From 7bbefa1fcc9a6006953eb0a79049ece9f05985de Mon Sep 17 00:00:00 2001 From: Leonard Richardson Date: Fri, 28 Jan 2011 11:39:36 -0500 Subject: Moved everything into the top-level directory and got rid of buildout. --- AUTHORS | 34 ++ CHANGELOG | 122 ++++++ README.txt | 26 ++ TODO | 45 +++ __init__.py | 375 ++++++++++++++++++ _bootstrap/COPYRIGHT.txt | 9 - _bootstrap/LICENSE.txt | 54 --- _bootstrap/bootstrap.py | 77 ---- bootstrap.py | 1 - buildout.cfg | 31 -- dammit.py | 292 ++++++++++++++ element.py | 851 +++++++++++++++++++++++++++++++++++++++++ lxml_test.py | 13 - setup.py | 44 --- src/beautifulsoup/AUTHORS | 34 -- src/beautifulsoup/CHANGELOG | 122 ------ src/beautifulsoup/README.txt | 26 -- src/beautifulsoup/TODO | 42 -- src/beautifulsoup/__init__.py | 375 ------------------ src/beautifulsoup/dammit.py | 292 -------------- src/beautifulsoup/element.py | 851 ----------------------------------------- src/beautifulsoup/python3.diff | 208 ---------- src/beautifulsoup/testing.py | 154 -------- src/beautifulsoup/util.py | 29 -- testall.sh | 2 - testing.py | 154 ++++++++ to3.sh | 9 - util.py | 29 ++ 28 files changed, 1928 insertions(+), 2373 deletions(-) create mode 100644 AUTHORS create mode 100644 CHANGELOG create mode 100644 README.txt create mode 100644 __init__.py delete mode 100644 _bootstrap/COPYRIGHT.txt delete mode 100644 _bootstrap/LICENSE.txt delete mode 100644 _bootstrap/bootstrap.py delete mode 120000 bootstrap.py delete mode 100644 buildout.cfg create mode 100644 dammit.py create mode 100644 element.py delete mode 100644 lxml_test.py delete mode 100644 setup.py delete mode 100644 src/beautifulsoup/AUTHORS delete mode 100644 src/beautifulsoup/CHANGELOG delete mode 100644 src/beautifulsoup/README.txt delete mode 100644 src/beautifulsoup/TODO delete mode 100644 src/beautifulsoup/__init__.py delete mode 100644 src/beautifulsoup/dammit.py delete mode 100644 src/beautifulsoup/element.py delete mode 100644 src/beautifulsoup/python3.diff delete mode 100644 src/beautifulsoup/testing.py delete mode 100644 src/beautifulsoup/util.py delete mode 100755 testall.sh create mode 100644 testing.py delete mode 100755 to3.sh create mode 100644 util.py diff --git a/AUTHORS b/AUTHORS new file mode 100644 index 0000000..d353253 --- /dev/null +++ b/AUTHORS @@ -0,0 +1,34 @@ +Behold, mortal, the origins of Beautiful Soup... +================================================ + +Leonard Richardson is the primary programmer. + +Sam Ruby helps with a lot of edge cases. + +Mark Pilgrim provided the encoding detection code that forms the base +of UnicodeDammit. + +Jonathan Ellis was awarded the prestigous Beau Potage D'Or for his +work in solving the nestable tags conundrum. + +The following people have contributed patches to Beautiful Soup: + + Istvan Albert, Andrew Lin, Anthony Baxter, Andrew Boyko, Tony Chang, + Zephyr Fang, Fuzzy, Roman Gaufman, Yoni Gilad, Richie Hindle, Peteris + Krumins, Kent Johnson, Ben Last, Robert Leftwich, Staffan Malmgren, + Ksenia Marasanova, JP Moins, Adam Monsen, John Nagle, "Jon", Ed + Oskiewicz, Greg Phillips, Giles Radford, Arthur Rudolph, Marko + Samastur, Jouni Seppänen, Alexander Schmolck, Andy Theyers, Glyn + Webster, Paul Wright, Danny Yoo + +The following people made suggestions or found bugs or found ways to +break Beautiful Soup: + + Hanno Böck, Matteo Bertini, Chris Curvey, Simon Cusack, Matt Ernst, + Michael Foord, Tom Harris, Bill de hOra, Donald Howes, Matt + Patterson, Scott Roberts, Steve Strassmann, Mike Williams, warchild + at redho dot com, Sami Kuisma, Carlos