Hi List, would please anyone review/comment this patch, Stipe?
On Monday 29 March 2004 19:14, J�rg Pommnitz wrote: > Hi List, > attached is a patch to the Kannel WML compiler that fixes the > problem described in my original post. > > With this patch applied, the following test WML document: > > <?xml version="1.0" encoding="ISO-8859-1"?> > <!DOCTYPE wml PUBLIC "-//WAPFORUM//DTD WML 1.1//EN" > "http://www.wapforum.org/DTD/wml_1.1.xml";> > <wml> > <card id="c1" title="A WMLC Testcase"> > <onevent type="onenterforward"> > <refresh> > <setvar name="XY" value="This is the content"/> > </refresh> > </onevent> > <p align="left"> > <a href="http://wap.url.mobile/path?XY=$XY";>$XY</a><br/> > <a href="http://wap.url.mobile/path?XY=$(XY)">$(XY)</a><br/> > <a > href="http://wap.url.mobile/path?XY=$(XY:escape)">$(XY:escape)</a><br/> > <a > href="http://wap.url.mobile/path?XY=$(XY:noesc)">$(XY:noesc)</a><br/> > <a > href="http://wap.url.mobile/path?XY=$(XY:unesc)">$(XY:unesc)</a><br/> > </p> > </card> > </wml> > > results in the following (decompiled) binary: > > <?xml version="1.0" encoding="ISO-8859-1"?> > <!DOCTYPE wml PUBLIC "-//WAPFORUM//DTD WML 1.1//EN" > "http://www.wapforum.org/DTD/wml_1.1.xml";> > <wml> > <card id="c1" title="A WMLC Testcase" > > <onevent type="onenterforward" > > <refresh> > <setvar name="XY" value="This is the content" /> > </refresh> > </onevent> > <p align="left" > > <a href="http://wap.url.mobile/path?XY=$(XY:escape)" >$(XY:noesc)</a> > <br /> > <a href="http://wap.url.mobile/path?XY=$(XY:escape)" >$(XY:noesc)</a> > <br /> > <a href="http://wap.url.mobile/path?XY=$(XY:escape)" >$(XY:escape)</a> > <br /> > <a href="http://wap.url.mobile/path?XY=$(XY:noesc)" >$(XY:noesc)</a> > <br /> > <a href="http://wap.url.mobile/path?XY=$(XY:unesc)" >$(XY:unesc)</a> > <br /> > </p> > </card> > </wml> > > As you can see, the compiler now chooses the right conversion for the > context, when no conversion is supplied. > > Regards > J�rg > > -----Urspr�ngliche Nachricht----- > Von: J�rg Pommnitz > Gesendet: Montag, 29. M�rz 2004 16:21 > An: Kannel-Devel (E-Mail) > Betreff: Bug in Kannel WML Compiler?? > > > Hello all, > I suspect that there is a small bug in the Kannel WML compiler. > The bug is related to variable substitution as specified in > http://www.openmobilealliance.org/tech/affiliates/wap/wap-191-wml-20000219- >a .pdf > . Section 10.3.1 Variable Substitution says: > > "If no conversion is specified, the variable is substituted using the > conversion format appropriate for the context. All attributes defined > as %HREF; default to escape conversion, elsewhere no conversion is > done. Specifying the noesc conversion disables context sensitive > escaping of a variable." > > The WML binary standard does not provide a token for a variable > substitution without conversion (the defined tokens are EXT_I_0 > ..EXT_I_2, EXT_T_0..EXT_T_2), that's why the WML compiler has to choose > the correct one. Currently the WML compiler unconditionaly uses > the "noesc" conversion for all contexts, even in the href attribute: > http://www.kannel.org/cgi-bin/lxr/source/gateway/gw/wml_compiler.c#L1107 > > Unfortunately the function "check_variable_syntax" does not even > know the context. Any suggestions? > > Regards > J�rg -- Best regards / Mit besten Gr��en aus D�sseldorf Dipl.-Ing. Alexander Malysh ___________________________________________ Centrium GmbH Vogelsanger Weg 80 40470 D�sseldorf Fon: +49 (0211) 74 84 51 80 Fax: +49 (0211) 277 49 109 email: a.malysh (at) centrium.de web: www.centrium.de msn: a.malysh (at) centrium.de icq: 98063111 ___________________________________________ Please avoid sending me Word, Excel or PowerPoint attachments. See http://www.fsf.org/philosophy/no-word-attachments.html
pgp00000.pgp
Description: signature
