Page MenuHomec4science
Files F69172156

search_engine_summarizer.py
No OneTemporary
Actions

File Metadata

Created
Sun, Jun 30, 14:07

search_engine_summarizer.py
View Options

# -*- coding: utf-8 -*-
## This file is part of CDS Invenio.
## Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007, 2008 CERN.
##
## CDS 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.
##
## CDS 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 CDS 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.bibrank_citation_searcher import get_cited_by_list
import search_engine
import invenio.template
websearch_templates = invenio.template.load('websearch')
## CFG_CITESUMMARY_COLLECTIONS -- how do we break down cite summary
## results according to collections?
if CFG_INSPIRE_SITE:
CFG_CITESUMMARY_COLLECTIONS = [['All papers', 'collection:citeable'],
['Published only', 'collection:citeable 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 summarize_records(recids, of, ln, searchpattern="", searchfield="", req=None):
"""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.
"""
import search_engine
if of == 'hcs':
# this is HTML cite summary
# 1) hcs prologue:
d_recids = {}
d_total_recs = {}
for coll, colldef in CFG_CITESUMMARY_COLLECTIONS:
if not colldef:
d_recids[coll] = recids
else:
d_recids[coll] = recids & search_engine.search_pattern(p=colldef)
d_total_recs[coll] = len(d_recids[coll])
req.write(websearch_templates.tmpl_citesummary_prologue(d_total_recs, CFG_CITESUMMARY_COLLECTIONS, searchpattern, searchfield, ln))
# 2) hcs overview:
d_recid_citers = {}
d_total_cites = {}
d_avg_cites = {}
for coll, colldef in CFG_CITESUMMARY_COLLECTIONS:
d_total_cites[coll] = 0
d_avg_cites[coll] = 0
d_recid_citers[coll] = get_cited_by_list(d_recids[coll])
for recid, lciters in d_recid_citers[coll]:
if lciters:
d_total_cites[coll] += len(lciters)
if d_total_cites[coll] != 0:
d_avg_cites[coll] = d_total_cites[coll] * 1.0 / d_total_recs[coll]
req.write(websearch_templates.tmpl_citesummary_overview(d_total_cites, d_avg_cites, CFG_CITESUMMARY_COLLECTIONS, ln))
# 3) hcs break down by fame:
for low, high, fame in CFG_CITESUMMARY_FAME_THRESHOLDS:
d_cites = {}
for coll, colldef in CFG_CITESUMMARY_COLLECTIONS:
d_cites[coll] = 0
for recid, lciters in d_recid_citers[coll]:
numcites = 0
if lciters:
numcites = len(lciters)
if numcites >= low and numcites <= high:
d_cites[coll] += 1
req.write(websearch_templates.tmpl_citesummary_breakdown_by_fame(d_cites, low, high, fame, CFG_CITESUMMARY_COLLECTIONS, searchpattern, searchfield, ln))
# 4) hcs epilogue:
req.write(websearch_templates.tmpl_citesummary_epilogue(ln))
return ''
elif of == 'xcs':
# this is XML cite summary
citedbylist = get_cited_by_list(recids)
return print_citation_summary_xml(citedbylist)
#for citation summary, code xcs/hcs (unless changed)
def print_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 low, high, name in CFG_CITESUMMARY_FAME_THRESHOLDS:
#get the name, print the value
if reciddict.has_key(name):
recs = reciddict[name]
outp += "<citationclass>"+name
outp += "<records>"+str(recs)+"</records>"
outp += "</citationclass>\n"
outp = outp + "</citationsummary>"
#req.write(outp)
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 reciddict.has_key(name):
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