websearch_templates = invenio.legacy.template.load('websearch')
## CFG_CITESUMMARY_COLLECTIONS -- how do we break down cite summary
## results according to collections?
if CFG_INSPIRE_SITE:
CFG_CITESUMMARY_COLLECTIONS = [['Citeable papers', 'collection:citeable'],
['Published only', 'collection:published']]
else:
CFG_CITESUMMARY_COLLECTIONS = [['All papers', ''],
['Published only', 'collection:article']]
## CFG_CITESUMMARY_FAME_THRESHOLDS -- how do we break down cite
## summary results into famous and less famous paper groups?
CFG_CITESUMMARY_FAME_THRESHOLDS = [
(500, 1000000, 'Renowned papers (500+)'),
(250, 499, 'Famous papers (250-499)'),
(100, 249, 'Very well-known papers (100-249)'),
(50, 99, 'Well-known papers (50-99)'),
(10, 49, 'Known papers (10-49)'),
(1, 9, 'Less known papers (1-9)'),
(0, 0, 'Unknown papers (0)')
]
def render_self_citations(d_recids, ln):
"""Render the html displayed for self-citations"""
d_total_cites = {}
d_avg_cites = {}
for coll, recids in d_recids.iteritems():
if recids:
d_total_cites[coll] = get_self_citations_count(recids)
d_avg_cites[coll] = d_total_cites[coll] * 1.0 / len(recids)
else:
d_total_cites[coll] = 0
d_avg_cites[coll] = 0
return websearch_templates.tmpl_citesummary_minus_self_cites(d_total_cites,
d_avg_cites,
ln)
def render_citations_breakdown(req, ln, collections, d_recid_citers,
search_patterns, searchfield):
"Render citations break down by fame"
header = websearch_templates.tmpl_citesummary_breakdown_header(ln)
req.write(header)
for low, high, fame in CFG_CITESUMMARY_FAME_THRESHOLDS:
d_cites = {}
for coll, citers in d_recid_citers.iteritems():
d_cites[coll] = 0
for dummy, lciters in citers:
numcites = 0
if lciters:
numcites = len(lciters)
if numcites >= low and numcites <= high:
d_cites[coll] += 1
fame_info = websearch_templates.tmpl_citesummary_breakdown_by_fame(
d_cites, low, high, fame, collections,
search_patterns, searchfield, ln)
req.write(fame_info)
def render_citation_summary(req, ln, recids, collections, searchpattern,
searchfield):
title = websearch_templates.tmpl_citesummary_title(ln)
req.write(title)
d_recids = get_recids(recids, collections)
d_recid_citers = get_citers(d_recids)
search_patterns = dict([(coll, searchpattern) \
for coll, dummy in collections])
render_citesummary_prologue(req,
ln,
recids,
collections,
search_patterns,
searchfield,
d_recids)
render_citesummary_overview(req,
ln,
collections,
d_recids,
d_recid_citers)
render_citations_breakdown(req,
ln,
collections,
d_recid_citers,
search_patterns,
searchfield)
render_h_index(req, ln, collections, d_recid_citers)
eplilogue = websearch_templates.tmpl_citesummary_epilogue(ln)
req.write(eplilogue)
links = websearch_templates.tmpl_citesummary_more_links(searchpattern, ln)
req.write(links)
def render_extended_citation_summary(req, ln, recids, initial_collections,
searchpattern, searchfield):
title = websearch_templates.tmpl_citesummary2_title(searchpattern, ln)
req.write(title)
d_recids = get_recids(recids, initial_collections)
def coll_self_cites(name):
return name + '<br />excluding self cites'
def coll_not_rpp(name):
return name + '<br />excluding RPP'
# Add self cites sets and "-title:rpp" sets
if CFG_INSPIRE_SITE:
notrpp_searchpattern = searchpattern + ' -title:rpp'
notrpp_recids = intbitset(perform_request_search(p=notrpp_searchpattern))
for coll, coll_recids in list(d_recids.iteritems()):
d_recids[coll_self_cites(coll)] = coll_recids
if CFG_INSPIRE_SITE:
d_recids[coll_not_rpp(coll)] = notrpp_recids & coll_recids
# Add self cites collections
collections = []
search_patterns = {}
for coll, query in initial_collections:
search_patterns[coll] = searchpattern
search_patterns[coll_self_cites(coll)] = searchpattern
if CFG_INSPIRE_SITE:
search_patterns[coll_not_rpp(coll)] = notrpp_searchpattern
collections += [
(coll, query),
(coll_self_cites(coll), query),
(coll_not_rpp(coll), query),
]
else:
collections += [
(coll, query),
(coll_self_cites(coll), query),
]
d_recid_citers = get_citers(d_recids)
for coll, dummy in initial_collections:
d_recid_citers[coll_self_cites(coll)] = [
(recid, range(get_self_citations_count([recid]))) \
for recid in d_recids[coll]
]
render_citesummary_prologue(req,
ln,
recids,
collections,
search_patterns,
searchfield,
d_recids)
render_citesummary_overview(req,
ln,
collections,
d_recids,
d_recid_citers)
render_citations_breakdown(req,
ln,
collections,
d_recid_citers,
search_patterns,
searchfield)
for coll, dummy in initial_collections:
d_recid_citers[coll_self_cites(coll)] = None
render_h_index(req, ln, collections, d_recid_citers)
# 6) hcs epilogue:
eplilogue = websearch_templates.tmpl_citesummary_epilogue(ln)
req.write(eplilogue)
back_link = websearch_templates.tmpl_citesummary_back_link(searchpattern, ln)
req.write(back_link)
