fix-bug: missing-pick-sentences

.gitignore

@@ -16,3 +16,4 @@ lib/
 local/
 man/
 share/
+.idea/

breadability/readable.py

@@ -21,7 +21,6 @@
 )
 from .utils import cached_property, shrink_text
 
-
 html_cleaner = Cleaner(
     scripts=True, javascript=True, comments=True,
     style=True, links=True, meta=False, add_nofollow=False,
@@ -30,7 +29,6 @@
     annoying_tags=False, remove_tags=None, kill_tags=("noscript", "iframe"),
     remove_unknown_tags=False, safe_attrs_only=False)
 
-
 SCORABLE_TAGS = ("div", "p", "td", "pre", "article")
 ANNOTATION_TAGS = (
     "a", "abbr", "acronym", "b", "big", "blink", "blockquote", "br", "cite",
@@ -193,6 +191,7 @@ def clean_document(node):
             logger.debug("Dropping <%s>, it's insignificant", n.tag)
             to_drop.append(n)
 
+        """ modified by guojw
         # drop block element without content and children
         if n.tag in ("div", "p"):
             text_content = shrink_text(n.text_content())
@@ -204,6 +203,7 @@ def clean_document(node):
         # finally try out the conditional cleaning of the target node
         if clean_conditionally(n):
             to_drop.append(n)
+        """
 
     drop_nodes_with_parents(to_drop)
 

breadability/scoring.py

@@ -13,7 +13,6 @@
 from ._compat import to_bytes
 from .utils import normalize_whitespace
 
-
 # A series of sets of attributes we check to help in determining if a node is
 # a potential candidate or not.
 CLS_UNLIKELY = re.compile(
@@ -140,7 +139,9 @@ def is_unlikely_node(node):
 
 def score_candidates(nodes):
     """Given a list of potential nodes, find some initial scores to start"""
-    MIN_HIT_LENTH = 25
+    # guojw
+    # MIN_HIT_LENTH = 25
+    MIN_HIT_LENTH = 1
     candidates = {}
 
     for node in nodes: