MapTool 1.3 Final, patch04 (build 87)

[jfrazierjr]

I wonder if the issue with AM is due to network binding order. If his NIC being bound changed, then that NIC might not be able to reach the router and thus throw that message...Of course... if that was the case, wouldn't he be totally unable to play over the internet?(at least thats what I thought would happen...) Hmmm on a side note, would sysinfo be able to tell which NIC is highest in the binding order and if so, could that be added as it would seem to be really really important thing to know.

[aliasmask]

/// MapTool Info \\\

==== MapTool Information ====
MapTool Version: 1.3.b87
MapTool Home...: D:\Documents and Settings\Chris\.maptool
Max mem avail..: 1,353M
Max mem used...: 62M

==== Java Information ====
Java Vendor.: Sun Microsystems Inc.
Java Home...: D:\Program Files\Java\jre6
Java Version: 1.6.0_26
............: Java(TM) SE Runtime Environment (build 1.6.0_26-b03)
............: Java HotSpot(TM) Client VM (build 20.1-b02, mixed mode, sharing)


==== OS Information ====
OS Name........: Windows XP
OS Version.....: 5.1
OS Architecture: x86
Processor......: x86 Family 15 Model 6 Stepping 2, GenuineIntel
PATH...........: null
Number of Procs: 2

==== User Information ====
User Name: Chris
User Home: D:\Documents and Settings\Chris
User Dir.: E:\GAMING\maptool-1.3.b87

==== Network Interfaces ====
Display Name..: MS TCP Loopback interface
Interface Name: lo
Address...: 127.0.0.1

Display Name..: Realtek RTL8139 Family PCI Fast Ethernet NIC - Packet Scheduler Miniport
Interface Name: eth0
Address...: 10.0.0.12

Host Address...: 10.0.0.12
Default Gateway: 10.0.0.1

==== Locale Information ====
Country.: United States
Language: English
Locale..: English (United States)
Variant.:

==== Display Information ====
Number of Displays: 1
Display 1: 1680x1050(32)

==== Internet Gateway Devices ====
No IGDs Found!

/// b87 Log \\\

2011-10-10 09:38:59,718 [upnp.Discovery$1:173] ERROR - Error occured during upnp root device object creation from location http://10.0.0.1:80/Public_UPNP_gatedesc.xml
org.apache.commons.jxpath.JXPathNotFoundException: No pointer for xpath: UPNPDevice/root
at org.apache.commons.jxpath.ri.JXPathContextReferenceImpl.getPointer(JXPathContextReferenceImpl.java:468)
at org.apache.commons.jxpath.ri.JXPathContextReferenceImpl.getPointer(JXPathContextReferenceImpl.java:452)
at net.sbbi.upnp.devices.UPNPRootDevice.<init>(UPNPRootDevice.java:138)
at net.sbbi.upnp.devices.UPNPRootDevice.<init>(UPNPRootDevice.java:100)
at net.sbbi.upnp.Discovery$1.discoveredDevice(Discovery.java:170)
at net.sbbi.upnp.DiscoveryListener.listenBroadCast(DiscoveryListener.java:273)
at net.sbbi.upnp.DiscoveryListener.run(DiscoveryListener.java:192)
at java.lang.Thread.run(Unknown Source)
2011-10-10 09:39:04,609 [client.MapTool:291] ERROR - UPnP Error - No Internet Gateway Devices found.<br><br>UPnP port mapping will not be available.
2011-10-10 09:39:12,671 [upnp.Discovery$1:173] ERROR - Error occured during upnp root device object creation from location http://10.0.0.1:80/Public_UPNP_gatedesc.xml
org.apache.commons.jxpath.JXPathNotFoundException: No pointer for xpath: UPNPDevice/root
at org.apache.commons.jxpath.ri.JXPathContextReferenceImpl.getPointer(JXPathContextReferenceImpl.java:468)
at org.apache.commons.jxpath.ri.JXPathContextReferenceImpl.getPointer(JXPathContextReferenceImpl.java:452)
at net.sbbi.upnp.devices.UPNPRootDevice.<init>(UPNPRootDevice.java:138)
at net.sbbi.upnp.devices.UPNPRootDevice.<init>(UPNPRootDevice.java:100)
at net.sbbi.upnp.Discovery$1.discoveredDevice(Discovery.java:170)
at net.sbbi.upnp.DiscoveryListener.listenBroadCast(DiscoveryListener.java:273)
at net.sbbi.upnp.DiscoveryListener.run(DiscoveryListener.java:192)
at java.lang.Thread.run(Unknown Source)