Rocha, Bob Hutchison, Joren Mc, + Michal Migurski, John Kleven, Tim Heaney, Tripp Lilley, Ed Summers, + Dennis Sutch, Chris Smith, Aaron Sweep^W Swartz, Stuart Turner, Greg + Edwards, Kevin J Kalupson, Nikos Kouremenos, Artur de Sousa Rocha, + Yichun Wei, Per Vognsen diff --git a/CHANGELOG b/CHANGELOG new file mode 100644 index 0000000..4e97e1b --- /dev/null +++ b/CHANGELOG @@ -0,0 +1,122 @@ += 3.1.0 = + +A hybrid version that supports 2.4 and can be automatically converted +to run under Python 3.0. There are three backwards-incompatible +changes you should be aware of, but no new features or deliberate +behavior changes. + +1. str() may no longer do what you want. This is because the meaning +of str() inverts between Python 2 and 3; in Python 2 it gives you a +byte string, in Python 3 it gives you a Unicode string. + +The effect of this is that you can't pass an encoding to .__str__ +anymore. Use encode() to get a string and decode() to get Unicode, and +you'll be ready (well, readier) for Python 3. + +2. Beautiful Soup is now based on HTMLParser rather than SGMLParser, +which is gone in Python 3. There's some bad HTML that SGMLParser +handled but HTMLParser doesn't, usually to do with attribute values +that aren't closed or have brackets inside them: + + baz + ', '"> + +A later version of Beautiful Soup will allow you to plug in different +parsers to make tradeoffs between speed and the ability to handle bad +HTML. + +3. In Python 3 (but not Python 2),HTMLParser converts entities within +attributes to the corresponding Unicode characters. In Python 2 it's +possible to parse this string and leave the é intact. + + + +In Python 3, the é is always converted to \xe9 during +parsing. + + += 3.0.7a = + +Added an import that makes BS work in Python 2.3. + + += 3.0.7 = + +Fixed a UnicodeDecodeError when unpickling documents that contain +non-ASCII characters. + +Fixed a TypeError that occured in some circumstances when a tag +contained no text. + +Jump through hoops to avoid the use of chardet, which can be extremely +slow in some circumstances. UTF-8 documents should never trigger the +use of chardet. + +Whitespace is preserved inside 
 and ")
-
-    def test_single_quote_attribute_values_become_double_quotes(self):
-        self.assertSoupEquals("",
-                              '')
-
-    def test_attribute_values_with_nested_quotes_are_left_alone(self):
-        text = """a"""
-        self.assertSoupEquals(text)
-
-    def test_attribute_values_with_double_nested_quotes_get_quoted(self):
-        text = """a"""
-        soup = self.soup(text)
-        soup.foo['attr'] = 'Brawls happen at "Bob\'s Bar"'
-        self.assertSoupEquals(
-            soup.foo.decode(),
-            """a""")
-
-    def test_ampersand_in_attribute_value_gets_quoted(self):
-        self.assertSoupEquals('',
-                              '')
-
-
-class BuilderInvalidMarkupSmokeTest(SoupTest):
-    """Tests of invalid markup.
-
-    These are very likely to give different results for different tree
-    builders. It's not required that a tree builder handle invalid
-    markup at all.
-    """
-
-    def test_unclosed_block_level_elements(self):
-        # Unclosed block-level elements should be closed.
-        self.assertSoupEquals(
-            '
Foo
Bar',
-            '
Foo
Bar
')
-
-    def test_fake_self_closing_tag(self):
-        # If a self-closing tag presents as a normal tag, the 'open'
-        # tag is treated as an instance of the self-closing tag and
-        # the 'close' tag is ignored.
