|
CVS notes? Any objection to starting a thread for CVS notes and comments while we are between betas? |
GUI notes All comments based on a G3 iBook running OS 10.3.9. I've updated the CVS each evening this past week, and run as an Ultrapeer. I know Tiger has fixed quite a few java bugs on OS X (like the Vmem problem), but here are some notes. Good to see Clear History implemented, though it would look less odd as a standard "Clear Menu" item menu, rather than the button. I like the "Search More." Woud prefer the keyword to be sent to the search box (for editing before searching), though. I hope the repeat search "search for keyword" will be extended to files in the download pane too. Better yet--implement "send to search box" for creating a slight variation of the search. Download pane--hope to see the tooltip extended to show truncated text in Name and Status columns. Actually, it would be nice if any truncated text could be shown in tooltips, no matter what Pane is being viewed. Copying entries: Like in a spreadsheet, allow a way to only copy the "cell(s)" selected, or the entire row(s). Monitor Pane: Pulsing blue bars should stop pulsing for complete uploads (as they do for 100% complete progress bars) All Panes: discontinuous selects should toggle the status of the command-clicked item. Currently, select-all and command-clicking an item to deselect it will deselt all the others and select the one that should be deselected. Search Pane: trying to close a search tab when several are open is often not immediate: several clicks and pauses may be needed. In general, closing search tabs has responded inconsistenty for a while Hope to see truncation of the text details for Direct Connect (overflow could be shown in tooltip). On a small size window, it can look really ugly. |
terminal notes Terminal fills up with "gnutella.ConnectionWatchdog.run - Potential dud: CONNECTION: host=<IP:Port> I get batches of 20 or so of these every minute or less (>1200/hour). Just wondering how these are triggered. |
SocketException: Invalid argument (anyone else getting dozens of these?) LimeWire version @version@ Java version 1.4.2_05 from Apple Computer, Inc. Mac OS X v. 10.3.9 on ppc Free/total memory: 7249536/25591808 java.lang.Error: java.net.SocketException: Invalid argument at sun.nio.ch.Net.localAddress(Net.java:127) at sun.nio.ch.SocketChannelImpl.localAddress(SocketCh annelImpl.java:390) at sun.nio.ch.SocketAdaptor.getLocalAddress(SocketAda ptor.java:142) at com.limegroup.gnutella.io.NIOSocket.getLocalAddres s(NIOSocket.java:380) at com.limegroup.gnutella.Acceptor.run(Acceptor.java: 577) at java.lang.Thread.run(Thread.java:552) at com.limegroup.gnutella.util.ManagedThread.managedR un(ManagedThread.java:60) at com.limegroup.gnutella.util.ManagedThread.run(Mana gedThread.java:49) Caused by: java.net.SocketException: Invalid argument at sun.nio.ch.Net.localInetAddress(Native Method) at sun.nio.ch.Net.localAddress(Net.java:124) ... 7 more |
couple of bugs maybe? after running overnight, several times the quit menu item is disabled. Quitting from the dock or Activity Monitor doesn't work either, even after waiting several minutes. Force-quit works. Connections Pane: Rarely see 30 leaves any more--often for hours there are less than a dozen leaves connected. |
This is as good a place as any for CVS notes. Hopefully we can figure out some way to distinguish between different days of CVS builds. Re: 'Clear History' -- my thought was that making it a menu item would confuse people because the actual entries are menu items. What happens if someone actually searches for "Clear History" -- then you see two menu items with that title. How do you distinguish between them? Re: 'Search More' -- perhaps below 'Repeat Search' there could be an option to 'Send To Search Box' which could be checked? Re: 'Search More in Downloads' -- Yup, that'll come. It was on our wishlist page and someone asked to collect the bounty on it, so we didn't implement it to give them a chance. I believe the patch is pending, we delayed a bit in order for the existing 'Search More' stuff to get merged in, and still are waiting on some changes to the downloader code, at which point we'll be ready to merge it in. Re: 'Truncated Text' -- It already does that. :) You just need to turn off 'Extended Tooltips' -- the extended info takes priority over extending truncated text. I suppose we could add the truncated info prior to the extended info, but that might look a little weird in some cases. Re: 'Copying Entries' -- It doesn't make much sense in LimeWire to select single cells, or whole columns. Although the info is displayed as a table, we're trying to get away from the "spreadsheet" look. Unfortunately, because selecting cells/columns really only makes sense for copying the information, we've disabled it (because it simplifies many other things and makes LimeWire overall easier to use). I believe you already can copy an entire row by selecting it and using your OS's copy shortcut (Ctrl-C on windows, Command-C on OSX). Re: 'Pulsing bars on Uploads' -- I don't believe Apple offers a way to do that for individual progressbars. It seems like they pulse if the status isn't 100%. Would setting all complete uploads to 100% work? Re: 'Discontinuous selects' -- Interesting. Not sure what's causing that. We'll take a look into it. Re: 'Odd Search Close Behaviour' -- I've noticed that on OSX also. I don't think there's anyway around it. On Windows, closing a tab works immediately when you click the 'X'. On OSX, the tab needs to be selected first. Re: 'Direct Connect Overflow' -- I've only noticed that when you make the app unbearably small. There are some things which we just can't support, because the app itself is pretty useless at that size anyway. If, however, it happens on a normal resolution like 800x600, then we could make the text scrollable. Re: 'Potential dud warning' -- That's just a message saying that a connection hasn't sent a message in awhile. LimeWire will send them a heartbeat ping to see if they're alive, and if there's no further messages sent from that host, we kill them (which will have a different message printed out). Re: 'Lots of java.lang.Errors' -- That's a bug in OSX's NIO implementation. It kinda sucks, but I'm not sure what to do. It's an error retrieving the localhost address. Are you able to maintain any connections or start any downloads when this happens? Can anyone connect to you? Re: 'Quit menu item disabled' -- That's odd. Does the UI refresh itself, or is everything frozen? Can you post the stacktrace printed out by 'kill -QUIT <pid>' on LimeWire (where <pid> is the process ID of LimeWire). Re: 'Not many leaves' -- How long have you been seeing little leaves? Thanks. |
Thanks for the comments Sam. for CVS builds, I log the terminal's date and time of the update. e.g "Sun Jun 5 10:48:01." Better yet, can the sample bug report add a line "last updated"? Re: 'Clear History', the standard in OSX works well across all the apps that use a menu clear: the "Clear Recent" or "Clear Menu" item is alway at the bottom and separated by a line (I think it's called a menu divider). This should work for LW, and should be always available, but greyed out if no entries are available as a confirmation that the history has been cleared. Re: 'Search More', the checkmark item for sending the term to the search box would be useful. good to see the bounty program working, though I can see the politics could be a bit tricky ;) Re: 'Truncated Text' -- thanks for the tip about extended tooltips overriding truncated tips. What about just having the extended info display over one of the default columns, like "Status" for the downloads, monitor, and connection panes, and truncated tips for the others? Re: 'Copying Entries' --sorry to hear about moving away from the spreadsheet look. As a new user, leveraging spreadsheets experience helped me get going with LW. iTunes uses it successfully too, so there's more leverage. The keyboard copying works, but I have to dump the rows into a spreadsheet to get the info, and edit out the unnecessary cells like the icon info. I was just looking to save a few steps, like when I wanted to copy just a filename to paste and edit in the keyword search box (especially when results are in Asian scripts). Re: 'Pulsing bars on Uploads' -- keep the current behaviour: I find it useful to see that some uploads are complete, yet not 100%.This gives a quick sense of what files swarmed, and which ones are unique or uploaded through browse-host. Well worth the minor irritation of the pulsing. Re: 'Odd Search Close Behaviour' -- Like Windows, closing a window works immediately when I click the red dot in Safari and other apps like the Finder. Just not in LW, but I haven't checked if it's the same with other java apps. Re: 'Lots of java.lang.Errors' -- I can maintain connections, and transfers continue when this happens. The connections Pane shows others connecting too. However, the dialog prevents switching panes or using any menu items. Re: 'Quit menu item disabled' -- The UI refreshes (Connection info updates, IIRC)and I'll post the stacktrace printed out by 'kill -QUIT <pid>' the next time it happens. I suspect it's a low hard drive space condition,since it hasn't happened with this last session. BTW, I noticed inconsistent greying menu behaviour is similar when Help->Generate Error is chosen. Re: 'Not many leaves' -- Not sure how long I have have been seeing fewer leaves. A few weeks at least, especially since I hadn't run LW's for a couple of weeks. Still, this last session (18 hrs), I never saw more than 29 leaves. cheers. |
The past two evenings I haven't been able to run the CVS version after updating. Here's the error Quote:
|
You can generally fix those errors by doing an "ant clean; ant" in both core & gui. |
Thanks. Did that, ran into java.lang.ExceptionInInitializerError, but after another update a few hours later, it's now working. btw--I ran the release version yesterday, saw few leafs there too for the several hours. After 8 hours, 28 leafs were active, but it took quite a while. Also, I switched skins to the default for the purposes of a screenshot, but when I switched back to the OS X Pinstripes, the search tabs remained in the Windows mode. I couldn't get them to back to the Pinstripe tabs before I had to shut down. |
Quit menu item disabled Here's the stacktrace (CVS version updated Tue Jun 7 23:09:58 CST) "IncommingConnectionThread" daemon prio=5 tid=0x1494a210 nid=0x15634400 in Object.wait() [f1346000..f1346b20] at java.lang.Object.wait(Native Method) at java.lang.Object.wait(Object.java:429) at com.limegroup.gnutella.io.BufferOutputStream.flush (BufferOutputStream.java:91) - locked <0x61d0f70> (a java.lang.Object) at com.limegroup.gnutella.util.ThrottledOutputStream. flush(ThrottledOutputStream.java:83) at com.limegroup.gnutella.Connection.writeLine(Connec tion.java:933) at com.limegroup.gnutella.Connection.sendHeaders(Conn ection.java:836) at com.limegroup.gnutella.Connection.concludeIncoming Handshake(Connection.java:740) at com.limegroup.gnutella.Connection.initializeIncomi ng(Connection.java:705) at com.limegroup.gnutella.Connection.initialize(Conne ction.java:473) at com.limegroup.gnutella.ManagedConnection.initializ e(ManagedConnection.java:295) at com.limegroup.gnutella.ConnectionManager.initializ eExternallyGeneratedConnection(ConnectionManager.j ava:1786) at com.limegroup.gnutella.ConnectionManager.acceptCon nection(ConnectionManager.java:334) at com.limegroup.gnutella.Acceptor$ConnectionDispatch Runner.run(Acceptor.java:692) at java.lang.Thread.run(Thread.java:552) at com.limegroup.gnutella.util.ManagedThread.managedR un(ManagedThread.java:60) at com.limegroup.gnutella.util.ManagedThread.run(Mana gedThread.java:49) "ConnectionDispatchRunner" daemon prio=5 tid=0x14949490 nid=0x13902400 in Object.wait() [f1040000..f1040b20] at java.lang.Object.wait(Native Method) at com.limegroup.gnutella.io.BufferInputStream.waitIm pl(BufferInputStream.java:127) at com.limegroup.gnutella.io.BufferInputStream.read(B ufferInputStream.java:87) - locked <0x7a8a850> (a java.lang.Object) at java.io.BufferedInputStream.fill(BufferedInputStre am.java:183) at java.io.BufferedInputStream.read(BufferedInputStre am.java:201) - locked <0x7a8acc8> (a java.io.BufferedInputStream) at com.limegroup.gnutella.util.IOUtils.readWord(IOUti ls.java:107) at com.limegroup.gnutella.UploadManager.acceptUpload( UploadManager.java:380) at com.limegroup.gnutella.Acceptor$ConnectionDispatch Runner.run(Acceptor.java:704) at java.lang.Thread.run(Thread.java:552) at com.limegroup.gnutella.util.ManagedThread.managedR un(ManagedThread.java:60) at com.limegroup.gnutella.util.ManagedThread.run(Mana gedThread.java:49) "HttpClient-IdleConnectionThread" daemon prio=5 tid=0x005fc7d0 nid=0x13f5f800 in Object.wait() [f15cb000..f15cbb20] at java.lang.Object.wait(Native Method) - waiting on <0x6fd26b8> (a org.apache.commons.httpclient.MultiThreadedHttpCon nectionManager$IdleConnectionThread) at org.apache.commons.httpclient.MultiThreadedHttpCon nectionManager$IdleConnectionThread.run(MultiThrea dedHttpConnectionManager.java:768) - locked <0x6fd26b8> (a org.apache.commons.httpclient.MultiThreadedHttpCon nectionManager$IdleConnectionThread) "HttpClient-ReferenceQueueThread" daemon prio=5 tid=0x005fc670 nid=0x13eb3200 in Object.wait() [f13c7000..f13c7b20] at java.lang.Object.wait(Native Method) - waiting on <0x6fd2748> (a java.lang.ref.ReferenceQueue$Lock) at java.lang.ref.ReferenceQueue.remove(ReferenceQueue .java:111) - locked <0x6fd2748> (a java.lang.ref.ReferenceQueue$Lock) at java.lang.ref.ReferenceQueue.remove(ReferenceQueue .java:127) at org.apache.commons.httpclient.MultiThreadedHttpCon nectionManager$ReferenceQueueThread.run(MultiThrea dedHttpConnectionManager.java:829) "MessageDispatch" daemon prio=5 tid=0x0058e3a0 nid=0x1f89200 in Object.wait() [f1a54000..f1a54b20] at java.lang.Object.wait(Native Method) at com.limegroup.gnutella.util.ProcessingQueue$Proces sor.run(ProcessingQueue.java:114) - locked <0x680ef08> (a com.limegroup.gnutella.util.ProcessingQueue) at java.lang.Thread.run(Thread.java:552) at com.limegroup.gnutella.util.ManagedThread.managedR un(ManagedThread.java:60) at com.limegroup.gnutella.util.ManagedThread.run(Mana gedThread.java:49) "DestroyJavaVM" prio=5 tid=0x00500d90 nid=0x1801200 waiting on condition [0..f0800270] "DaapServerThread" daemon prio=5 tid=0x005757a0 nid=0x1c1f200 runnable [f164c000..f164cb20] at sun.nio.ch.PollArrayWrapper.poll0(Native Method) at sun.nio.ch.PollArrayWrapper.poll(PollArrayWrapper. java:100) at sun.nio.ch.PollSelectorImpl.doSelect(PollSelectorI mpl.java:64) at sun.nio.ch.SelectorImpl.lockAndDoSelect(SelectorIm pl.java:59) - locked <0x6a41540> (a java.util.HashSet) - locked <0x6a41550> (a java.util.HashSet) - locked <0x6a414f0> (a sun.nio.ch.PollSelectorImpl) at sun.nio.ch.SelectorImpl.select(SelectorImpl.java:7 0) at de.kapsi.net.daap.nio.DaapServerNIO.process(DaapSe rverNIO.java:639) at de.kapsi.net.daap.nio.DaapServerNIO.run(DaapServer NIO.java:718) at java.lang.Thread.run(Thread.java:552) at com.limegroup.gnutella.util.ManagedThread.managedR un(ManagedThread.java:60) at com.limegroup.gnutella.gui.DaapManager$1.managedRu n(DaapManager.java:194) at com.limegroup.gnutella.util.ManagedThread.run(Mana gedThread.java:49) "JmDNS.RecordReaper" prio=5 tid=0x0056c1e0 nid=0x1d9ea00 in Object.wait() [f11c3000..f11c3b20] at java.lang.Object.wait(Native Method) - waiting on <0x6a1f6c0> (a javax.jmdns.JmDNS) at javax.jmdns.JmDNS$RecordReaper.run(Unknown Source) - locked <0x6a1f6c0> (a javax.jmdns.JmDNS) at java.lang.Thread.run(Thread.java:552) "JmDNS.SocketListener" prio=5 tid=0x0056be60 nid=0x1da8200 runnable [f1142000..f1142b20] at java.net.PlainDatagramSocketImpl.receive(Native Method) - locked <0x6a02cb0> (a java.net.PlainDatagramSocketImpl) at java.net.DatagramSocket.receive(DatagramSocket.jav a:711) - locked <0x6a24a98> (a java.net.DatagramPacket) - locked <0x6a215d0> (a java.net.MulticastSocket) at javax.jmdns.JmDNS$SocketListener.run(Unknown Source) at java.lang.Thread.run(Thread.java:552) "Thread-7" daemon prio=5 tid=0x00569140 nid=0x1938400 runnable [0..f0f3e9d0] "ManualGC" daemon prio=5 tid=0x00566060 nid=0x1d4bc00 waiting on condition [f154a000..f154ab20] at java.lang.Thread.sleep(Native Method) at com.limegroup.gnutella.RouterService$1.run(RouterS ervice.java:428) at java.lang.Thread.run(Thread.java:552) at com.limegroup.gnutella.util.ManagedThread.managedR un(ManagedThread.java:60) at com.limegroup.gnutella.util.ManagedThread.run(Mana gedThread.java:49) "HTTPAcceptor" daemon prio=5 tid=0x00565290 nid=0x1d28c00 runnable [f14c9000..f14c9b20] at java.net.PlainSocketImpl.socketAccept(Native Method) at java.net.PlainSocketImpl.accept(PlainSocketImpl.ja va:353) - locked <0x69d36f0> (a java.net.PlainSocketImpl) at java.net.ServerSocket.implAccept(ServerSocket.java :448) at java.net.ServerSocket.accept(ServerSocket.java:419 ) at com.limegroup.gnutella.browser.HTTPAcceptor.run(HT TPAcceptor.java:174) - locked <0x69d37c0> (a java.lang.Object) at java.lang.Thread.run(Thread.java:552) at com.limegroup.gnutella.util.ManagedThread.managedR un(ManagedThread.java:60) at com.limegroup.gnutella.util.ManagedThread.run(Mana gedThread.java:49) "QueryUnicaster" daemon prio=5 tid=0x00565090 nid=0x1d28800 in Object.wait() [f1448000..f1448b20] at java.lang.Object.wait(Native Method) - waiting on <0x675e940> (a java.util.Hashtable) at java.lang.Object.wait(Object.java:429) at com.limegroup.gnutella.QueryUnicaster.waitForQueri es(QueryUnicaster.java:312) - locked <0x675e940> (a java.util.Hashtable) at com.limegroup.gnutella.QueryUnicaster.queryLoop(Qu eryUnicaster.java:228) at com.limegroup.gnutella.QueryUnicaster.access$000(Q ueryUnicaster.java:37) at com.limegroup.gnutella.QueryUnicaster$1.managedRun (QueryUnicaster.java:192) at com.limegroup.gnutella.util.ManagedThread.run(Mana gedThread.java:49) "Acceptor" daemon prio=5 tid=0x00563400 nid=0x1d08c00 in Object.wait() [f12c5000..f12c5b20] at java.lang.Object.wait(Native Method) - waiting on <0x6769728> (a java.lang.Object) at com.limegroup.gnutella.io.NIOServerSocket.accept(N IOServerSocket.java:99) - locked <0x6769728> (a java.lang.Object) at com.limegroup.gnutella.Acceptor.run(Acceptor.java: 532) - locked <0x675ac88> (a java.lang.Object) at java.lang.Thread.run(Thread.java:552) at com.limegroup.gnutella.util.ManagedThread.managedR un(ManagedThread.java:60) at com.limegroup.gnutella.util.ManagedThread.run(Mana gedThread.java:49) "MulticastService" daemon prio=5 tid=0x00563220 nid=0x1d04200 runnable [f1244000..f1244b20] at java.net.PlainDatagramSocketImpl.receive(Native Method) - locked <0x6769550> (a java.net.PlainDatagramSocketImpl) at java.net.DatagramSocket.receive(DatagramSocket.jav a:711) - locked <0x78630c0> (a java.net.DatagramPacket) - locked <0x67689f8> (a java.net.MulticastSocket) at com.limegroup.gnutella.MulticastService.run(Multic astService.java:190) - locked <0x6768a28> (a java.lang.Object) at java.lang.Thread.run(Thread.java:552) at com.limegroup.gnutella.util.ManagedThread.managedR un(ManagedThread.java:60) at com.limegroup.gnutella.util.ManagedThread.run(Mana gedThread.java:49) "QRPPropagator" daemon prio=5 tid=0x00563020 nid=0x1d03e00 waiting on condition [f0ebd000..f0ebdb20] at java.lang.Thread.sleep(Native Method) at com.limegroup.gnutella.MessageRouter$QRPPropagator .managedRun(MessageRouter.java:2423) at com.limegroup.gnutella.util.ManagedThread.run(Mana gedThread.java:49) "Thread-3" daemon prio=5 tid=0x00529000 nid=0x18e1e00 runnable [f10c1000..f10c1b20] at java.lang.Object.wait(Native Method) - waiting on <0x675a640> (a java.util.TaskQueue) at java.util.TimerThread.mainLoop(Timer.java:429) - locked <0x675a640> (a java.util.TaskQueue) at java.util.TimerThread.run(Timer.java:382) "TimerQueue" daemon prio=5 tid=0x005249a0 nid=0x1971e00 runnable [f0fbf000..f0fbfb20] at java.lang.Object.wait(Native Method) at javax.swing.TimerQueue.run(TimerQueue.java:231) - locked <0x673cd20> (a javax.swing.TimerQueue) at java.lang.Thread.run(Thread.java:552) "NIODispatcher" prio=5 tid=0x00520a10 nid=0x18e7c00 waiting on condition [f0c08000..f0c08b20] at java.lang.Thread.sleep(Native Method) at com.limegroup.gnutella.io.NIODispatcher.process(NI ODispatcher.java:342) at com.limegroup.gnutella.io.NIODispatcher.run(NIODis patcher.java:424) at java.lang.Thread.run(Thread.java:552) at com.limegroup.gnutella.util.ManagedThread.managedR un(ManagedThread.java:60) at com.limegroup.gnutella.util.ManagedThread.run(Mana gedThread.java:49) "AWT-EventQueue-0" prio=5 tid=0x005119d0 nid=0x18e2200 in Object.wait() [f0e3c000..f0e3cb20] at java.lang.Object.wait(Native Method) at java.lang.Object.wait(Object.java:429) at java.awt.EventQueue.getNextEvent(EventQueue.java:3 39) - locked <0x6668bb8> (a java.awt.EventQueue) at java.awt.EventDispatchThread.pumpOneEventForHierar chy(EventDispatchThread.java:195) at java.awt.EventDispatchThread.pumpEventsForHierarch y(EventDispatchThread.java:184) at java.awt.EventDispatchThread.pumpEvents(EventDispa tchThread.java:178) at java.awt.EventDispatchThread.pumpEvents(EventDispa tchThread.java:170) at java.awt.EventDispatchThread.run(EventDispatchThre ad.java:100) "AWT-Shutdown" prio=5 tid=0x00511520 nid=0x18d4400 in Object.wait() [f0dbb000..f0dbbb20] at java.lang.Object.wait(Native Method) - waiting on <0x666b430> (a java.lang.Object) at java.lang.Object.wait(Object.java:429) at sun.awt.AWTAutoShutdown.run(AWTAutoShutdown.java:2 