/// b86 Log \\\

2011-10-10 09:41:55,593 [util.UPnPUtil:47] INFO - UPnP: Found device: RangeMax(TM) Wireless Router
2011-10-10 09:41:55,671 [util.UPnPUtil:89] INFO - UPnP: Port 6800 mapped
2011-10-10 09:42:01,578 [util.UPnPUtil:114] INFO - UPnP: Port unmapped

/// Public_UPNP_gatedesc \\\

<?xml version="1.0"?>
<root xmlns="urn:schemas-upnp-org:device-1-0">
<specVersion>
<major>1</major>
<minor>0</minor>
</specVersion>
<URLBase>http://10.0.0.1:80/</URLBase>
<device>
<deviceType>urn:schemas-upnp-org:device:InternetGatewayDevice:1</deviceType>
<presentationURL>/start.htm</presentationURL>
<friendlyName>Residential Gateway</friendlyName>
<manufacturer>NETGEAR, Inc.</manufacturer>
<manufacturerURL>http://www.NETGEAR.com</manufacturerURL>
<modelDescription>RangeMax(TM) Wireless Router</modelDescription>
<modelNumber>WPN824v2</modelNumber>
<modelName>RangeMax(TM) Wireless Router</modelName>
<UDN>uuid:c69a5d6a-b30a-b563-3ea5-6622ad886868</UDN>
<serviceList>
<service>
<serviceType>urn:schemas-upnp-org:service:Layer3Forwarding:1</serviceType>
<serviceId>urn:upnp-org:serviceId:L3Forwarding1</serviceId>
<SCPDURL>/Public_UPNP_Layer3F.xml</SCPDURL>
<controlURL>/Public_UPNP_C1</controlURL>
<eventSubURL>/Public_UPNP_Event_1</eventSubURL>
</service>
</serviceList>
<deviceList>
<device>
<deviceType>urn:schemas-upnp-org:device:WANDevice:1</deviceType>
<friendlyName>WAN Device</friendlyName>
<manufacturer>NETGEAR, Inc.</manufacturer>
<manufacturerURL>http://www.NETGEAR.com</manufacturerURL>
<modelDescription>RangeMax(TM) Wireless Router</modelDescription>
<modelNumber>WPN824v2</modelNumber>
<modelName>RangeMax(TM) Wireless Router</modelName>
<UDN>uuid:63ba3211-1c05-a0e1-3732-85c65c74677d</UDN>
<serviceList>
<service>
<serviceType>urn:schemas-upnp-org:service:WANCommonInterfaceConfig:1</serviceType>
<serviceId>urn:upnp-org:serviceId:WANCommonIFC1</serviceId>
<SCPDURL>/Public_UPNP_WAND.xml</SCPDURL>
<controlURL>/Public_UPNP_C2</controlURL>
<eventSubURL>/Public_UPNP_Event_2</eventSubURL>
</service>
</serviceList>
<deviceList>
<device>
<deviceType>urn:schemas-upnp-org:device:WANConnectionDevice:1</deviceType>
<friendlyName>WAN Connection Device</friendlyName>
<manufacturer>NETGEAR, Inc.</manufacturer>
<manufacturerURL>http://www.NETGEAR.com</manufacturerURL>
<modelDescription>RangeMax(TM) Wireless Router</modelDescription>
<modelNumber>WPN824v2</modelNumber>
<modelName>RangeMax(TM) Wireless Router</modelName>
<UDN>uuid:ca22b020-d07d-b2ec-34a1-85107400e4be</UDN>
<serviceList>
<service>
<serviceType>urn:schemas-upnp-org:service:WANIPConnection:1</serviceType>
<serviceId>urn:upnp-org:serviceId:WANIPConn1</serviceId>
<SCPDURL>/Public_UPNP_WANIPConn.xml</SCPDURL>
<controlURL>/Public_UPNP_C3</controlURL>
<eventSubURL>/Public_UPNP_Event_3</eventSubURL>
</service>
</serviceList>
</device>
</deviceList>
</device>
</deviceList>
</device>
</root>

[Azhrei]

Very interesting. The error is in the Apache Commons JXPath library, which was updated for b87. It's apparently complaining that it can't find a root entry in the XML document returned by the router, yet the content you posted for the return data clearly shows that there is a <root> element.

