Hi Alex,
I am aware and agree with you about de-compiling un-licensed source.
Just trying to understand what is happening inside.
I am planning to move as you suggested : The implementation can be empty
for now. Once you figure out what APIs you need, let’s take a closer
look at them and see if you really need them.
Thanks,
Serkan
8.11.2018 01:57 tarihinde Alex Harui yazdı:
Hi Serkan,
I didn’t think the Fiber code was open source either. De-compiling is
probably not in conformance with the Adobe license agreement.
If you are using HTTPService with Fiber, it is probably in your best
interest to identify the API surface of Fiber that you need and start
from empty files and fill in only what you need. De-compiling is
going to probably create a lot of unnecessary dependencies.
To do that, remove the Fiber classes and FDS.swc see what compiler
errors you get. Then only bring in the APIs the compiler complains
about, just like we are doing for many of the other emulation
components. The implementation can be empty for now. Once you figure
out what APIs you need, let’s take a closer look at them and see if
you really need them. IIRC, Fiber does a lot of wrapping but I’m not
sure you really need it.
-Alex
*From: *Serkan Taş <[email protected]>
*Reply-To: *"[email protected]" <[email protected]>
*Date: *Wednesday, November 7, 2018 at 11:01 AM
*To: *"[email protected]" <[email protected]>
*Subject: *Re: flex access modifier and usage within library
Hi Alex,
I used a de-compiler for extraction of the classes that is way i am
keeping it separately.
I am not sure how the fds is used for now, because my project does not
directly refer the library but the fiber which is i am dependent.
I am accessing to the server direct http post/get via HttpService.
If i can compile successfully fiber adn have more control over the
libraries i may have the ability to wrap all in Royale and remove fds.swc.
Note : I am waiting your confirmation for the PR.
Thanks,
Serkan.
7.11.2018 08:31 tarihinde Alex Harui yazdı:
Hi Serkan,
I’m not sure where you found the source for ConcreteDataService.
I didn’t think the code in fds.swc was open source.
Regarding your specific question, when I look at the emulation for
AsyncToken, setResult is already mx_internal and not private.
Does ConcreteDataService have:
import mx.core.mx_internal;
use namespace mx_internal;
That said, the key question for you is what part of FDS you are
using? We have RemoteObject working with AMF, but I think FDS has
other protocols. If you didn’t go with Royale, what would you
have done to access the server? Whatever that is, you should put
that together and wrap it in Royale APIs and replace fds.swc.
Thanks,
-Alex
*From: *Serkan Taş <[email protected]>
<mailto:[email protected]>
*Reply-To: *"[email protected]"
<mailto:[email protected]> <[email protected]>
<mailto:[email protected]>
*Date: *Tuesday, November 6, 2018 at 1:17 PM
*To: *"[email protected]" <mailto:[email protected]>
<[email protected]> <mailto:[email protected]>
*Subject: *flex access modifier and usage within library
Hi,
I extracted the sources of fds.swc and library class
*ConcreteDataService.as *uses method *mx_internal function
setResult(newResult:Object):void* .
public function applyChangedItems(changedItems:ChangedItems) :
AsyncToken
{
var applyToken:AsyncTokenChain = null;
var refreshAllLocalFills:Function = function():void
{
var currDataList:DataList = null;
for each(currDataList in _dataLists.toArray())
{
applyToken.addPendingInvocation(refreshLocalFill,currDataList);
}
};
applyToken = new AsyncTokenChain();
applyToken.setResult(changedItems);
According to the definition of the method in***mx.rpc.AsyncToken*
the function is private and i can not compile the fds.
In flex, the fsd.swc is included in my project and everything is
working fine. How can this happen any idea ?
Thanks,
Serkan