59) - locked <0x666b430> (a java.lang.Object) at java.lang.Thread.run(Thread.java:552) "Java2D Disposer" daemon prio=10 tid=0x00510bd0 nid=0x18cb400 in Object.wait() [f0d3a000..f0d3ab20] at java.lang.Object.wait(Native Method) - waiting on <0x666b4a0> (a java.lang.ref.ReferenceQueue$Lock) at java.lang.ref.ReferenceQueue.remove(ReferenceQueue .java:111) - locked <0x666b4a0> (a java.lang.ref.ReferenceQueue$Lock) at java.lang.ref.ReferenceQueue.remove(ReferenceQueue .java:127) at sun.java2d.Disposer.run(Disposer.java:116) at java.lang.Thread.run(Thread.java:552) "AWT-AppKit" daemon prio=5 tid=0x0050c540 nid=0xa000b2a4 runnable [0..bfffe678] "Signal Dispatcher" daemon prio=10 tid=0x00508230 nid=0x1809c00 waiting on condition [0..0] "Finalizer" daemon prio=8 tid=0x005068a0 nid=0x180e000 in Object.wait() [f0a04000..f0a04b20] at java.lang.Object.wait(Native Method) - waiting on <0x6657b40> (a java.lang.ref.ReferenceQueue$Lock) at java.lang.ref.ReferenceQueue.remove(ReferenceQueue .java:111) - locked <0x6657b40> (a java.lang.ref.ReferenceQueue$Lock) at java.lang.ref.ReferenceQueue.remove(ReferenceQueue .java:127) at java.lang.ref.Finalizer$FinalizerThread.run(Finali zer.java:159) "Reference Handler" daemon prio=10 tid=0x00506570 nid=0x1802200 in Object.wait() [f0983000..f0983b20] at java.lang.Object.wait(Native Method) - waiting on <0x6657ba8> (a java.lang.ref.Reference$Lock) at java.lang.Object.wait(Object.java:429) at java.lang.ref.Reference$ReferenceHandler.run(Refer ence.java:115) - locked <0x6657ba8> (a java.lang.ref.Reference$Lock) "VM Thread" prio=5 tid=0x00505df0 nid=0x1802e00 runnable "VM Periodic Task Thread" prio=10 tid=0x005081a0 nid=0x181a200 waiting on condition "Exception Catcher Thread" prio=10 tid=0x00500ef0 nid=0x1802a00 runnable |
No Share Menu items ArrayIndexOutOfBoundsException Tonight's build (updated Thu Jun 9 19:17:25 CST -6) just has the regular menu items--without the extra commands like "Add Folder to Share" Clicking a file in the Library triggers the following exception. Quote:
|
Pinstripe skin problems 1 Attachment(s) I can't reliably refresh the Pinstripe skin with the builds for the last two nights--the search tabs in particular retain characteristics of the previously applied skin. Here's an example of the Pin skin after trying the Red. |
relief/irritation btw--I've only had one of the java.lang.Error: java.net.SocketException: Invalid argument errors tonight (while connecting after startup). Did you find a way to fix this on Panther, Sam, or is it just being silently discarded? Either way, I don't miss the irritation of frequnetly discarding those errors :) Thanks --on the other hand, I was hoping today's Library changes would stop sharing files downloaded to the UNshared download folder, with no way within LimeWire to prevent them being shared. I assume Justin knows about this, but just in case it's not on the agenda yet . . . |
It's a new option in 'Sharing' options. You can also right-click on the file in the library and choose 'Stop Sharing'. Those errors may also go away if you do a clean compile. It may be some old constants that weren't refreshed. Those particular items not updating with theme changes are long-standing bugs. Not high on the list of things to be fixed, but they will be fixed at some point. We haven't done anything in particular to get rid of the java.lang.Errors. We're going to have to, most likely, prior to releasing. |
I thought I did a clean compile (ant clean && ant for core and gui) :confused: time to man cvs Anyway, I checked the new sharing options, but couldn't find a way to stop the media folders from being automatically shared. The stop sharing item in the contextual menu is greyed out currently, but I'll keep checking. I like to check out a file before sharing it back, so I hope that will be an available option. re the SocketException: Invalid argument errors, they're back again to their usual frequency. Perhaps restarting the iBook earlier today made an initial difference. Thanks for the tips. |
Re: Library table AIOOBE stief, Thanks for finding this! Seems to be a little bit of a problem when running without the media player. Should be easy to fix by this afternoon. In the meantime, things should be ok if you turn the media player on in the Options. Justin |