I'll look at any outstanding bug reports for JXPath and then trace the code to see what the net.sbbi.upnp library is trying to do -- perhaps the API for the JXPath library has changed and the library needs to be updated to match.

Thanks!

[jfrazierjr]

Very interesting. The error is in the Apache Commons JXPath library, which was updated for b87. It's apparently complaining that it can't find a root entry in the XML document returned by the router, yet the content you posted for the return data clearly shows that there is a <root> element.

I'll look at any outstanding bug reports for JXPath and then trace the code to see what the net.sbbi.upnp library is trying to do -- perhaps the API for the JXPath library has changed and the library needs to be updated to match.

Thanks!

I think you might need to specify the namespace. xpath needs to know the namespace for navigation purposes. If the old version was not NS complicant... it probably just worked, but was "broken". Now it's fixed to know about ns's, but that broke the existing code.

[dorpond]

Good catch, Aliasmask!

[Azhrei]

Previously the OK button was defaulted to be selected so when I hit enter on the keyboard, OK was depressed. Now, I have to tab through a whole input tab to select OK to hit enter.

Hm. I did see a note in the TinyLAF Release Notes that the handling of the default button had changed slightly. Let me take a look -- I think it's a checkbox in the TinyLAF control panel now... If so then it can probably be changed by the program as well.

The code that handles the input function is pretty intense. What I mean is, there's a lot of "this is a bug in Swing so I did this to work around it" type of code.

I'm not sure I've fixed the OK button having the default focus, but when I use one of the many input-based macros in the D&D3.5/PF framework I can just hit Enter to close the panel.

I've also set the maximum size of the dialog to be 75% of the screen height. There was already code that did that for the content of the dialog, but that didn't take into account the height of the tabs. For some reason the dialog won't resize horizontally properly so I'm leaving that alone. (Hm, it just occurred to me that a dialog opened on a secondary window might not be properly height-constrained. I'll check that.)

I did set it up so that the scrolled area will always be at the top when an input opens. As I said, I tested it with the PF framework using the ActiveMods and Edit Equip > Add New... macros.

These fixes are checked in as SVN 5808.

Very interesting. The error is in the Apache Commons JXPath library, which was updated for b87. It's apparently complaining that it can't find a root entry in the XML document returned by the router, yet the content you posted for the return data clearly shows that there is a <root> element.

I'll look at any outstanding bug reports for JXPath and then trace the code to see what the net.sbbi.upnp library is trying to do -- perhaps the API for the JXPath library has changed and the library needs to be updated to match.

Thanks!

I think you might need to specify the namespace. xpath needs to know the namespace for navigation purposes. If the old version was not NS complicant... it probably just worked, but was "broken". Now it's fixed to know about ns's, but that broke the existing code.

Yep, that was exactly my thinking. I'm going to work on this next. We'll see if mfrizzell can replicate his wrong-settings-for-FOW-and-vision bug...

[wolph42]

I was wondering, did anyone notice my earlier benchmark post?
I thought it at least peculiar that the new libs in this build should increase the speed for json object operations, while the benchmark showed that (at least for creating json objects) its actually 20% slower, which is rather significant and contradictory.

[Azhrei]

I said that because the json-lib library was updated and because of the Release Notes for that library that I expected JSON operations to be faster. That's not to say that converting strings to JSONs or JSONs to strings would improve, but that operations like Wiki: json.diff() or Wiki: json.merge() might be faster. Of course, MT has it's own overhead for these operations so any potential speedup could get lost in the MT overhead.

[Tyshalle]

This build still hasn't fixed the really weird issue that I think started in b86, where when you draw something and then undo it, and then draw something else, it immediately reverts back to the image you just deleted. This is fixed simply by zooming in and out a bit, but it's still pretty annoying to have to constantly zoom in and out every time you want to draw a new thing, especially if you're making a very elaborate map with a lot of details. I find that the only real workaround to maintain my sanity is to jump back into b84 and draw my maps there, and then export them to the newer builds. That's fine for when I'm drawing new maps ahead of time, but if I need to make any edits in-game, or god forbid if my players need to do any drawing, this can all get incredibly irritating and confusing for everyone.