-        self.assertSoupEquals(
-            "http://foo.com/",
-            "http://foo.com/")
-
-    def test_boolean_attribute_with_no_value_gets_empty_value(self):
-        soup = self.soup("
foo
")
-        self.assertEquals(soup.table.td['nowrap'], '')
-
-    def test_incorrectly_nested_tables(self):
-        self.assertSoupEquals(
-            '
',
-            '
')
-
-
-
diff --git a/src/beautifulsoup/util.py b/src/beautifulsoup/util.py
deleted file mode 100644
index 693a7e2..0000000
--- a/src/beautifulsoup/util.py
+++ /dev/null
@@ -1,29 +0,0 @@
-# Helper functions and mixin classes for Beautiful Soup
-
-import types
-try:
-    set
-except NameError:
-    from sets import Set as set
-
-def isList(l):
-    """Convenience method that works with all 2.x versions of Python
-    to determine whether or not something is listlike."""
-    return ((hasattr(l, '__iter__') and not isString(l))
-            or (type(l) in (types.ListType, types.TupleType)))
-
-def isString(s):
-    """Convenience method that works with all 2.x versions of Python
-    to determine whether or not something is stringlike."""
-    try:
-        return isinstance(s, unicode) or isinstance(s, basestring)
-    except NameError:
-        return isinstance(s, str)
-
-def buildSet(args=None):
-    """Turns a list or a string into a set."""
-    if isinstance(args, str):
-        return set([args])
-    if args is None:
-        return set()
-    return set(args)
diff --git a/testall.sh b/testall.sh
deleted file mode 100755
index 801124f..0000000
--- a/testall.sh
+++ /dev/null
@@ -1,2 +0,0 @@
-#!/bin/sh
-python BeautifulSoupTests.py && sh to3.sh && cd python3 && python3 BeautifulSoupTests.py
diff --git a/testing.py b/testing.py
new file mode 100644
index 0000000..20d087e
--- /dev/null
+++ b/testing.py
@@ -0,0 +1,154 @@
+"""Helper classes for tests."""
+
+import unittest
+from beautifulsoup import BeautifulSoup
+from beautifulsoup.element import Comment, SoupStrainer
+from beautifulsoup.builder.lxml_builder import LXMLTreeBuilder
+
+class SoupTest(unittest.TestCase):
+
+    def setUp(self):
+        # LXMLTreeBuilder won't handle bad markup, but that's fine,
+        # since all the parsing tests take place in parser-specific
+        # test suites that override default_builder.
+        self.default_builder = LXMLTreeBuilder()
+
+    def soup(self, markup, **kwargs):
+        """Build a Beautiful Soup object from markup."""
+        return BeautifulSoup(markup, builder=self.default_builder, **kwargs)
+
+    def document_for(self, markup):
+        """Turn an HTML fragment into a document.
+
+        The details depend on the builder.
+        """
+        return self.default_builder.test_fragment_to_document(markup)
+
+    def assertSoupEquals(self, to_parse, compare_parsed_to=None):
+        builder = self.default_builder
+        obj = BeautifulSoup(to_parse, builder=builder)
+        if compare_parsed_to is None:
+            compare_parsed_to = to_parse
+
+        self.assertEquals(obj.decode(), self.document_for(compare_parsed_to))
+
+
+
+class BuilderSmokeTest(SoupTest):
+    """A generic smoke test for tree builders.
+
+    Subclasses of this test ensure that all of Beautiful Soup's tree
+    builders generate more or less the same trees. It's okay for trees
+    to differ, especially when given invalid markup--just override the
+    appropriate test method to demonstrate how one tree builder
+    differs from others.
+    """
+
+    def test_bare_string(self):
+        # A bare string is turned into some kind of HTML document or
+        # fragment recognizable as the original string.
+        self.assertSoupEquals("A bare string")
+
+    def test_mixed_case_tags(self):
+        # Mixed-case tags are folded to lowercase.
+        self.assertSoupEquals(
+            "",
+            "")
+
+    def test_self_closing(self):
+        # HTML's self-closing tags are recognized as such.
+        self.assertSoupEquals(
+            "
A  tag
", "
A  tag
")
+
+        self.assertSoupEquals(
+            "
Foo
bar
", "
Foo
bar
")
+
+    def test_comment(self):
+        # Comments are represented as Comment objects.
+        markup = "
foobaz
"
+        self.assertSoupEquals(markup)
+
+        soup = self.soup(markup)
+        comment = soup.find(text="foobar")
+        self.assertEquals(comment.__class__, Comment)
+
+    def test_nested_inline_elements(self):
+        # Inline tags can be nested indefinitely.
