Homec4science
Diffusion invenio-infoscience (maint-2.0) Browse

R3600/invenio/legacy/bibrank/downloads_grapher.py
dd44fdec50bbmaint-2.0

Last Change
Raw File

downloads_grapher.py
View Options

# -*- coding: utf-8 -*-
##
## This file is part of Invenio.
## Copyright (C) 2005, 2006, 2007, 2008, 2009, 2010, 2011 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.
__revision__ = "$Id$"
import string
import os
import time
import calendar
from invenio.config import CFG_SITE_URL, CFG_SITE_LANG, CFG_BIBRANK_SHOW_DOWNLOAD_GRAPHS, CFG_BIBRANK_SHOW_DOWNLOAD_GRAPHS_CLIENT_IP_DISTRIBUTION, CFG_WEBDIR
from invenio.base.i18n import gettext_set_language
from intbitset import intbitset
from invenio.legacy.dbquery import run_sql
from invenio.legacy.bibrank.downloads_indexer import database_tuples_to_single_list
from invenio.legacy.bibrank.grapher import *
color_line_list = ['9', '19', '10', '15', '21', '18']
cfg_id_bibdoc_id_bibrec = 5
def create_download_history_graph_and_box(id_bibrec, ln=CFG_SITE_LANG):
"""Create graph with citation history for record ID_BIBREC (into a
temporary file) and return HTML box refering to that image.
Called by Detailed record pages.
Notes:
if id_bibdoc=0 : its an oustide-stored document and it has no id_bibdoc --> only one line
if len(id_bibdocs) <= cfg_id_bibdoc_id_bibrec draw one line per id_bibdoc
if len(id_bibdocs) > cfg_id_bibdoc_id_bibrec draw only one line which hold simultaneously the downloads for all id_bibdoc
Each time this function is called, all the images older than 10 minutes are deleted.
"""
_ = gettext_set_language(ln)
out = ""
# Prepare downloads history graph:
if CFG_BIBRANK_SHOW_DOWNLOAD_GRAPHS:
html_content = ""
# remove images older than 10 minutes
remove_old_img("download")
# download count graph
id_bibdocs = intbitset(run_sql("select distinct id_bibdoc from rnkDOWNLOADS where id_bibrec=%s", (id_bibrec, )))
id_existing_bibdocs = intbitset(run_sql("SELECT id_bibdoc FROM bibrec_bibdoc JOIN bibdoc ON id_bibdoc=id WHERE id_bibrec=%s AND status<>'DELETED'", (id_bibrec, )))
## FIXME: when bibdocs are deleted we loose the stats. What shall we do with them?
id_bibdocs &= id_existing_bibdocs
history_analysis_results = ()
if not id_bibdocs:
pass
elif len(id_bibdocs) <= cfg_id_bibdoc_id_bibrec and 0 not in id_bibdocs:
history_analysis_results = draw_downloads_statistics(id_bibrec, list(id_bibdocs))
else:
history_analysis_results = draw_downloads_statistics(id_bibrec, [])
if history_analysis_results and history_analysis_results[0]:
if CFG_BIBRANK_SHOW_DOWNLOAD_GRAPHS == 2:
graph_file_history = CFG_WEBDIR + "/img/" + history_analysis_results[0]
html_content += """<tr><td valign=center align=center>%s</td>""" % open(graph_file_history).read()
else:#gnuplot
graph_file_history = CFG_SITE_URL + "/img/" + history_analysis_results[0]
html_content += """<tr><td valign=center align=center><img src='%s'/></td>""" % graph_file_history
file_to_close_history = history_analysis_results[1]
if file_to_close_history :
if os.path.exists(file_to_close_history):
os.unlink(file_to_close_history)
if html_content != "":
out += """<table border="0" cellspacing="1" cellpadding="1">"""
out += html_content + "</table>"
if CFG_BIBRANK_SHOW_DOWNLOAD_GRAPHS_CLIENT_IP_DISTRIBUTION:
# do we show also user IP repartition?
html_content = ""
remove_old_img("download")
#Users analysis graph
ips = database_tuples_to_single_list(run_sql("select client_host from rnkDOWNLOADS where id_bibrec=%s;" % id_bibrec))
if ips:
users_analysis_results = create_users_analysis_graph(id_bibrec, ips)
if users_analysis_results[0]:
file_to_close_users = users_analysis_results[1]
if CFG_BIBRANK_SHOW_DOWNLOAD_GRAPHS_CLIENT_IP_DISTRIBUTION == 1:
html_content += """<tr><td valign=center align=center><img src='%s/img/%s' align="center" alt=""></td>""" % (CFG_SITE_URL, users_analysis_results[0])
elif CFG_BIBRANK_SHOW_DOWNLOAD_GRAPHS_CLIENT_IP_DISTRIBUTION == 2:
html_content += """<tr><td valign=center align=center>%s</td>""" % open(CFG_WEBDIR + "/img/" + users_analysis_results[0]).read()
if file_to_close_users:
if os.path.exists(file_to_close_users):
os.unlink(file_to_close_users)
if html_content != "":
out += """<br/><br/><table><tr><td class="blocknote">
%s</td></tr><tr><td>
<table border="0" cellspacing="1" cellpadding="1">""" % _("Download user distribution:")
out += html_content
out += "</table></td></tr></table>"
# return html code used by get_file or search_engine
return out
def draw_downloads_statistics(id_bibrec, id_bibdoc_list):
"""Create a graph of download history using a temporary file to store datas
and a new png file for each id_bibrec to store the image of the graph which will
be referenced by html code."""
intervals = []
#used to name the different curves when len(id_bibdoc_list)>1
docfile_name_list = []
#used to name the uniquecurve when len(id_bibdoc_list)=0 or > cfg_id_bibdoc_id_bibrec
record_name = ""
record_name_query = run_sql("select value from bibrec_bib24x,bib24x where id_bibrec=%s and id_bibxxx=id;" % id_bibrec)
if record_name_query:
record_name = record_name_query[0][0]
#list of lists of tuples: [[("09/2004",4),..],[(..,..)]..]
#Each list element of the list is represented by a curve
#each elem of each list is a point on the graph
coordinates_list = []
#If the document is not stored in internally it has id_bibrec 0 and no creation date
#In this case the beginning date is the first time the document has been downloaded
local_time = time.localtime()
local_month = local_time.tm_mon
local_year = local_time.tm_year
creation_date_res = run_sql("""SELECT DATE_FORMAT(creation_date,"%%Y-%%m") FROM bibrec WHERE id=%s;""" % id_bibrec)
if creation_date_res == ():
creation_date_res = run_sql("""SELECT DATE_FORMAT(MIN(download_time),"%%Y-%%m") FROM rnkDOWNLOADS where id_bibrec=%s;""" % id_bibrec)
if creation_date_res == (('0000-00',),):
creation_date_year = local_year - 1
creation_date_month = local_month
else :
creation_date_year, creation_date_month = string.split(creation_date_res[0][0], "-")
creation_date_year = int(creation_date_year)
creation_date_month = int(creation_date_month)
#create intervals and corresponding values
res = create_tic_intervals(local_year, local_month, creation_date_year, creation_date_month)
intervals = res[1]
tic_list = res[0]
if id_bibdoc_list == []:
coordinates_list.append(create_list_tuple_data(intervals, id_bibrec))
docfile_name_list = record_name
else :
for i in range(len(id_bibdoc_list)):
datas = create_list_tuple_data(intervals, id_bibrec, id_bibdoc_query_addition="and id_bibdoc=%s" % id_bibdoc_list[i])
coordinates_list.append(datas)
docname = run_sql("select docname from bibrec_bibdoc where id_bibdoc=%s and id_bibrec=%s;" % (id_bibdoc_list[i], id_bibrec))
docfile_name_list.append(docname[0][0])
#In case of multiple id_bibdocs datas_max will be used to draw a line which is the total of the others lines
if not (len(intervals)==1 or len(id_bibdoc_list)==1):
datas_max = create_list_tuple_total(intervals, coordinates_list)
coordinates_list.append(datas_max)
#write coordinates_list in a temporary file
result2 = write_coordinates_in_tmp_file(coordinates_list)
fname = result2[0]
y_max = result2[1]
#Use create the graph from the temporary file
return create_temporary_image(id_bibrec, 'download_history', fname, ' ', 'Times downloaded', [0, 0], y_max, id_bibdoc_list, docfile_name_list, tic_list)
def create_list_tuple_data(intervals, id_bibrec, id_bibdoc_query_addition=""):
"""-Return a list of tuple of the form [('10/2004',3),(..)] used to plot graph
Where 3 is the number of downloads between 01/10/2004 and 31/10/2004"""
list_tuple = []
for elem in intervals:
main_date_end = string.split(elem[1], '/')
end_of_month_end = calendar.monthrange(int(main_date_end[1]), int(main_date_end[0]))[1]
s0 = string.split(elem[0], "/")
s1 = string.split(elem[1], "/")
elem0 = s0[1] + "-" + s0[0]
elem1 = s1[1] + "-" + s1[0]
date1 = "%s%s" % (elem0, "-01 00:00:00")
date2 = "%s%s" % (elem1, "-%s 00:00:00" % str(end_of_month_end))
sql_query = "select count(*) from rnkDOWNLOADS where id_bibrec=%s %s and download_time>='%s' and download_time<'%s';" % (id_bibrec, id_bibdoc_query_addition, date1, date2)
res = run_sql(sql_query)[0][0]
list_tuple.append((elem[0], res))
#list_tuple = sort_list_tuple_by_date(list_tuple)
return (list_tuple)
def sort_list_tuple_by_date(list_tuple):
"""Sort a list of tuple of the forme ("09/2004", 3)according to the
year of the first element of the tuple"""
list_tuple.sort(lambda x, y: (cmp(string.split(x[0], '/')[1],
string.split(y[0], '/')[1])))
return list_tuple
def create_list_tuple_total(intervals, list_data):
"""In the case of multiple id_bibdocs, a last paragraph is added
at the end to show the global evolution of the record"""
list_tuple = []
if len(intervals)==1:
res = 0
for j in range(len(list_data)):
res += list_data[j][1]
list_tuple.append((intervals[0][0], res))
else :
for i in range(len(intervals)):
res = 0
for j in range(len(list_data)):
res += list_data[j][i][1]
list_tuple.append((intervals[i][0], res))
#list_tuple = sort_list_tuple_by_date(list_tuple)
return list_tuple
def create_tic_intervals(local_year, local_month, creation_date_year, creation_date_month):
"""Create intervals since document creation date until now
Return a list of the tics for the graph of the form ["04/2004","05/2004"), ...]
And a list of tuple(each tuple stands for a period) of the form [("04/2004", "04/2004"),.]
to compute the number of downloads in each period
For the very short periods some tics and tuples are added to make sure that
at least two dates are returned. Useful for drawing graphs.
"""
# okay, off we go
tic_list = []
interval_list = []
original_date = (creation_date_month, creation_date_year)
while (creation_date_year, creation_date_month) <= (local_year, local_month) and creation_date_month <= 12:
date_elem = "%s/%s" % (creation_date_month, creation_date_year)
tic_list.append(date_elem)
interval_list.append((date_elem, date_elem))
if creation_date_month != 12:
creation_date_month = creation_date_month+1
else :
creation_date_year = creation_date_year+1
creation_date_month = 1
next_period = (creation_date_month, creation_date_year)
#additional periods for the short period
if len(interval_list) <= 2:
period_before = "%s/%s" % (sub_month(original_date[0], original_date[1]))
period_after = "%s/%s" % next_period
interval_list.insert(0, (period_before, period_before))
interval_list.append((period_after, period_after))
tic_list.insert(0, period_before)
tic_list.append(period_after)
return (tic_list, interval_list)
def add_month(month, year):
"""Add a month and increment the year if necessary"""
if month == 12:
month = 1
year += 1
else :
month += 1
return (month, year)
def sub_month(month, year):
"""Add a month and decrease the year if necessary"""
if month == 1:
month = 12
year = year -1
else :
month -= 1
return (month, year)
def create_users_analysis_graph(id_bibrec, ips):
"""For a given id_bibrec, classify cern users and other users
Draw a percentage graphic reprentation"""
cern_users = 0
other_users = 0
coordinates_list = []
#compute users repartition
for i in range(len(ips)):
if 2307522817 <= ips[i] <= 2307588095 or 2156724481 <= ips[i] <= 2156789759:
cern_users += 1
else :
other_users += 1
tot = float(cern_users+other_users)
#prepare coordinates list
coordinates_list.append((1, str(float(cern_users)/tot*100)))
coordinates_list.append((3, str(float(other_users)/tot*100)))
#write coordinates in a temporary file
result2 = write_coordinates_in_tmp_file([coordinates_list])
#result2 example: [/path/to-invenio/var/www/img/tmpeC9GP5,'100.0']
#the file contains e.g.
#1 100.0
#3 0.0
#plot the graph
return create_temporary_image(id_bibrec, 'download_users', result2[0], ' ', 'User distribution', (0, 0), result2[1], [], [], [1, 3])
c4science · Help