Page 10 of 46 FirstFirst ... 5678910111213141520 ... LastLast
Results 181 to 200 of 913
Like Tree8Likes
  1. #181  
    Quote Originally Posted by milliman View Post
    Looks like we have to wait for the 1.2.1 SDK to see the source that needs to be modified. I wish I had Carl's source as a basis to see how he addressed the Google Maps issue. The easiest way is to exclude the URL for Google Maps.
    Clipcarl didn't make this change via source. His work is simply a binary patch of libWebKitLuna.so, replacing the default user agent string with a modified one. It doesn't add any logic though - no branching is possible based on the url - it simply replaces the original string with a new one.

    The real trick came when hofs1 found that a *******ized user agent string could allow iphone sites to recognize the phone properly and still not break Google Maps, but ClipCarl didn't use the user agent string hofs1 supplied because it was too long to simply replace the original string with. With hofs1's discovery ClipCarl continued to investigate until he found a modified version of the hofs1 user agent string that was no longer than the original string but still worked. The nature of this kind of change means it is easy if the string is equal or smaller in length than the original it's replacing. Adding a longer string or adding logic would make it vastly more complex as a binary patch. You really would need the libWebKitLuna source code for that.

    We should be able to simply patch the new libWebKitLuna.so file with the same string ClipCarl used, compute the md5sum of the new and old libWebKitLuna.so file, compute the new binary patch file and update ClipCarl's installer. It appears several people are already hot on this trail. I'm pretty sure I can do it but I'm on the road and don't have the user agent string ClipCarl used in his final solution handy. I do have it in the office though I won't be back there until Friday. Hopefully someone else will have figured it out by then.
  2. #182  
    Quote Originally Posted by Trekker View Post
    Assuming someone can get this working again, is it possible to turn it on/off from within the browser itself...or do you have to un-install the tweak?

    I was thinking that there may be times when people want to see the full version of a website, so it would be nice to have a menu option that would temporarily disable the tweak and then be able to re-enable it.
    That would be nice, but it's not possible based on ClipCarl's original solution. The user agent string isn't located in the browser code but in a shared library file instead. Adding that kind of logic would be a significant effort with all the source code, and this particular patch was made without source code and only possible because the new user agent string was shorter than the one replaced.
  3. #183  
    This is the user agent string ClipCarl used in his most recent patch that worked with Google Maps:

    Mozilla/5.0 (iPhone; U; en)(webOS/1.1; U; %s) AppleWebKit/525.27.1 Version/1.0 Safari/525.27.1 Pre/1.0
  4. #184  
    wtgreen: Makes sense that he replaced the binary to keep it the same size as not to break it although the checksum changes. Changing the binary should be easy now that you posted the string. Thanks!
  5. #185  
    Quote Originally Posted by milliman View Post
    wtgreen: Makes sense that he replaced the binary to keep it the same size as not to break it although the checksum changes. Changing the binary should be easy now that you posted the string. Thanks!
    Since this string is shorter than the original, be sure to add extra null characters to pad it out to the same length of the original. If you don't, it'll break things.
  6. #186  
    "We should be able to simply patch the new libWebKitLuna.so file with the same string ClipCarl used, compute the md5sum of the new and old libWebKitLuna.so file, compute the new binary patch file and update ClipCarl's installer. It appears several people are already hot on this trail. I'm pretty sure I can do it but I'm on the road and don't have the user agent string ClipCarl used in his final solution handy. I do have it in the office though I won't be back there until Friday. Hopefully someone else will have figured it out by then."

    Wow thanks wtgreen for fixing this as soon as you can but are we allowed? I asked a computer guy at work and he said you can't just modify programs without permission and he wouldn't do it.
  7. #187  
    You guys are amazing! Hopefully you guys won't spaz out and run away like clipcarl...

    How long do you think it'll will take before the average user can apply this, WebOS QI or copy/paste command?
  8. #188  
    Quote Originally Posted by 1KRUNCH View Post
    "We should be able to simply patch the new libWebKitLuna.so file with the same string ClipCarl used, compute the md5sum of the new and old libWebKitLuna.so file, compute the new binary patch file and update ClipCarl's installer. It appears several people are already hot on this trail. I'm pretty sure I can do it but I'm on the road and don't have the user agent string ClipCarl used in his final solution handy. I do have it in the office though I won't be back there until Friday. Hopefully someone else will have figured it out by then."

    Wow thanks wtgreen for fixing this as soon as you can but are we allowed? I asked a computer guy at work and he said you can't just modify programs without permission and he wouldn't do it.
    Although I have not read the SDK license, modifying the code should not be a problem since Palm has released WebOS to the open source community. Patching a binary from Windows 7 violates their copyright and license agreement but it is done all of the time. As long as the source and binaries are available to the public and the person that modifies the code returns it to the open source community and does not charge for it, modification is allowed. Palm has been very supportive of this community and its modifications to its functionality. We should be in the clear modifying the binary.
  9. #189  
    "Although I have not read the SDK license, modifying the code should not be a problem since Palm has released WebOS to the open source community. Patching a binary from Windows 7 violates their copyright and license agreement but it is done all of the time. As long as the source and binaries are available to the public and the person that modifies the code returns it to the open source community and does not charge for it, modification is allowed. Palm has been very supportive of this community and its modifications to its functionality. We should be in the clear modifying the binary."

