well i was getting more into this issue, and i figured out the the whole string table is 180Kb and most of it is just needed. using the -XdisableClassMetada flag reduced the code by 35Kb, ie still like 8% of the whole project. additionally the -XdisableCastChecking flag reduced the code by 8Kb more, so at the end i am saving 10%!!
the things i thought was a css class name lookup table was just a list of (mostly obfuscated) class names, i just got confused by the external class names. > My point was that this particular feature is a > performance issue and in no way a security issue. sure, i understand, still to carry 40Kb+ of data for nothing is not good... thanks Michael On May 17, 10:26 pm, Sripathi Krishnan <sripathi.krish...@gmail.com> wrote: > > ... it is 100% server-side where to use reflection is ok. > > Correct. I was only trying to say its worthless to think of obfuscation as a > security measure, because there are several ways to bypass it, syncproxy > being one of them. My point was that this particular feature is a > performance issue and in no way a security issue. > > --Sri > > On 18 May 2010 01:35, mmoossen <mmoos...@gmail.com> wrote: > > > > > Hi again! > > > btw, i checked the synproxy project, which looks interesting but as i > > understood it is 100% server-side where to use reflection is ok. > > > thanks for sharing > > Michael > > > On May 17, 4:09 pm, Sripathi Krishnan <sripathi.krish...@gmail.com> > > wrote: > > > > Could somebody please explain to me why this is needed? > > > > AFAIK, this wasn't always the case. Issue > > > 370<http://code.google.com/p/google-web-toolkit/issues/detail?id=370> > > > has > > > some information on why this was introduced. The class name are required > > so > > > that you can use the getClass() method on an object. > > > > As Olivier pointed out, there is a way to disable this behavior. All > > google > > > websites I have seen disable class names. You can take a look at orkut or > > > wave for example. I believe it isn't being done by default because it > > could > > > break some websites that depend on classnames. If you don't depend on > > > getClass(), then you could benefit from the optimization that > > > -XdisableClassMetada provides. > > > > That said, there are other ways to extract out information about classes > > and > > > methods. For example, it is possible to extract the complete signature of > > a > > > RPC method and reverse engineer the RemoteService interface, such that > > you > > > can use a library like > > > syncproxy< > >http://www.gdevelop.com/w/blog/2010/01/10/testing-gwt-rpc-services/>to > > > make RPC calls to any server. So, treat > > > -XdisableClassMetada as a way to improve performance, and not as a way to > > > completely obfuscate all class and method names. > > > > --Sri > > > > On 17 May 2010 18:15, Olivier Monaco <olivier.mon...@free.fr> wrote: > > > > > Hi, > > > > > Maybe it's about this : > > > >http://code.google.com/p/google-web-toolkit/wiki/NoClassMetadataOptim. > > .. > > > > > Olivier > > > > > On 17 mai, 12:18, mmoossen <mmoos...@gmail.com> wrote: > > > > > Dear all! > > > > > > i just found this:http://degwt.googlecode.com > > > > > > and i was really surprised to actually find a classname table in the > > > > > generated production js code. > > > > > > i know that the rpc classes and methods are used (which i do not like > > > > > very much, but that is another story). > > > > > > but i do not see any need for a general classname table, i mean every > > > > > single class even enums are listed there. > > > > > additionally there seems to be also a kind of lookup table for style > > > > > names!!? > > > > > > could somebody please explain to me why this is needed? > > > > > > i mean all this data is taking about 180Kb of 500Kb of my production > > > > > cache files, and i would really like to know what is the idea behind > > > > > that... > > > > > > Thanks > > > > > Michael > > > > > > -- > > > > > You received this message because you are subscribed to the Google > > Groups > > > > "Google Web Toolkit" group. > > > > > To post to this group, send email to > > google-web-toolkit@googlegroups.com > > > > . > > > > > To unsubscribe from this group, send email to > > > > google-web-toolkit+unsubscr...@googlegroups.com<google-web-toolkit%2bunsubscr...@googlegroups.com> > > <google-web-toolkit%2bunsubscr...@googlegroups.com<google-web-toolkit%252bunsubscr...@googlegroups.com> > > > > > . > > > > > For more options, visit this group athttp:// > > > > groups.google.com/group/google-web-toolkit?hl=en. > > > > > -- > > > > You received this message because you are subscribed to the Google > > Groups > > > > "Google Web Toolkit" group. > > > > To post to this group, send email to > > google-web-tool...@googlegroups.com. > > > > To unsubscribe from this group, send email to > > > > google-web-toolkit+unsubscr...@googlegroups.com<google-web-toolkit%2bunsubscr...@googlegroups.com> > > <google-web-toolkit%2bunsubscr...@googlegroups.com<google-web-toolkit%252bunsubscr...@googlegroups.com> > > > > > . > > > > For more options, visit this group at > > > >http://groups.google.com/group/google-web-toolkit?hl=en. > > > > -- > > > You received this message because you are subscribed to the Google Groups > > "Google Web Toolkit" group. > > > To post to this group, send email to google-web-toolkit@googlegroups.com > > . > > > To unsubscribe from this group, send email to > > google-web-toolkit+unsubscr...@googlegroups.com<google-web-toolkit%2bunsubscr...@googlegroups.com> > > . > > > For more options, visit this group athttp:// > > groups.google.com/group/google-web-toolkit?hl=en. > > > -- > > You received this message because you are subscribed to the Google Groups > > "Google Web Toolkit" group. > > To post to this group, send email to google-web-tool...@googlegroups.com. > > To unsubscribe from this group, send email to > > google-web-toolkit+unsubscr...@googlegroups.com<google-web-toolkit%2bunsubscr...@googlegroups.com> > > . > > For more options, visit this group at > >http://groups.google.com/group/google-web-toolkit?hl=en. > > -- > You received this message because you are subscribed to the Google Groups > "Google Web Toolkit" group. > To post to this group, send email to google-web-tool...@googlegroups.com. > To unsubscribe from this group, send email to > google-web-toolkit+unsubscr...@googlegroups.com. > For more options, visit this group > athttp://groups.google.com/group/google-web-toolkit?hl=en. -- You received this message because you are subscribed to the Google Groups "Google Web Toolkit" group. To post to this group, send email to google-web-tool...@googlegroups.com. To unsubscribe from this group, send email to google-web-toolkit+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/google-web-toolkit?hl=en.