def render_citesummary_overview(req, ln, collections, recids, recid_citers):
"""Citations overview: total citations"""
total_cites = {}
avg_cites = {}
for coll, citers in recid_citers.iteritems():
total_cites[coll] = sum([len(lciters) for dummy, lciters in citers])
try:
avg_cites[coll] = float(total_cites[coll]) / len(recids[coll])
except ZeroDivisionError:
avg_cites[coll] = 0
overview = websearch_templates.tmpl_citesummary_overview(collections,
total_cites,
avg_cites,
ln)
req.write(overview)
def get_recids(recids, collections):
"""Compute recids for each column"""
d_recids = {}
for coll, colldef in collections:
if not colldef:
d_recids[coll] = recids
else:
d_recids[coll] = recids & search_pattern(p=colldef)
return d_recids
def get_citers(d_recids):
"""For each recid fetches the list of citing papers"""
d_recid_citers = {}
for coll, recids in d_recids.iteritems():
d_recid_citers[coll] = get_cited_by_list(recids)
return d_recid_citers
def render_citesummary_prologue(req, ln, recids, collections, search_patterns,
searchfield, d_recids):
total_count = len(recids)
citable_recids = recids & search_pattern(p='collection:citeable')
prologue = websearch_templates.tmpl_citesummary_prologue(d_recids,
collections,
search_patterns,
searchfield,
citable_recids,
total_count,
ln)
req.write(prologue)
def render_h_index(req, ln, collections, d_recid_citers):
"Calculate and Render h-hep index"
d_recid_citecount_l = {}
for coll, citers in d_recid_citers.iteritems():
if citers is None:
d_recid_citecount_l[coll] = None
else:
d_recid_citecount_l[coll] = []
for recid, lciters in citers:
d_recid_citecount_l[coll].append((recid, len(lciters)))
d_h_factors = {}
def comparator(x, y):
if x[1] > y[1]:
return -1
elif x[1] == y[1]:
return 0
else:
return 1
for coll, citecount in d_recid_citecount_l.iteritems():
if citecount is None:
d_h_factors[coll] = 'n/a'
else:
d_h_factors[coll] = 0
d_recid_citecount_l[coll].sort(cmp=comparator)
for citecount in citecount:
d_h_factors[coll] += 1
if d_h_factors[coll] > citecount[1]:
d_h_factors[coll] -= 1
break
h_idx = websearch_templates.tmpl_citesummary_h_index(collections,
d_h_factors,
ln)
req.write(h_idx)
def summarize_records(recids, of, ln, searchpattern="", searchfield="",
req=None, collections=CFG_CITESUMMARY_COLLECTIONS):
"""Write summary report for records RECIDS in the format OF in language LN.
SEARCHPATTERN and SEARCHFIELD are search query that led to RECIDS,
for instance p='Smith, Paul' and f='author'. They are used for links.
REQ is the Apache/mod_python request object.
"""
if of == 'xcs':
# this is XML cite summary
citedbylist = get_cited_by_list(recids)
return render_citation_summary_xml(citedbylist)
has_req = req is not None
if not has_req:
req = StringIO()
if of == 'hcs':
renderer = render_citation_summary
else:
renderer = render_extended_citation_summary
renderer(req,
ln,
recids,
collections,
searchpattern,
searchfield)
req.write(websearch_templates.tmpl_citesummary_footer())
if has_req:
return ''
else:
return req.getvalue()
# For citation summary, code xcs/hcs (unless changed)
def render_citation_summary_xml(citedbylist):
"""Prints citation summary in xml."""
alldict = calculate_citations(citedbylist)
avgstr = str(alldict['avgcites'])
totalcites = str(alldict['totalcites'])
# format avg so that it does not span 10 digits
avgstr = avgstr[0:4]
reciddict = alldict['reciddict']
# output formatting
outp = "<citationsummary records=\"" + str(len(citedbylist))
outp += "\" citations=\"" + str(totalcites) + "\">"
for dummy, dummy, name in CFG_CITESUMMARY_FAME_THRESHOLDS:
# get the name, print the value
if name in reciddict:
recs = reciddict[name]
outp += "<citationclass>" + name
outp += "<records>%s</records>" % recs
outp += "</citationclass>\n"
outp = outp + "</citationsummary>"
return outp # just to return something
def calculate_citations(citedbylist):
"""calculates records in classes of citations
defined by thresholds. returns a dictionary that
contains total, avg, records and a dictionary
of threshold names and number corresponding to it"""
totalcites = 0
avgcites = 0
reciddict = {}
for recid, cites in citedbylist:
numcites = 0
if cites:
numcites = len(cites)
totalcites = totalcites + numcites
# take the numbers in CFG_CITESUMMARY_FAME_THRESHOLDS
for low, high, name in CFG_CITESUMMARY_FAME_THRESHOLDS:
if (numcites >= low) and (numcites <= high):
if name in reciddict:
tmp = reciddict[name]
tmp.append(recid)
reciddict[name] = tmp
else:
reciddict[name] = [recid]
if (len(citedbylist) == 0):
avgcites = 0
else:
avgcites = totalcites * 1.0 / len(citedbylist)
# create a dictionary that contains all the values
alldict = {}
alldict['records'] = len(citedbylist)
alldict['totalcites'] = totalcites
alldict['avgcites'] = avgcites
alldict['reciddict'] = reciddict
return alldict

Event Timeline

c4science · Help