This same issue also seems to affect all the images on a map when it first loads. Almost everything is blank, except objects you've drawn when you first boot up a campaign file, until you zoom in and out a bit. This is less of a big deal and more of a nitpick, since it won't actually affect anyone beyond the first few seconds they load up a campaign, but I do have a player who plays on a laptop without a mouse, and so he doesn't zoom in or out very often, and this might cause occasional confusion for him. It's still not that big of a deal though, compared to the drawing bug.

Also the token movement is kind of a nitpick, but since it's a nitpick that's likely going to come up all the time, I can see it becoming pretty irritating for some people. Thankfully not me, as I almost never use the grid!


A couple other nitpicks:

I'm not sure I love the font change. It's decent, but it doesn't do well with bold text in certain situations. Also, it takes up a marginal amount of extra vertical real estate, and so for frameworks with a lot of buttons on them, it sort of unnecessarily throws everything off IMO. It's not a huge deal, but I definitely think that I prefer the old font.

The chat box still spazzes out when you put more than two lines of text into it. It's mostly okay if you are just typing away, but if you click off the screen at all, it suddenly goes crazy, resizing again and again as you move your cursor around the map. It also prevents you from scrolling if you do this, so if you type 6 lines of text and click away, you can't easily get to the bottom of your chat box again. Instead you'll have to hold down the right arrow until you get to the bottom, since you can't use the up or down arrows for some reason.


I'll probably give this a whirl in my game soon, and report more info then.

[Azhrei]

This build still hasn't fixed the really weird issue that I think started in b86, where when you draw something and then undo it, and then draw something else, it immediately reverts back to the image you just deleted.

This is covered in a separate thread: this is integral to the way Undo/Redo work and MapTool's implementation is terrible -- it won't be fixed in 1.3.

This same issue also seems to affect all the images on a map when it first loads.

This is a different issue. I have a patch for it, but it doesn't seem to 100% solve it. I plan to have this patch in b88.

I'm not sure I love the font change.

It's the default for the new TinyLAF library. If you don't like it you can create your own "theme" file and drop it in your home directory. Now it'll use whatever font you want.

There will be a note in the README of the next build zip that describes where to find the new TinyLAF control panel software for making such a change.

The chat box still spazzes out when you put more than two lines of text into it.

Hm, I've never heard of this. But I leave my chat window docked so maybe that's the difference?

It also prevents you from scrolling if you do this, so if you type 6 lines of text and click away, you can't easily get to the bottom of your chat box again. Instead you'll have to hold down the right arrow until you get to the bottom, since you can't use the up or down arrows for some reason.

Or use the left mouse button and select all of the text. Left dragging on the chat text will autoscroll and when you get to the end press the right arrow once to remove the selection. (The up and down arrows move one *logical* line at a time, not one *physical* line.)

In any case, this is part of the built-in JEditPane functionality and won't be changing in 1.3.

[aliasmask]

One last request to fix this for final. When using [frame:], the height, width and temporary string props are ignored. I also get an error if I use a title of an existing frame:

Could not execute the command: "s [frame(d1000,"temporary=1;title=Edit: image:Unknown"): {Test}]", exception = The DockableFrame exists with key "Edit: image:Unknown". Please specify a new key for the frame.

This is the case whether the frame is visible, open or not. Using closeFrame or resetFrame doesn't help. If it has been opened once before, specifying title will blow it up if title has existed before. Shouldn't the key be the var name rather than the title?

resetFrame puts the frame in center of screen, which I like. Can opening the frame for first time do the same thing?

[neofax]

Can someone confirm that the filled circle for VBL doesn't show the area while drawing the VBL? I was making maps last night and tried to use it, but couldn't see where the final VBL was going to be.

[CoveredInFish]

It seems to work fine here.

[Azhrei]

One last request to fix this for final. When using [frame:], the height, width and temporary string props are ignored. I also get an error if I use a title of an existing frame:
[...]
resetFrame puts the frame in center of screen, which I like. Can opening the frame for first time do the same thing?

Hm, I've added this to my hitlist. I'll check into it before putting out b88.

Can someone confirm that the filled circle for VBL doesn't show the area while drawing the VBL? I was making maps last night and tried to use it, but couldn't see where the final VBL was going to be.

Works fine here, too. I tried both filled and unfilled circles, both drawing and erasing. When drawing I get a red "rubberband" showing where the VBL will be placed and when erasing it's white instead.

[neofax]

Strange. Doesn't for me and that is the only one that does that. No problem as I can just use the freehand fill tool.