Re: Library table AIOOBE stief, I just merged in a fix to your problem when selecting a library file. Please let me know if it still doesn't work for you after updating! =) Justin |
Thanks Justin--no more errors with no player enabled (build Fri Jun 10 16:31:32 CST-6). Also, the multiple selection "stop sharing file" works too. However, I still can't see how to disable the automatic sharing of downloaded files. Could the "stop sharing folder" be enabled for the Individually Shared Files " item? Better yet, Individually Shared Files should respect the shared status of the download folder. thanks (the Library sure shows promise,and must reflect lots of frustrating hours!) |
There's a new option under Sharing in the Options, at the bottom. Is it not working? You can also select all with the "Individually Shared Files" folder showing and right-click to stop sharing all of them. I guess we could probably make this a little more clear, huh? Quote:
|
Doh! Found it, it works, and is perfectly clear enough. btw, do you look after magnets too? I wondered if MagnetClipboardListener could be easily enabled for OSX and Windows, as a way of testing magnets with the CVS versions. |
Felix is fixing up a lot of magnet stuff. |
QUIT/ Library Thanks. Sam, the OS X "Quit" menu problem didn't show up today, though the conditions were similar to before. Just wondering if it was fixed, or if I should continue trying to repeat it. a few Library notes: --I assume DnDing a file from one Library folder to another isn't enabled, and that command-clicking a file among a selection still deselects all the others instead of the expected file. (just checking). --it would be good if the virtual media folders were also created in the OS. That would make the Explore button more functional. --I REALLY like the fact that individual files can be shared or unshared without losing all the hits/locations data. :D --adding a Shared/Unshared column to the Library would help check which files need to be added. A "Date Modified" (or Added) column, as in iTunes or Finder windows would also be useful. --when sharing a group of unshared files, any already shared files among the selection end up with all their hits/locations data reset. Would be nice if already shared files were ignored in this minor case. -the duplicate files handling on the Search pane is a very welcome addition. The dialog L&F is relatively cluttered on OSX, and perhaps default settings could be optimized with automatic defaults. |
LAN transfers/connections 1 Attachment(s) Any chance LAN transfers can be fixed for this setup? I realize the upload limits, which are set for uploads to external clients on both machines is the problem. -IBook G3 and an iMac G4; both running 10.3.9; both have separate static local IP's to enable uploads; both can accept incoming; both have upload bandwidth set at 90KB/s--about 85% -G3 has a static IP of 192.168.59.2 and is running as an UP with the CVS version -G4 is 192.168.59.3, running LW 4.8.1 Pro as a leaf. Search results from either machine only sometime show the other's files as blue stars, so that part works OK, but when I started LW in a different user account on the G4, the results just showed as the regular T1 speed on the G3, but the G4's search for "stief" quickly showed blue files. :confused: Searching with the G3 for the name of the G4's user account quickly returns a list of shared files. Grr :mad: Please remove user account names from the keywords advertised in the path. When the G4 downloaded a video from the G3, the screenshot taken from the G3's monitor pane shows the upload split between the router's static local IP and the proper static IP of the server. Any chance such transfers could bypass the upload controls, and be shown as a single transfer? Sorry I can't explain this more clearly, but I hope LAN transfers can be revisited before LW 5 Thanks |
bug report streamlined? --is there an automatic limit to max of three "internal error" dialogs? When LW is the foremost window, I get a bug dialog every few minutes, but overnight I've never had to get rid of more than three. --If the preference for Bug Reports is set to "Always Review", could LW just skip the initial dialog and just pop up the review window with the Discard button (and Send button, for regular release versions)? This would be easier, and would help ensure that I didn't review the same error report more than once Thanks [edit--overly large "discard button example" removed to save thread space] |
