The staff object is 75% complete. Friday was a learning experience for Maria. A step was overlooked and the scan caught it. I tried tweaking the application object but that didn't work. To keep it from becoming an exercise in futility I decided I would start from square one. Today I made another snAPPShot and recreated the application object. I included the catalog search and set the default homepage to ACPL.
The application distributed flawlessly onto a computer with no previous Firefox browser installed. It also went well on top of our old Firefox browser (Firefox 1.0.6). Speed bump is Firefox 2.X. When distributing the application to a computer with Firefox 2.X I get a cannot read configuration error. I have looked at a clean install of 3.0 vs 2.x. With 1.06 I had to manipulate the registry. These settings carried through with 2.x. However, I don't think they are needed for 3.0.
More testing is needed and I am not confident the answer is in the registry. I did finally get it to install over 2.x on my computer but I think it was a fluke. I will see on Tuesday.
One problem is that I created the first Firefox object so staff could update it. So there may be many flavors of Firefox out there. Not a wise choice as I look back on it now. But I am hoping I am, once again, overlooking the obvious, making a mountain out of a mole hill. If someone writes a book about how the main character does it the hard way. That character would be me.
2 comments:
So in the case of staff machines with updated Firefoxes with 2.x, does removing Firefox completely before installing the thing you're working on bypass the problem? Just wondering.
Sorry this is late but it effects our current deployment of 3.x. The runtime environment is updated with each update to Firefox. This means the public profile is looking for the previous environment and it cannot locate it so it will not launch. I am working to fix this problem. I plan to post more about it later this month.
Post a Comment