Add provider for ECCC (partially addresses #104)

4 files changed, 118 insertions, 0 deletions

diff --git a/paper2remarkable/providers/__init__.py b/paper2remarkable/providers/__init__.py
index 5130147..e574b80 100644
--- a/paper2remarkable/providers/__init__.py
+++ b/paper2remarkable/providers/__init__.py
@@ -5,6 +5,7 @@ from .acm import ACM
 from .arxiv import Arxiv
 from .citeseerx import CiteSeerX
 from .cvf import CVF
+from .eccc import ECCC
 from .html import HTML
 from .jmlr import JMLR
 from .local import LocalFile
@@ -28,6 +29,7 @@ providers = [
     Arxiv,
     CiteSeerX,
     CVF,
+    ECCC,
     JMLR,
     Nature,
     NBER,
diff --git a/paper2remarkable/providers/eccc.py b/paper2remarkable/providers/eccc.py
new file mode 100644
index 0000000..f6a6bae
--- /dev/null
+++ b/paper2remarkable/providers/eccc.py
@@ -0,0 +1,86 @@
+# -*- coding: utf-8 -*-
+
+"""Provider for Electronic Colloquium on Computational Complexity
+
+Author: G.J.J. van den Burg
+License: See LICENSE file
+Copyright: 2021, G.J.J. van den Burg
+
+"""
+
+import bs4
+import re
+
+from ._info import Informer
+from ._base import Provider
+from ..exceptions import URLResolutionError
+from ..log import Logger
+
+logger = Logger()
+
+
+class ECCCInformer(Informer):
+    def _get_paper_div(self, soup):
+        h3 = soup.find(lambda t: t.name == "h3" and t.get_text() == "Paper:")
+        div = h3.find_next_sibling("div")
+        return bs4.BeautifulSoup(div.prettify(), "html.parser")
+
+    def get_title(self, soup):
+        divsoup = self._get_paper_div(soup)
+        h4 = divsoup.find("h4")
+        if not h4:
+            logger.warning(
+                "Couldn't determine title information, maybe provide the desired filename using '--filename'?"
+            )
+            return ""
+        return h4.get_text().strip()
+
+    def get_authors(self, soup):
+        divsoup = self._get_paper_div(soup)
+        aa = divsoup.find_all(
+            lambda t: t.name == "a" and t.get("href").startswith("/author/")
+        )
+        if not aa:
+            logger.warning(
+                "Couldn't determine author information, maybe provide the desired filename using '--filename'?"
+            )
+            return ""
+        authors = [a.get_text() for a in aa]
+        return self._format_authors(authors, sep=" ", idx=-1)
+
+    def get_year(self, soup):
+        divsoup = self._get_paper_div(soup)
+        line = next(
+            (l for l in divsoup.text.split("\n") if "Publication: " in l), None
+        )
+        if line is None:
+            logger.warning(
+                "Couldn't determine year information, maybe provide the desired filename using '--filename'?"
+            )
+            return ""
+        year = line.strip().split(" ")[3]  # bit lazy
+        return year
+
+
+class ECCC(Provider):
+
+    re_abs = "https?://eccc.weizmann.ac.il/report/\d{4}/\d+/?$"
+    re_pdf = "https?://eccc.weizmann.ac.il/report/\d{4}/\d+/download/?$"
+
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.informer = ECCCInformer()
+
+    def get_abs_pdf_urls(self, url):
+        if re.match(self.re_abs, url):
+            abs_url = url
+            pdf_url = url.rstrip("/") + "/download"
+        elif re.match(self.re_pdf, url):
+            abs_url = url.rstrip("/")[: -len("/download")]
+            pdf_url = url
+        else:
+            raise URLResolutionError("ECCC", url)
+        return abs_url, pdf_url
+
+    def validate(src):
+        return re.match(ECCC.re_abs, src) or re.match(ECCC.re_pdf, src)
diff --git a/tests/test_providers.py b/tests/test_providers.py
index 1c1e1e6..efc4be1 100644
--- a/tests/test_providers.py
+++ b/tests/test_providers.py
@@ -21,6 +21,7 @@ from paper2remarkable.providers import (
     Arxiv,
     CVF,
     CiteSeerX,
+    ECCC,
     HTML,
     JMLR,
     LocalFile,
@@ -478,6 +479,29 @@ class TestProviders(unittest.TestCase):
             with pdf.open_outline() as outline:
                 assert len(outline.root) > 0
 
+    def test_eccc_1(self):
+        prov = ECCC(upload=False, verbose=VERBOSE)
+        url = "https://eccc.weizmann.ac.il/report/2021/063/"
+        exp = "Chou_et_al_-_Approximability_of_All_Finite_CSPs_in_the_Dynamic_Streaming_Setting_2021.pdf"
+        filename = prov.run(url)
+        self.assertEqual(exp, os.path.basename(filename))
+
+    def test_eccc_2(self):
+        prov = ECCC(upload=False, verbose=VERBOSE)
+        url = "https://eccc.weizmann.ac.il/report/2007/003/"
+        exp = "Cai_Lu_-_Bases_Collapse_in_Holographic_Algorithms_2007.pdf"
+        filename = prov.run(url)
+        self.assertEqual(exp, os.path.basename(filename))
+
+    def test_eccc_3(self):
+        prov = ECCC(upload=False, verbose=VERBOSE)
+        url = "https://eccc.weizmann.ac.il/report/1998/052/download"
+        exp = (
+            "Hemkemeier_Vallentin_-_On_the_Decomposition_of_Lattices_1998.pdf"
+        )
+        filename = prov.run(url)
+        self.assertEqual(exp, os.path.basename(filename))
+
 
 if __name__ == "__main__":
     unittest.main()
diff --git a/tests/test_ui.py b/tests/test_ui.py
index 86a3c8e..87b66d8 100644
--- a/tests/test_ui.py
+++ b/tests/test_ui.py
@@ -22,6 +22,7 @@ from paper2remarkable.providers import (
     Arxiv,
     CiteSeerX,
     CVF,
+    ECCC,
     HTML,
     JMLR,
     LocalFile,
@@ -203,6 +204,11 @@ class TestUI(unittest.TestCase):
                 "https://www.aclweb.org/anthology/2020.sigmorphon-1.29v2.pdf",
                 "https://www.aclweb.org/anthology/2020.sigmorphon-1.29v2.pdf",
             ),
+            (
+                ECCC,
+                "https://eccc.weizmann.ac.il/report/2019/006/",
+                "https://eccc.weizmann.ac.il/report/2019/006/",
+            ),
         ]
         for exp_prov, url, exp_url in tests:
             prov, new_url, jar = choose_provider(url)