The bug report dialog was limited two three at once since its inception. The problem with popping up the 'Review' dialog immediately is that it is very ugly and filled with information that most users don't care about. It's useful for us, and for beta testers / CVS users, but that's about it. Also, 'Always Review' is the default option. |
Thanks again for the info. Makes sense. Either adding the discard button to the reviewed report, or adding the exception type to the initial dialog would be welcome. I forgot to say earlier that I like the Help->Generate Error menu item, and hope it makes it to the release version (or something similar that can be copied and posted easily in the forums). |
"Generate Error Report" is never going to be in a released version. It's there for debugging. We could, however, have something that generates the sample information from the bottom. We'll look into adding the exception type to the initial dialog as well as the discard button to the review window. Both are good ideas. |
updated CVS Fri Jun 17 16-02-26 re the generated bug report, thought you might be interested to know that under low HD conditions, it did not pop up a window (but the terminal recorded the info). LW quit just fine. --this CVS now refreshes all the Library when the refresh button is used. With only the Incomplete folder selected, that was, previously, the only Library refreshed. Actually, I miss that, but if that's the price of having the new dynamic updating of the Shared Library whenever a file is added from the Finder, that's fine by me. --I LIKE the option to choose the download folder for individual files from the Downloads Panel. Nice! --Glad to see the bounty for client side queuing: I hope LW'll also consider Pipes rather than slots? I'm happily wasting so much time downloading picture collections from single hosts :) Most only allow me to download three pics at a time, so it's too easy to gas hours on a single collection. I wonder if "pipes" might be more efficient than "slots". I.E, once I've opened a pipe to a host, the bitstream can be maintained irrespective of the number of files that make up that bitstream. My upstream bandwidth is greatly under-utilized given the kinds of files I share, but even so I've seen a few banned greedy clients who really aren't and wish I could allow their client a "download pipe" (timed?) rather than a three file slot limit. Just wishing and dreaming ;) --re the gui slowdown for closing search tabs: it looks like it is proportionate to the number of entries under that tab. Closing a browsed host tab with 8000 entries can lock up the gui for quite a while, but a tab with just a few entries closes quickly. Can the 'clearing' of entries (or whatever is going on) be deferred to the background, so the active window can respond without waiting for the clearing? --this gui lockup also happens while waiting for a browsed host's results to show and sort for the first time. i.e, a gui lock happens while, say, 8,000 files show in the browsed host tab; another lock when I filter for Images, and then another while sorting by size. Other tabs or panes are completey unresponsive until the results are displayed several minutes later. However, uploads, connections and downloads are still progressing (seen in Activity Monitor), althought the GUI shows them frozen. lastly, this latest CVS is just as stable here as the regular release. IMHO, it's past beta already, although I am hoping the spam filters make it to the beta first. Cheers. btw, I noticed acqx has released a beta that sounds like it has tapped into all the countless hours Sam and Justin have done (without mentioning them specifically, of course) http://www.newsfirex.com/blog/?cat=3 No doubt many more contributors are also finding their work quietly appearing elsewhere. Well, I know congrats aren't $, but congrats and appreciation anyway :) |
| All times are GMT -7. The time now is 05:15 AM. |
Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2025, vBulletin Solutions, Inc.
SEO by vBSEO 3.6.0 ©2011, Crawlability, Inc.
Copyright © 2020 Gnutella Forums.
All Rights Reserved.