From 45e0625bcbc2c7f041b8c5d177c5dcf487f07d26 Mon Sep 17 00:00:00 2001
From: Jules Laplace <julescarbon@gmail.com>
Date: Fri, 14 Dec 2018 02:31:14 +0100
Subject: new reports

---
 scraper/s2-citation-report.py | 37 ++++++++++++++++++++++---------------
 1 file changed, 22 insertions(+), 15 deletions(-)

(limited to 'scraper/s2-citation-report.py')

diff --git a/scraper/s2-citation-report.py b/scraper/s2-citation-report.py
index 526cf778..5c5fae9a 100644
--- a/scraper/s2-citation-report.py
+++ b/scraper/s2-citation-report.py
@@ -5,7 +5,7 @@ import simplejson as json
 import math
 import operator
 import click
-import builder
+#import builder
 from util import *
 
 @click.command()
@@ -14,9 +14,10 @@ def s2_citation_report():
   megapixels = load_megapixels_queries()
   successful_geocodes = {}
   papers = []
-  for fn in glob.iglob('datasets/s2/papers/**/*.json', recursive=True):
-    paper_data = process_paper(fn, addresses, megapixels, successful_geocodes)
-    papers.append(paper_data)
+  for row in megapixels:
+    paper_data = process_paper(row, addresses, successful_geocodes)
+    if paper_data is not None:
+      papers.append(paper_data)
   write_papers_report('reports/report_index.html', 'All Papers', papers, 'title')
   write_papers_report('reports/report_coverage.html', 'Coverage', papers, 'citations_geocoded', reverse=True)
 
@@ -35,6 +36,7 @@ def write_papers_report(fn, title, papers, key, reverse=False):
     sorted_papers.append([
       paper['paperId'],
       paper['key'],
+      paper['name'],
       LinkLine(paper['report_link'], paper['title']),
       LinkLine(paper['pdf_link'], '[pdf]'),
       paper['journal'],
@@ -53,6 +55,7 @@ def write_papers_report(fn, title, papers, key, reverse=False):
   sorted_paper_keys = [
     'Paper ID',
     'Megapixels Key',
+    'Megapixels Name',
     'Report Link',
     'PDF Link',
     'Journal',
@@ -70,7 +73,7 @@ def write_papers_report(fn, title, papers, key, reverse=False):
   ]
   write_report(fn, title=title, keys=sorted_paper_keys, rows=sorted_papers)
 
-def process_paper(fn, addresses, megapixels, success):
+def process_paper(row, addresses, success):
   res = {
     'paperId': '',
     'key': '',
@@ -98,14 +101,18 @@ def process_paper(fn, addresses, megapixels, success):
   doi_count = 0
   address_count = 0
 
+  fn = file_path('papers', row['paper_id'], 'paper.json')
+
   with open(fn, 'r') as f:
     data = json.load(f)
     print('>> {}'.format(data['paperId']))
     paper = load_paper(data['paperId'])
-    if paper.data is None:
+    if paper is None:
       print("Paper missing! {}".format(data['paperId']))
       return
     
+    res['key'] = row['key']
+    res['name'] = row['name']
     res['paperId'] = paper.paper_id
     res['title'] = paper.title
     res['journal'] = paper.journal
@@ -114,9 +121,6 @@ def process_paper(fn, addresses, megapixels, success):
     # res['authors'] = ', '.join(paper.authors)
     # res['citations'] = []
 
-    if res['title'] in megapixels:
-      res['key'] = megapixels[res['title']]['Database Name']
-
     paper_institutions = load_institutions(paper.paper_id)
     paper_address = None
     for inst in sorted(paper_institutions, key=operator.itemgetter(1)):
@@ -126,7 +130,7 @@ def process_paper(fn, addresses, megapixels, success):
         paper_address = addresses.find(institution)
 
     if paper_address:
-      print(paper_address)
+      # print(paper_address)
       res['address'] = paper_address[0]
       res['lat'] = paper_address[3]
       res['lng'] = paper_address[4]
@@ -235,7 +239,10 @@ def process_paper(fn, addresses, megapixels, success):
       f.write('<li>Address: {}</li>'.format(paper_address[2]))
       f.write('<li>Lat/Lng: {}, {}</li>'.format(paper_address[3], paper_address[4]))
     f.write('<li>Year: {}</li>'.format(paper.year))
-    f.write('<li>Coverage: {} / {} citations were located ({} %).</li>'.format(len(geocoded_citations), total_citations, math.floor(len(geocoded_citations) / total_citations * 100)))
+    if total_citations == 0:
+      f.write('<li>Coverage: No citations found!</li>')
+    else:
+      f.write('<li>Coverage: {} / {} citations were located ({} %).</li>'.format(len(geocoded_citations), total_citations, math.floor(len(geocoded_citations) / total_citations * 100)))
     f.write('</ul>')
     f.write('<h3>{}</h3>'.format('Geocoded Citations'))
     write_table(f, keys=None, rows=sorted(display_geocoded_citations, key=operator.itemgetter(0)))
@@ -258,14 +265,14 @@ def process_paper(fn, addresses, megapixels, success):
   return res
 
 def load_megapixels_queries():
-  keys, rows = read_csv('datasets/citations-2018310.csv')
-  lookup = {}
+  keys, rows = read_csv('datasets/citation_lookup.csv')
+  recs = []
   for row in rows:
     rec = {}
     for index, key in enumerate(keys):
       rec[key] = row[index]
-    lookup[rec['Title'].strip()] = rec
-  return lookup
+    recs.append(rec)
+  return recs
 
 def load_institutions(paperId):
   if os.path.exists(file_path('pdf', paperId, 'institutions.json')):
-- 
cgit v1.2.3-70-g09d2