Add a new AdvancedHTMLMiniFormatter to Formatter.py, which will only output the function parts of whitespace stripping all excess and not pretty-printing. This is exposed on AdvancedHTMLParser.AdvancedHTMLParser via a new method 'getMiniHTML'

Author: kata198

AdvancedHTMLParser/Formatter.py

@@ -27,7 +27,7 @@
 import codecs
 
 
-__all__ = ('AdvancedHTMLFormatter', )
+__all__ = ('AdvancedHTMLFormatter', 'AdvancedHTMLMiniFormatter')
 
 class AdvancedHTMLFormatter(HTMLParser):
     '''
@@ -37,10 +37,11 @@ class AdvancedHTMLFormatter(HTMLParser):
 
     def __init__(self, indent='  ', encoding='utf-8'):
         '''
-            Create a formatter.
+            Create a pretty formatter.
 
-            @param indent - Either a space/tab/newline that represents one level of indent, or an integer to use that number of spaces
-            @param encoding - Use this encoding for the document.
+            @param indent <str/int>, Default '    ' [4 spaces] - Either a space/tab/newline that represents one level of indent, or an integer to use that number of spaces
+
+            @param encoding <str/None>, Default 'utf-8', - Use this encoding for the document. None to not mess with encoding
         '''
         HTMLParser.__init__(self)
 
@@ -307,4 +308,22 @@ def parseStr(self, html):
         else:
             self.feed(html)
 
+
+class AdvancedHTMLMiniFormatter(AdvancedHTMLFormatter):
+    '''
+        AdvancedHTMLMiniFormatter - A formatter that will reformat a document, keeping only functional
+            whitespace and removing any and all indentation and nesting spaces.
+    '''
+
+    def __init__(self, encoding='utf-8'):
+        '''
+            Create a mini formatter.
+
+            @param encoding <str/None>, Default 'utf-8', - Use this encoding for the document. None to not mess with encoding
+        '''
+        AdvancedHTMLFormatter.__init__(self, indent='', encoding=encoding)
+
+    def _getIndent(self):
+        return ''
+
 #vim: set ts=4 sw=4 expandtab

AdvancedHTMLParser/Parser.py

@@ -767,8 +767,15 @@ def doMatchFunc(em):
 
     def getHTML(self):
         '''
-            getHTML - Get the full HTML as contained within this tree
-                @returns - String
+            getHTML - Get the full HTML as contained within this tree.
+
+                If parsed from a document, this will contain the original whitespacing.
+
+                @returns - <str> of html
+
+                    @see getFormattedHTML
+
+                    @see getMiniHTML
         '''
         root = self.getRoot()
         if root is None:
@@ -799,18 +806,35 @@ def getHTML(self):
 
     def getFormattedHTML(self, indent='  '):
         '''
-            getFormattedHTML - Get formatted and xhtml of this document
+            getFormattedHTML - Get formatted and xhtml of this document, replacing the original whitespace
+                with a pretty-printed version
 
             @param indent - space/tab/newline of each level of indent, or integer for how many spaces per level
         
-            @return - Formatted html as string
+            @return - <str> Formatted html
+
+            @see getHTML - Get HTML with original whitespace
+
+            @see getMiniHTML - Get HTML with only functional whitespace remaining
         '''
         from .Formatter import AdvancedHTMLFormatter
         html = self.getHTML()
         formatter = AdvancedHTMLFormatter(indent, None) # Do not double-encode
         formatter.feed(html)
         return formatter.getHTML()
-    
+
+    def getMiniHTML(self):
+        '''
+            getMiniHTML - Gets the HTML representation of this document without any pretty formatting
+                and disregarding original whitespace beyond the functional.
+
+                @return <str> - HTML with only functional whitespace present
+        '''
+        from .Formatter import AdvancedHTMLMiniFormatter
+        html = self.getHTML()
+        formatter = AdvancedHTMLMiniFormatter(None) # Do not double-encode
+        formatter.feed(html)
+        return formatter.getHTML()
 
     def _reset(self):
         '''