From 8f61b92000c9a03a8d0740f268dc1befca66ad88 Mon Sep 17 00:00:00 2001
From: Ken Kellner <ken@kenkellner.com>
Date: Fri, 20 Apr 2018 14:52:58 -0400
Subject: Add Nature Publishing support; links to tables only

---
 article_epub/publishers/__init__.py |  1 +
 article_epub/publishers/nature.py   | 97 +++++++++++++++++++++++++++++++++++++
 article_epub/utilities.py           |  7 ++-
 3 files changed, 104 insertions(+), 1 deletion(-)
 create mode 100644 article_epub/publishers/nature.py

diff --git a/article_epub/publishers/__init__.py b/article_epub/publishers/__init__.py
index d153124..206b7bc 100644
--- a/article_epub/publishers/__init__.py
+++ b/article_epub/publishers/__init__.py
@@ -9,3 +9,4 @@ from article_epub.publishers.nrc import NRC
 from article_epub.publishers.royalsociety import RoyalSociety
 from article_epub.publishers.tandf import TandF
 from article_epub.publishers.annualreviews import AnnualReviews
+from article_epub.publishers.nature import Nature
diff --git a/article_epub/publishers/nature.py b/article_epub/publishers/nature.py
new file mode 100644
index 0000000..77e1f9c
--- /dev/null
+++ b/article_epub/publishers/nature.py
@@ -0,0 +1,97 @@
+from article_epub.publisher import Publisher, register_publisher
+import sys
+import copy
+
+class Nature(Publisher):
+    """Class for Nature Publishing articles"""
+
+    name = "Nature Publishing"
+    domains = ["nature.com"]
+
+    def check_fulltext(self):
+        test = self.soup.find('a',{'data-track-action':'subscribe'})
+        if test != None:
+            sys.exit('Error: Can\'t access fulltext of article')
+        else:
+            return(True)
+
+    def get_doi(self):
+        if self.doi == None:
+            self.doi = str(self.soup.find('meta',{'name':'DOI'})['content'])
+
+    def get_abstract(self):
+        """Get article abstract"""
+        abstract_raw = self.soup.find('div',{'id':'abstract-section'})
+        try:
+            abstract_raw.find('span').decompose()
+        except:
+            pass
+
+        self.abstract = str(abstract_raw)
+
+    def get_keywords(self):
+        """Get article keywords"""
+        self.keywords = []
+        try:
+            keywords_raw = self.soup.find_all('a',class_='subject-tag-link')
+            for i in keywords_raw:
+                self.keywords.append(i.text)
+        except:
+            pass
+
+    def get_body(self):
+        """Get body of article"""
+        body_raw = copy.copy(self.soup.find('div',class_='article-body'))
+
+        try:
+            body_raw.find('section',{'aria-labelledby':'abstract'}).decompose()
+        except:
+            pass
+
+        try:
+            body_raw.find('section',{'aria-labelledby':'references'}).decompose()
+        except:
+            pass
+
+        try:
+            body_raw.find('section', \
+                    {'aria-labelledby':'author-information'}).decompose()
+            body_raw.find('section',{'aria-labelledby':'rightslink'}) \
+                    .decompose()
+            body_raw.find('section',{'aria-labelledby':'article-comments'}) \
+                    .decompose()
+        except:
+            pass
+
+        for i in body_raw.find_all('span',class_='js-section-title-label'):
+            i.decompose()
+
+        for i in body_raw.find_all('a',{'data-track-action':'view table'}):
+            link = 'https://www.nature.com'+i['href']
+            i['href'] = link
+
+        for i in body_raw.find_all('a',{'data-track-action':'reference anchor'}):
+            part = i['href'].split('#')[1]
+            i['href'] = '#'+part
+
+        for i in body_raw.find_all('a',{'data-track-action':'view figure'}):
+            link = 'https://www.nature.com'+i['href']
+            i['href'] = link
+
+        self.body = str(body_raw)
+
+    def get_references(self):
+        """Get references list"""
+        ref_all = self.soup.find('div',{'id':'references-section'})
+        ref_all.find('span',class_='js-section-title-label').decompose()
+        refs = ref_all.find('ol').find_all('li',recursive=False)
+        for i in refs:
+            try:
+                i.find('span').decompose()
+                i.find('ul',class_='js-ref-links').decompose()
+            except:
+                pass
+
+        self.references = str(ref_all)
+
+register_publisher(Nature)
diff --git a/article_epub/utilities.py b/article_epub/utilities.py
index 1fa1f61..60a2d97 100644
--- a/article_epub/utilities.py
+++ b/article_epub/utilities.py
@@ -14,11 +14,16 @@ def url_from_title(title):
             .find('div',class_='gs_ri').find('a')
         possible_title = result.text
         possible_link = result['href']
+        
+        if possible_title == '':
+            print('No matching link available.')
+            sys.exit('Getting URL from title failed')
+        
         print('Provided title:')
         print(title)
         print('Found following article:')
         print(possible_title)
-        choice = input("Is this correct (y/n)? ")
+        choice = input("\033[0;37m"+"Is this correct (y/n)? "+"\033[00m")
         if choice == 'y':
             return(possible_link)
         else:
-- 
cgit v1.2.3