I’m looking to implement a configuration and asset loader and have come upon the AppCoreLib.
On it’s face this it like a useful contribution but I’m unclear as to usage details. For example, in the sample source we see tags for both <swfs> and <urls> but nothing for image formats (gif, jpg, png).
It’d be very useful to understand the intent of ‘<urls>’ in both content.xml and config.xml.
I’m attempting to use a provider that had worked under SQL2K but is tossing an error under 05 Express Edition. I’m on a completely new server so perhaps it’s just a matter or replicating some libraries or other related files that don’t exist out of the box.
The connection string that had worked is:
“Provider=MSDataShape; Data Provider=SQLOLEDB.1;Persist Security Info=True;User Id=me;PASSWORD=foo;Database=bar”
Error reads:
80004005 [DBNETLIB][ConnectionOpen_(Invalid_Instance()).]Invalid_connection. 605
The resolution is:
“Provider=MSDataShape; Data Provider=SQLNCLI;Persist Security Info=True;User Id=me;PASSWORD=foo;Database=bar”
Data Services Made Easy for Adobe Flex Applications
— Flex has gotten popular lately because of its rich GUI capabilities. It also comes in handy with HTTPService and Web Service components connecting to back-end servers to fetch and update data. But using this mechanism to talk to the back-end server requires formulating a unique service object from the Flex side, making a request, and getting back data from the back-end either in XML or plain text format. The response data then has to be parsed and fed to the Flex objects to update the UI. For small to medium-size Flex projects it’s a viable solution, but for enterprise projects with thousands of external service calls it will get quite repetitive and could result in a lot of unmanageable, buggy code. – http://flex.sys-con.com/read/418939.htm
A really provacative idea being presented here – synchronize object structure between server-side and the client.
This is one of those ideas that entice us designer crossovers with visions of easily discoverable, accessable, post-processable server-side stuff. And if we know enough T-SQL we can really make some noise at an architectural level.
Can’t help but wonder at the lack of followup/comments by the community.
What would the reverse look like? if one were defining the server-side objects from the native Flex? Couldn’t we literally feed mxml components in a digestible way?
Posted in Flex Builder 2, Remoting
