Created
August 19, 2015 07:29
-
-
Save theSage21/e1f23c1cc88c6f06e49a to your computer and use it in GitHub Desktop.
Results after changes
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Wrote profile results to testing.py.lprof | |
Timer unit: 1e-06 s | |
Total time: 0.001185 s | |
File: html2text/__init__.py | |
Function: feed at line 121 | |
Line # Hits Time Per Hit % Time Line Contents | |
============================================================== | |
121 @profile | |
122 def feed(self, data): | |
123 2 6 3.0 0.5 data = data.replace("</' + 'script>", "</ignore>") | |
124 2 1179 589.5 99.5 HTMLParser.HTMLParser.feed(self, data) | |
Total time: 0.195888 s | |
File: html2text/__init__.py | |
Function: handle at line 126 | |
Line # Hits Time Per Hit % Time Line Contents | |
============================================================== | |
126 @profile | |
127 def handle(self, data): | |
128 1 1183 1183.0 0.6 self.feed(data) | |
129 1 20 20.0 0.0 self.feed("") | |
130 1 194685 194685.0 99.4 return self.optwrap(self.close()) | |
Total time: 0.194552 s | |
File: html2text/__init__.py | |
Function: optwrap at line 786 | |
Line # Hits Time Per Hit % Time Line Contents | |
============================================================== | |
786 @profile | |
787 def optwrap(self, text): | |
788 """ | |
789 Wrap all paragraphs in the provided text. | |
790 | |
791 :type text: str | |
792 | |
793 :rtype: str | |
794 """ | |
795 1 3 3.0 0.0 if not self.body_width: | |
796 return text | |
797 | |
798 1 2 2.0 0.0 assert wrap, "Requires Python 2.3." | |
799 1 1 1.0 0.0 result = '' | |
800 1 1 1.0 0.0 newlines = 0 | |
801 # I cannot think of a better solution for now. | |
802 # To avoid the non-wrap behaviour for entire paras | |
803 # because of the presence of a link in it | |
804 1 2 2.0 0.0 if not self.wrap_links: | |
805 self.inline_links = False | |
806 3 13 4.3 0.0 for para in text.split("\n"): | |
807 2 4 2.0 0.0 if len(para) > 0: | |
808 1 210 210.0 0.1 if not skipwrap(para, self.wrap_links): | |
809 1 194295 194295.0 99.9 temp = (result, "\n".join(wrap(para, self.body_width))) | |
810 1 7 7.0 0.0 result = ''.join(temp) | |
811 1 5 5.0 0.0 if para.endswith(' '): | |
812 result += " \n" | |
813 newlines = 1 | |
814 else: | |
815 1 4 4.0 0.0 result += "\n\n" | |
816 1 2 2.0 0.0 newlines = 2 | |
817 else: | |
818 # Warning for the tempted!!! | |
819 # Be aware that obvious replacement of this with | |
820 # line.isspace() | |
821 # DOES NOT work! Explanations are welcome. | |
822 if not config.RE_SPACE.match(para): | |
823 result += para + "\n" | |
824 newlines = 1 | |
825 else: | |
826 1 2 2.0 0.0 if newlines < 2: | |
827 result += "\n" | |
828 newlines += 1 | |
829 1 1 1.0 0.0 return result | |
Total time: 0.197733 s | |
File: html2text/__init__.py | |
Function: html2text at line 831 | |
Line # Hits Time Per Hit % Time Line Contents | |
============================================================== | |
831 @profile | |
832 def html2text(html, baseurl='', bodywidth=config.BODY_WIDTH): | |
833 1 1834 1834.0 0.9 h = HTML2Text(baseurl=baseurl, bodywidth=bodywidth) | |
834 | |
835 1 195899 195899.0 99.1 return h.handle(html) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment