#52 'IllegalStateException: Already has script with id xxx' is thrown when page is refreshed in browser

[Anonymous]

Originally created by: nik...@gmail.com

SDK/Debugger version: SDK 0.3.0 with WIP backend wk93101
Google Chrome/V8 Embedder + version: Chrome 15.0.874.106
OS + version: Ubuntu 11.04.

What steps will reproduce the problem?
1. Start Chrome with enabled remote debugging and open some page containing scripts.
2. Connect to it from Java program using WIP backend.
3. Refresh page in Chrome.

What happens instead?

The following exception in Java process is thrown (looks like SDK assumes that ScriptParsed events come after FrameNavigated event but actually they come first):

java.lang.IllegalStateException: Already has script with id 22
    at org.chromium.sdk.internal.wip.WipScriptManager.scriptIsReportedParsed(WipScriptManager.java:125)
    at org.chromium.sdk.internal.wip.WipCommandProcessor$3.accept(WipCommandProcessor.java:213)
    at org.chromium.sdk.internal.wip.WipCommandProcessor$3.accept(WipCommandProcessor.java:1)
    at org.chromium.sdk.internal.wip.WipCommandProcessor$EventMap$InternalHandler.handle(WipCommandProcessor.java:282)
    at org.chromium.sdk.internal.wip.WipCommandProcessor$EventMap.handleEvent(WipCommandProcessor.java:255)
    at org.chromium.sdk.internal.wip.WipCommandProcessor.processEvent(WipCommandProcessor.java:121)
    at org.chromium.sdk.internal.wip.WipCommandProcessor.access$3(WipCommandProcessor.java:112)
    at org.chromium.sdk.internal.wip.WipCommandProcessor$WipMessageTypeHandler.acceptNonSeq(WipCommandProcessor.java:175)
    at org.chromium.sdk.internal.wip.WipCommandProcessor$WipMessageTypeHandler.acceptNonSeq(WipCommandProcessor.java:1)
    at org.chromium.sdk.internal.BaseCommandProcessor.processIncoming(BaseCommandProcessor.java:110)
    at org.chromium.sdk.internal.wip.WipCommandProcessor.acceptResponse(WipCommandProcessor.java:105)
    at org.chromium.sdk.internal.wip.WipTabImpl$2.textMessageRecieved(WipTabImpl.java:98)
    at org.chromium.sdk.internal.websocket.WsConnection$7$1.dispatch(WsConnection.java:210)
    at org.chromium.sdk.internal.websocket.WsConnection$8.runImpl(WsConnection.java:271)
    at org.chromium.sdk.internal.websocket.WsConnection$8.run(WsConnection.java:262)
    at java.lang.Thread.run(Thread.java:662)

[Anonymous]

Originally posted by: peter.ry...@gmail.com

I tried:
Google Chrome on Windows: 15.0.874.106
Eclipse features:
    Chromium JavaScript Remote Debugger: 0.3.0....
    ChromeDevTools SDK WIP Backends: 0.1.2
Backend: WK93101

Sites:
http://jetbrains.com (several refreshes)
http://design.ru (several refreshes)

SDK does assume that FrameNavigated comes first. Have you seen any evidence of opposite?
Note that future versions of protocol wouldn't work with WK93101 because they don't send FrameNavigated at all (unless explicitly required), but not 15.0.874.*

Status: NotReproducible

[Anonymous]

Originally posted by: nikolay....@gmail.com

I've reproduced the problem using 'WebKit Protocol' debug configuration in Eclipse. Open the attached index.html file in Chrome, start debug configuration in Eclipse and refresh the page. After that the stacktrace should be printed to stdout. Also I've attached two txt files containing messages printed to 'Console' window in Eclipse before and after refresh. According to that ScriptParsed indeed comes before FrameNavigated.

[Anonymous]

Originally posted by: peter.ry...@gmail.com

This seems to be non-fatal exception. Nikolay, can you confirm this?

Status: Accepted

[Anonymous]

Originally posted by: nikolay....@gmail.com

Yes, it seems that this exception doesn't hurt.