From f25cf2121af72e681f9e8c1ae1fd6b6fa6fb5b03 Mon Sep 17 00:00:00 2001
From: Bob Mottram <bob@libreserver.org>
Date: Wed, 4 Jan 2023 13:33:05 +0000
Subject: [PATCH] Link to DOI scientific references

---
 content.py        | 36 +++++++++++++++++++++++++++++++++++-
 tests.py          | 19 +++++++++++++++++++
 webapp_podcast.py |  2 +-
 3 files changed, 55 insertions(+), 2 deletions(-)

diff --git a/content.py b/content.py
index 31d86978e..c80dd4229 100644
--- a/content.py
+++ b/content.py
@@ -585,6 +585,25 @@ def _shorten_linked_urls(content: str) -> str:
     return content
 
 
+def _contains_doi_reference(wrd: str, replace_dict: {}) -> bool:
+    """Handle DOI scientific references
+    """
+    if not wrd.startswith('doi:') and \
+       not wrd.startswith('DOI:'):
+        return False
+
+    doi_ref_str = wrd.split(':', 1)[1]
+    doi_site = 'https://sci-hub.ru'
+    markup = '<a href="' + doi_site + '/' + \
+        doi_ref_str + '" tabindex="10" ' + \
+        'rel="nofollow noopener noreferrer" ' + \
+        'target="_blank">' + \
+        '<span class="ellipsis">doi:' + doi_ref_str + \
+        '</span></a>'
+    replace_dict[wrd] = markup
+    return True
+
+
 def _contains_arxiv_reference(wrd: str, replace_dict: {}) -> bool:
     """Handle arxiv scientific references
     """
@@ -620,6 +639,19 @@ def _contains_arxiv_reference(wrd: str, replace_dict: {}) -> bool:
     return True
 
 
+def _contains_academic_references(content: str) -> bool:
+    """Does the given content contain academic references
+    """
+    prefixes = (
+        'arXiv:', 'arx:', 'arxiv:',
+        'doi:', 'DOI:'
+    )
+    for reference in prefixes:
+        if reference in content:
+            return True
+    return False
+
+
 def add_web_links(content: str) -> str:
     """Adds markup for web links
     """
@@ -639,7 +671,7 @@ def add_web_links(content: str) -> str:
 
     # if there are no prefixes then just keep the content we have
     if not prefix_found:
-        if 'arXiv:' in content or 'arx:' in content or 'arxiv:' in content:
+        if _contains_academic_references(content):
             prefix_found = True
         else:
             return content
@@ -652,6 +684,8 @@ def add_web_links(content: str) -> str:
             continue
         if _contains_arxiv_reference(wrd, replace_dict):
             continue
+        if _contains_doi_reference(wrd, replace_dict):
+            continue
         # does the word begin with a prefix?
         prefix_found = False
         for prefix in prefixes:
diff --git a/tests.py b/tests.py
index 2bacd9d6b..c99597fee 100644
--- a/tests.py
+++ b/tests.py
@@ -3623,6 +3623,11 @@ def _test_web_links():
         print(linked_text)
     assert expected_text in linked_text
 
+    # NOTE: it is difficult to find academic studies of the fediverse which
+    # do not in some way violate consent or embody an arrogant status
+    # quo attitude. Did all those scraped accounts agree to be part of
+    # an academic study? Did they even consider consent as an issue?
+    # It seems doubtful. We are just like algae under a microscope to them.
     example_text = \
         'This post has an arxiv link arXiv:2203.15752 some other text'
     linked_text = add_web_links(example_text)
@@ -3635,6 +3640,20 @@ def _test_web_links():
         print(linked_text)
     assert expected_text in linked_text
 
+    example_text = \
+        'This post has an doi link ' + \
+        'doi:10.1109/INFCOMW.2019.8845221 some other text'
+    linked_text = add_web_links(example_text)
+    expected_text = \
+        '<a href="https://sci-hub.ru/10.1109/INFCOMW.2019.8845221" ' + \
+        'tabindex="10" rel="nofollow noopener noreferrer"' + \
+        ' target="_blank"><span class="ellipsis">' + \
+        'doi:10.1109/INFCOMW.2019.8845221</span></a>'
+    if expected_text not in linked_text:
+        print(expected_text + '\n')
+        print(linked_text)
+    assert expected_text in linked_text
+
     example_text = \
         'This post has a very long web link\n\nhttp://' + \
         'cbwebewuvfuftdiudbqd33dddbbyuef23fyug3bfhcyu2fct2' + \
diff --git a/webapp_podcast.py b/webapp_podcast.py
index 4388b4d83..e54739dbc 100644
--- a/webapp_podcast.py
+++ b/webapp_podcast.py
@@ -299,7 +299,7 @@ def html_podcast_episode(translate: {},
                          text_mode_banner: str, access_keys: {},
                          session, session_onion, session_i2p,
                          http_prefix: str, debug: bool) -> str:
-    """Returns html for a podcast episode, giebn an item from the newswire
+    """Returns html for a podcast episode, an item from the newswire
     """
     css_filename = base_dir + '/epicyon-podcast.css'
     if os.path.isfile(base_dir + '/podcast.css'):