Improve robustness of springer provider

Adds support for downloading chapters

2 files changed, 40 insertions, 6 deletions

diff --git a/paper2remarkable/providers/springer.py b/paper2remarkable/providers/springer.py
index 5ce2564..dea8bd5 100644
--- a/paper2remarkable/providers/springer.py
+++ b/paper2remarkable/providers/springer.py
@@ -10,10 +10,12 @@ Copyright: 2019, G.J.J. van den Burg
 
 import re
 import urllib
+import requests
 
 from ._base import Provider
 from ._info import Informer
 from ..exceptions import URLResolutionError
+from ..utils import HEADERS
 
 
 class SpringerInformer(Informer):
@@ -26,24 +28,49 @@ class SpringerInformer(Informer):
 
 class Springer(Provider):
 
-    re_abs = "https?:\/\/link.springer.com\/article\/10\.\d{4}\/[a-z0-9\-]+"
-    re_pdf = "https?:\/\/link\.springer\.com\/content\/pdf\/10\.\d{4}(%2F|\/)[a-z0-9\-]+\.pdf"
+    re_abs_1 = "https?:\/\/link.springer.com\/article\/10\.\d{4}\/[a-z0-9\-]+"
+    re_abs_2 = "https?:\/\/link.springer.com\/chapter\/10\.\d{4}\/[a-z0-9\-]+"
+    re_pdf = "https?:\/\/link\.springer\.com\/content\/pdf\/10\.\d{4}(%2F|\/)[a-z0-9\-\_]+\.pdf"
 
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
         self.informer = SpringerInformer()
 
+    def _get_abs_url(self, pdf_url):
+        article_url = pdf_url.replace("content/pdf", "article")[: -len(".pdf")]
+        req = requests.head(
+            article_url, headers=HEADERS, cookies=self.cookiejar
+        )
+        if req.status_code == 200:
+            return article_url
+
+        chapter_url = pdf_url.replace("content/pdf", "chapter")[: -len(".pdf")]
+        req = requests.head(
+            chapter_url, headers=HEADERS, cookies=self.cookiejar
+        )
+        if req.status_code == 200:
+            return chapter_url
+
+        raise URLResolutionError("Springer", pdf_url)
+
     def get_abs_pdf_urls(self, url):
         """ Get the pdf and abstract urls from a Springer url """
-        if re.match(self.re_abs, url):
+        if re.match(self.re_abs_1, url):
             abs_url = url
             pdf_url = url.replace("article", "content/pdf")
+        elif re.match(self.re_abs_2, url):
+            abs_url = url
+            pdf_url = url.replace("chapter", "content/pdf")
         elif re.match(self.re_pdf, url):
-            abs_url = url.replace("content/pdf", "article")[: -len(".pdf")]
+            abs_url = self._get_abs_url(url)
             pdf_url = urllib.parse.unquote(url)
         else:
             raise URLResolutionError("Springer", url)
         return abs_url, pdf_url
 
     def validate(src):
-        return re.match(Springer.re_abs, src) or re.match(Springer.re_pdf, src)
+        return (
+            re.match(Springer.re_abs_1, src)
+            or re.match(Springer.re_abs_2, src)
+            or re.match(Springer.re_pdf, src)
+        )
diff --git a/tests/test_providers.py b/tests/test_providers.py
index 1a6f84f..5c8a8e4 100644
--- a/tests/test_providers.py
+++ b/tests/test_providers.py
@@ -126,13 +126,20 @@ class TestProviders(unittest.TestCase):
         filename = prov.run(url)
         self.assertEqual(exp_filename, os.path.basename(filename))
 
-    def test_springer(self):
+    def test_springer_1(self):
         prov = Springer(upload=False, verbose=VERBOSE)
         url = "https://link.springer.com/article/10.1007/s10618-019-00631-5"
         exp_filename = "Mauw_Ramirez-Cruz_Trujillo-Rasua_-_Robust_Active_Attacks_on_Social_Graphs_2019.pdf"
         filename = prov.run(url)
         self.assertEqual(exp_filename, os.path.basename(filename))
 
+    def test_springer_2(self):
+        prov = Springer(upload=False, verbose=VERBOSE)
+        url = "https://link.springer.com/content/pdf/10.1007%2F11681878_14.pdf"
+        exp_filename = "Dwork_et_al_-_Calibrating_Noise_to_Sensitivity_in_Private_Data_Analysis_2006.pdf"
+        filename = prov.run(url)
+        self.assertEqual(exp_filename, os.path.basename(filename))
+
     def test_local(self):
         local_filename = "test.pdf"
         with open(local_filename, "w") as fp: