Re: [sqlmap-users] --skip-urlencode forcing content type to text/plain

[Brandon P. <bpe...@gm...>]

I am magically unable to reproduce this at the moment.

If I end up seeing the behavior again, I will get more information to reproduce with.

However, I did realize that I no longer actually need —skip-urlencode. At some point between when I last touched this code and now, POST request bodies are no longer URL encoded.

In any case, sorry for the noise.


> On Oct 19, 2015, at 12:08 PM, Miroslav Stampar <mir...@gm...> wrote:
> 
> Either through request file or headers. Otherwise, sqlmap automatically sets it based on recognized type (e.g. application/json for JSON)
> 
> Bye
> 
> On Oct 19, 2015 7:05 PM, "Brandon Perry" <bpe...@gm... <mailto:bpe...@gm...>> wrote:
> Just curious, how do you expect a user to set a specific content type? Is there are argument I am missing, or is --headers the expected way?
> 
> On Mon, Oct 19, 2015 at 9:41 AM, Miroslav Stampar <mir...@gm... <mailto:mir...@gm...>> wrote:
> I really can't enforce this behavior. Waiting for your sample.
> 
> Bye
> 
> On Mon, Oct 19, 2015 at 4:04 PM, Brandon Perry <bpe...@gm... <mailto:bpe...@gm...>> wrote:
> Yes, I can, but it will have to be when I get home this evening.
> 
> FWIW, I am interacting with the sqlmap API, so not passing it a request file. I am building the SOAP XML programmatically, then setting it as the 'data' in the options (along with headers to specify text/xml and SOAPAction), with skipUrlEncode.
> 
> On Mon, Oct 19, 2015 at 9:02 AM, Miroslav Stampar <mir...@gm... <mailto:mir...@gm...>> wrote:
> Hi.
> 
> But sqlmap should automatically skip the url encoding of such request bodies if the content-type has been set to the proper value from start (or if there was no content-type from the beginning).
> 
> Can you please send a sample request file and/or used sqlmap options.
> 
> Bye
> 
> On Mon, Oct 19, 2015 at 4:00 PM, Brandon Perry <bpe...@gm... <mailto:bpe...@gm...>> wrote:
> The actual request is a SOAP payload, which requires a content type of XML, and no URL encoding (which, if performed, returns a 50x).
> 
> On Mon, Oct 19, 2015 at 6:37 AM, Miroslav Stampar <mir...@gm... <mailto:mir...@gm...>> wrote:
> Hi Brandon.
> 
> Sorry for late reply. It goes like this.
> 
> Your header value for content-type should be propagated/used, even in this case, in all cases THAN one.
> 
> If you use --skip-urlencode and you (or your request file) state that the content-type should be "urlencoded" sqlmap forces switch to either the "recognized" (e.g. json, xml,...) or the "plain". So, that line that you've pinpointed will be triggered only in described situation.
> 
> Can you please describe what are you trying to accomplish? I believe that you are trying to leave some parts (non-payload) url encoded, while you want payload to not be url encoded.
> 
> Bye
> 
> On Sun, Oct 18, 2015 at 11:35 AM, Miroslav Stampar <mir...@gm... <mailto:mir...@gm...>> wrote:
> Will patch it later today.
> 
> Bye
> 
> On Oct 17, 2015 04:32, "Brandon Perry" <bpe...@gm... <mailto:bpe...@gm...>> wrote:
> I tracked it down to ./lib/request/connect.py, line 726.
> 
> contentType = POST_HINT_CONTENT_TYPES.get(kb.postHint, PLAIN_TEXT_CONTENT_TYPE)
> 
> I am specifying a content type explicitly with —headers, so commenting this line out allowed sqlmap to detect the injections (the server returns 50x if the content type isn't right).
> 
> Not sure what the correct solution is to this, as I understand the intent. Would this be more useful as a github issue?
> ------------------------------------------------------------------------------
> _______________________________________________
> sqlmap-users mailing list
> sql...@li... <mailto:sql...@li...>
> https://lists.sourceforge.net/lists/listinfo/sqlmap-users <https://lists.sourceforge.net/lists/listinfo/sqlmap-users>
> 
> 
> 
> -- 
> Miroslav Stampar
> http://about.me/stamparm <http://about.me/stamparm>
> 
> 
> -- 
> http://volatile-minds.blogspot.com <http://volatile-minds.blogspot.com/> -- blog
> http://www.volatileminds.net <http://www.volatileminds.net/> -- website
> 
> 
> 
> -- 
> Miroslav Stampar
> http://about.me/stamparm <http://about.me/stamparm>
> 
> 
> -- 
> http://volatile-minds.blogspot.com <http://volatile-minds.blogspot.com/> -- blog
> http://www.volatileminds.net <http://www.volatileminds.net/> -- website
> 
> 
> 
> -- 
> Miroslav Stampar
> http://about.me/stamparm <http://about.me/stamparm>
> 
> 
> -- 
> http://volatile-minds.blogspot.com <http://volatile-minds.blogspot.com/> -- blog
> http://www.volatileminds.net <http://www.volatileminds.net/> -- website