From f2eb10a42b8772c082875296f2d426ecd85e7770 Mon Sep 17 00:00:00 2001
From: Ken Kellner <ken@kenkellner.com>
Date: Fri, 20 Apr 2018 12:01:46 -0400
Subject: Support for Annual Reviews; tables don't work

---
 article_epub/publishers/__init__.py      |   1 +
 article_epub/publishers/annualreviews.py | 107 +++++++++++++++++++++++++++++++
 2 files changed, 108 insertions(+)
 create mode 100644 article_epub/publishers/annualreviews.py

diff --git a/article_epub/publishers/__init__.py b/article_epub/publishers/__init__.py
index 2fa5e09..d153124 100644
--- a/article_epub/publishers/__init__.py
+++ b/article_epub/publishers/__init__.py
@@ -8,3 +8,4 @@ from article_epub.publishers.nih import NIH
 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
diff --git a/article_epub/publishers/annualreviews.py b/article_epub/publishers/annualreviews.py
new file mode 100644
index 0000000..d5ebf01
--- /dev/null
+++ b/article_epub/publishers/annualreviews.py
@@ -0,0 +1,107 @@
+from article_epub.publisher import Publisher, register_publisher
+import sys
+import copy
+import requests
+
+class AnnualReviews(Publisher):
+    """Class for Annual Reviews articles"""
+
+    name = "Annual Reviews"
+    domains = ["annualreviews.org"]
+
+    def get_final_url(self):
+        pass
+   
+    def check_fulltext(self):
+        test = self.soup.find_all('div',class_='hlFld-Fulltext')
+        if len(test) < 1:
+            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',{'scheme':'doi'})['content'])
+            
+    def get_abstract(self):
+        """Get article abstract"""
+        abstract_raw = self.soup.find('div',class_='hlFld-Abstract')
+        try:
+            abstract_raw.find('iframe').decompose()
+        except:
+            pass
+        try:
+            abstract_raw.find('span',class_='title').decompose()
+        except:
+            pass
+        
+        self.abstract = str(abstract_raw)
+
+    def get_keywords(self):
+        """Get article keywords"""
+        self.keywords = []
+        try:
+            keywords_raw = self.soup.find('div',class_='hlFld-KeywordText') \
+                .find_all('a')
+            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_='hlFld-Fulltext'))
+
+        try:
+            body_raw.find('div',class_='lit-cited').decompose()
+            body_raw.find('div',{'id':'citations'}).decompose()
+        except:
+            pass
+
+        for i in body_raw.find_all('a',class_='scrollRef'):
+            i['href'] = '#'+i['refid']+'ref'
+
+        for i in body_raw.find_all('a',class_='scrollFig'):
+            try:
+                i['href'] = '#'+i['data-figindex']
+            except:
+                pass
+
+        for i in body_raw.find_all('figure'):
+            oldlink = 'https://www.annualreviews.org'+ \
+                    i.find('a').find('img')['src']
+            newlink = oldlink.replace('small','medium')
+            response = requests.head(newlink).headers['content-type']
+            if 'image' in response:
+                i.find('a').find('img')['src'] = newlink
+            else:
+                newlinkjpeg = newlink.replace('.gif','.jpeg')
+                i.find('a').find('img')['src'] = newlinkjpeg
+        
+        for i in body_raw.find_all('span',class_='NLM_inline-graphic'):
+            link = 'https://www.annualreviews.org'+\
+                    i.find('img')['src']
+            i.find('img')['src'] = link
+
+        for i in body_raw.find_all('div',class_='equation'):
+            link = 'https://www.annualreviews.org'+i.find('img')['src']
+            i.find('img')['src'] = link
+
+        self.body = str(body_raw)
+    
+    def get_references(self):
+        """Get references list"""
+        references = self.soup.find('div',class_='lit-cited')
+
+        for i in references.find_all('ul',class_='off-links'):
+            i.decompose()
+        for i in references.find_all('div',class_='article-locations'):
+            i.decompose()
+        for i in references.find_all('a',class_='ar-modal-link citation'):
+            i.decompose()
+        for i in references.find_all('div',class_='citation-content'):
+            i.decompose()
+
+        self.references = str(references)
+
+register_publisher(AnnualReviews)
-- 
cgit v1.2.3