    I told the guy at work this and he said that even thought webos open source that doesn't mean this tweak program is so I guess I will have to wait for wtgreen to fix it.
  10. #190  
    Quote Originally Posted by 1KRUNCH View Post
    "I told the guy at work this and he said that even thought webos open source that doesn't mean this tweak program is so I guess I will have to wait for wtgreen to fix it.
    Tweak program? You mean changing the user agent string? I don't think that guy knows what he's talking about. Unless he IS clipcarl and doesn't want anyone modding their phones without his permission.
  11. #191  
    Quote Originally Posted by 1KRUNCH View Post
    Wow thanks wtgreen for fixing this as soon as you can but are we allowed? I asked a computer guy at work and he said you can't just modify programs without permission and he wouldn't do it.
    I'm not sure exactly what you're referring to in reference to allowed, but I'll give you my take on it.

    First, are we "allowed" to modify the libWebKitLuna.so file? Technically doing so might void your warranty, as rooting your Pre and some of the other hacks around here might. I haven't read the user agreement with our Sprint contract or Palm Warranty myself so I don't know, nor do I care too much. The great thing about the Pre is it's virtually impossible to permanantly damage via software changes, thanks to Palm's WebOSDoctor. But does changing the file infringe Palm's copyright? No, I don't think so, no more so than marking out or changing text in a book you purchased does. Passing the work of on our own would, but we aren't doing that. And it's not legal to distribute copies of the libWebKitLuna.so file either - modified or not - but this patch doesn't do that. It distributes a binary diff patch file that instructs the bspatch program on how to modify the libWebKitLuna.so file, but it doesn't distribute the file itself.

    Next, are we allowed to modify the patch ClipCarl supplied? First, I doubt he cares. Second, while he does claim a copyright to the scripts, we aren't changing the logic in it or claiming credit for it. He didn't charge for the script to begin with and distributes it freely and it's still included in the OP, so clearly he's not trying to limit it's distribution, and I'll simply be modifying the md5sum and providing a slightly modified data file - the binary patch file. Third, even if he did have a valid infringment claim - which he won't - what damages could he sue for? He didn't charge anything for the original and we aren't charging anything for the modified script. And this would be worthwhile to sue someone for?

    If worse came to worse and ClipCarl contacts me and says he doesn't want me to distribute an update to his patch using his scripts, I'd honor his request and just re-write it anyway. His scripts are quite clever and I learned a lot by reading them, but there are many ways to skin a cat and I could write the same thing from scratch. Mine just wouldn't be as slick as ClipCarl's.

    Anyway, I'll provide an update later today, or at worst tomorrow. Being gone a week from work I'm a bit behind and probably ought to do something productive for a little while.
  12. #192  
    Thank you
  13. gage006's Avatar
    Posts
    543 Posts
    Global Posts
    576 Global Posts
    #193  
    Quote Originally Posted by wtgreen View Post
    Anyway, I'll provide an update later today, or at worst tomorrow. Being gone a week from work I'm a bit behind and probably ought to do something productive for a little while.
    Can't wait! I miss this patch.
  14. hofs1's Avatar
    Posts
    460 Posts
    Global Posts
    473 Global Posts
    #194  
    Thanks !!!!!!!!!
    T300 - T600 - T650 - T700p - T755p - T800w - Pre
  15. #195  
    Ok, so I've updated the patch for 1.2.1 and it seems to work fine. Follow the instructions included in the first post of this thread. As with any patch, understand things can and do go wrong, so use at your own risk.

    The patch is based on ClipCarl's excellent original work. I've simply modified the patch file for the 1.2.1 version of libWebKitLuna.so, updated the scripts to check for version 1.2.1 of webOS and repackaged it. The attached file will not work with later or earlier versions of webOS.

    The script updates the file /usr/lib/libWebKitLuna.so and creates a backup as /usr/lib/libWebKitLuna.so-iphone_user_agent.bak. When you run the script with the uninstall command, it simply restores the original file.

    If you want to be extra careful before installing, you should backup /usr/lib/libWebKitLuna.so yourself, and it's easy enough to reverse the install without using the supplied script by simply restoring your backed up version of the same file.

    Anyway remember, ClipCarl did all the hard and clever work. He deserves the credit even if he isn't around to update this for the latest webOS version. I learned a lot myself just by reading his scripts. Enjoy!
    Attached Files Attached Files
  16. gage006's Avatar
    Posts
    543 Posts
    Global Posts
    576 Global Posts
    #196  
    Yay it works! Thanks carl and wtgreen.
  17. kenford's Avatar
    Posts
    32 Posts
    Global Posts
    41 Global Posts
    #197  
    Just tried it and it works GREAT! Thanks much.
  18. jghacker's Avatar
    Posts
    53 Posts
    Global Posts
    55 Global Posts
    #198  
    This is exactly what I've been waiting for since it broke with webOS 1.2.0

    Does anyone know if it will be incorporated for install via preware soon? Love the new preware patch installs - and so much less hassle than other methods.
  19. #199  
    thank you wtgreen!
  20. #200  
    is this not the same patch as the iphone spoof agent???

Posting Permissions