+        b_tag = "Inside a B tag"
+        self.assertSoupEquals(b_tag)
+
+        nested_b_tag = "
A nested tag
"
+        self.assertSoupEquals(nested_b_tag)
+
+        double_nested_b_tag = "
A doubly nested tag
"
+        self.assertSoupEquals(nested_b_tag)
+
+    def test_nested_block_level_elements(self):
+        soup = self.soup('
Foo
')
+        blockquote = soup.blockquote
+        self.assertEqual(blockquote.p.b.string, 'Foo')
+        self.assertEqual(blockquote.b.string, 'Foo')
+
+    def test_collapsed_whitespace(self):
+        """In most tags, whitespace is collapsed."""
+        self.assertSoupEquals("
   
", "
 
")
+
+    def test_preserved_whitespace_in_pre_and_textarea(self):
+        """In 
 and ")
+
+    def test_single_quote_attribute_values_become_double_quotes(self):
+        self.assertSoupEquals("",
+                              '')
+
+    def test_attribute_values_with_nested_quotes_are_left_alone(self):
+        text = """a"""
+        self.assertSoupEquals(text)
+
+    def test_attribute_values_with_double_nested_quotes_get_quoted(self):
+        text = """a"""
+        soup = self.soup(text)
+        soup.foo['attr'] = 'Brawls happen at "Bob\'s Bar"'
+        self.assertSoupEquals(
+            soup.foo.decode(),
+            """a""")
+
+    def test_ampersand_in_attribute_value_gets_quoted(self):
+        self.assertSoupEquals('',
+                              '')
+
+
+class BuilderInvalidMarkupSmokeTest(SoupTest):
+    """Tests of invalid markup.
+
+    These are very likely to give different results for different tree
+    builders. It's not required that a tree builder handle invalid
+    markup at all.
+    """
+
+    def test_unclosed_block_level_elements(self):
+        # Unclosed block-level elements should be closed.
+        self.assertSoupEquals(
+            '
Foo
Bar',
+            '
Foo
Bar
')
+
+    def test_fake_self_closing_tag(self):
+        # If a self-closing tag presents as a normal tag, the 'open'
+        # tag is treated as an instance of the self-closing tag and
+        # the 'close' tag is ignored.
+        self.assertSoupEquals(
+            "http://foo.com/",
+            "http://foo.com/")
+
+    def test_boolean_attribute_with_no_value_gets_empty_value(self):
+        soup = self.soup("
foo
")
+        self.assertEquals(soup.table.td['nowrap'], '')
+
+    def test_incorrectly_nested_tables(self):
+        self.assertSoupEquals(
+            '
',
+            '
')
+
+
+
diff --git a/to3.sh b/to3.sh
deleted file mode 100755
index 26b3246..0000000
--- a/to3.sh
+++ /dev/null
@@ -1,9 +0,0 @@
-#!/bin/sh
-mkdir python3
-for i in BeautifulSoupTests.py builder.py element.py dammit.py
-do
-    cp $i python3/
-    2to3-3.0 -x next $i | patch -p0 python3/$i
-    cp python3/$i python3/$i.orig
-    patch -p0 python3/$i < $i.3.diff
-done
\ No newline at end of file
diff --git a/util.py b/util.py
new file mode 100644
index 0000000..693a7e2
--- /dev/null
+++ b/util.py
@@ -0,0 +1,29 @@
+# Helper functions and mixin classes for Beautiful Soup
+
+import types
+try:
+    set
+except NameError:
+    from sets import Set as set
+
+def isList(l):
+    """Convenience method that works with all 2.x versions of Python
+    to determine whether or not something is listlike."""
+    return ((hasattr(l, '__iter__') and not isString(l))
+            or (type(l) in (types.ListType, types.TupleType)))
+
+def isString(s):
+    """Convenience method that works with all 2.x versions of Python
+    to determine whether or not something is stringlike."""
+    try:
+        return isinstance(s, unicode) or isinstance(s, basestring)
+    except NameError:
+        return isinstance(s, str)
+
+def buildSet(args=None):
+    """Turns a list or a string into a set."""
+    if isinstance(args, str):
+        return set([args])
+    if args is None:
+        return set()
+    return set(args)
-- 
cgit v1.2.3