[WIP] Provider for Taylor and Francis Online

4 files changed, 94 insertions, 0 deletions

diff --git a/paper2remarkable/providers/__init__.py b/paper2remarkable/providers/__init__.py
index fabdcfe..53fda1f 100644
--- a/paper2remarkable/providers/__init__.py
+++ b/paper2remarkable/providers/__init__.py
@@ -10,6 +10,7 @@ from .pdf_url import PdfUrl
 from .pmlr import PMLR
 from .pubmed import PubMed
 from .springer import Springer
+from .tandfonline import TandFOnline
 
 # NOTE: Order matters here, PdfUrl should be last
 providers = [
@@ -21,6 +22,7 @@ providers = [
     PMLR,
     PubMed,
     Springer,
+    TandFOnline,
     LocalFile,
     PdfUrl,
 ]
diff --git a/paper2remarkable/providers/_base.py b/paper2remarkable/providers/_base.py
index b2f584c..596af98 100644
--- a/paper2remarkable/providers/_base.py
+++ b/paper2remarkable/providers/_base.py
@@ -116,7 +116,9 @@ class Provider(metaclass=abc.ABCMeta):
         self.initial_dir = os.getcwd()
         with tempfile.TemporaryDirectory(prefix="p2r_") as working_dir:
             os.chdir(working_dir)
+            print("working_dir:", working_dir)
             self.retrieve_pdf(pdf_url, tmp_filename)
+
             assert_file_is_pdf(tmp_filename)
 
             intermediate_fname = tmp_filename
diff --git a/paper2remarkable/providers/tandfonline.py b/paper2remarkable/providers/tandfonline.py
new file mode 100644
index 0000000..9f9c59e
--- /dev/null
+++ b/paper2remarkable/providers/tandfonline.py
@@ -0,0 +1,75 @@
+# -*- coding: utf-8 -*-
+
+"""Provider for Taylor and Francis Online
+
+Author: G.J.J. van den Burg
+License: See LICENSE file
+Copyright: 2020, G.J.J. van den Burg
+
+"""
+
+import re
+
+from ._base import Provider
+from ._info import Informer
+from ..exceptions import URLResolutionError
+from ..log import Logger
+
+logger = Logger()
+
+
+class TandFOnlineInformer(Informer):
+    meta_title_key = "dc.Title"
+    meta_author_key = "dc.Creator"
+    meta_date_key = "dc.Date"
+
+    def _format_authors(self, soup_authors):
+        return super()._format_authors(soup_authors, sep=" ", idx=-1)
+
+    def _format_year(self, soup_date):
+        return soup_date.strip().split(" ")[-1].strip()
+
+
+class TandFOnline(Provider):
+
+    re_abs = "^https?://www.tandfonline.com/doi/(full|abs)/(?P<doi>\d+\.\d+/\d+\.\d+\.\d+)"
+    re_pdf = "^https?://www.tandfonline.com/doi/(full|pdf)/(?P<doi>\d+\.\d+/\d+\.\d+\.\d+)"
+
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.informer = TandFOnlineInformer()
+
+    def _get_doi(self, url):
+        m = re.match(self.re_abs, url) or re.match(self.re_pdf, url)
+        if m:
+            return m["doi"]
+        raise URLResolutionError(
+            "TandFOnline", url, reason="Failed to retrieve DOI."
+        )
+
+    def get_abs_pdf_urls(self, url):
+        if re.match(self.re_abs, url):
+            abs_url = url
+            doi = self._get_doi(url)
+            pdf_url = "https://www.tandfonline.com/doi/pdf/{doi}?needAccess=true".format(
+                doi=doi
+            )
+        elif re.match(self.re_pdf, url):
+            doi = self._get_doi(url)
+            pdf_url = "https://www.tandfonline.com/doi/pdf/{doi}?needAccess=true".format(
+                doi=doi
+            )
+            # full redirects to abs if we don't have access
+            abs_url = "https://www.tandfonline.com/doi/full/{doi}".format(
+                doi=doi
+            )
+        else:
+            raise URLResolutionError("TandFOnline", url)
+        print("abs:", abs_url, "pdf:", pdf_url)
+        return abs_url, pdf_url
+
+    def validate(src):
+        m = re.match(TandFOnline.re_abs, src) or re.match(
+            TandFOnline.re_pdf, src
+        )
+        return not m is None
diff --git a/tests/test_providers.py b/tests/test_providers.py
index e256eec..3204768 100644
--- a/tests/test_providers.py
+++ b/tests/test_providers.py
@@ -23,6 +23,7 @@ from paper2remarkable.providers import (
     PdfUrl,
     PubMed,
     Springer,
+    TandFOnline,
 )
 from paper2remarkable.providers.arxiv import DEARXIV_TEXT_REGEX
 
@@ -206,6 +207,20 @@ class TestProviders(unittest.TestCase):
         filename = prov.run(url)
         self.assertEqual(exp, os.path.basename(filename))
 
+    def test_tandfonline_1(self):
+        prov = TandFOnline(upload=False, verbose=VERBOSE)
+        url = "https://www.tandfonline.com/doi/full/10.1080/01621459.2017.1385466"
+        exp = "Fearnhead_Rigaill_-_Changepoint_Detection_in_the_Presence_of_Outliers_2018.pdf"
+        filename = prov.run(url)
+        self.assertEqual(exp, os.path.basename(filename))
+
+    def test_tandfonline_2(self):
+        prov = TandFOnline(upload=False, verbose=VERBOSE)
+        url = "https://www.tandfonline.com/doi/pdf/10.1080/03610918.2012.625790?scroll=top&needAccess=true"
+        exp = "Huskova_Marusiakova_-_M-Procedures_for_Detection_of_Changes_for_Dependent_Observations_2012.pdf"
+        filename = prov.run(url)
+        self.assertEqual(exp, os.path.basename(filename))
+
 
 if __name__ == "__main__":
     unittest.main()