[ccan] It was only a matter of time...

Daniel Burke dan.p.burke at gmail.com
Mon May 16 19:45:38 EST 2011


As <http://www.grinninglizard.com/tinyxml/> it's<http://expat.sourceforge.net/>
 common <http://www.minixml.org/> knowledge <http://xmlsoft.org/>
that<http://xerces.apache.org/xerces-c/>
the <http://www.cs.rpi.edu/~tayloj/CL-RDFXML/>
world<http://common-lisp.net/project/cl-xml/>
doesn't <http://common-lisp.net/project/s-xml/>
have<http://www.common-lisp.net/project/xmls/>
enough <http://sourceforge.net/projects/pyxml/> XML <http://lxml.de/>
parsers <http://wiki.python.org/moin/PythonXml>,
I<http://library.forum.nokia.com/index.jsp?topic=/S60_5th_Edition_Cpp_Developers_Library/GUID-35228542-8C95-4849-A73F-2B4F082F0C44/sdk/doc_source/reference/reference-cpp/XML_Framework/Xml_1_1CParserClass.html>
decided <http://cookxml.yuanheng.org/> to <http://woodstox.codehaus.org/>
fix <http://php.net/manual/en/book.xml.php>
that<http://infinitesque.net/projects/4Suite/docs/CoreManual.html>
.

Well, my real motivation was to have a pure C xml parser that I could drop
into an existing project, that had no dependencies and is very small. Think
"I have 4Mb of RAM" small. It uses buffer size + final data structure
memory, and as far as I can tell it should function correctly with a buffer
as small as 3 bytes, although I found about 3k works fastest on my setup.

It's 340 lines, plus 60 lines of test harness, and doesn't leak a byte.
Malicious XML should do no worse than create extra text nodes. It also
totally ignores character sets; it's 7bit ASCII or the highway.

I figure CCAN would be the place to put it. How do we go about that?

Oh yes... I think the name for it should be TTXML... the Trivial Trusted XML
parser... emphasis on the trusted part, as dealing with random files on the
'net is done better elsewhere.

regards,

dan
--
"Within C++, there is a much smaller and cleaner language struggling to get
out"
--Bjarne Stroustrup
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ozlabs.org/pipermail/ccan/attachments/20110516/3322969d/attachment.html>


More information about the ccan mailing list