NPE at DxfImportTask.realRun

[jjaf.de]

Trying to load a .DXF via DxfImport from ​Open GeoData of Hanover, e.g. ​SKH 1000 Teil Nord 4597D.DXF. After conforming the scale 1 unit(s) = 1 m:

An unexpected exception occured that may have come from the 'DxfImport' plugin.

URL:http://josm.openstreetmap.de/svn/trunk
Repository:UUID: 0c6e7542-c601-0410-84e7-c038aed88b3b
Last:Changed Date: 2017-10-29 19:52:48 +0100 (Sun, 29 Oct 2017)
Build-Date:2017-10-29 19:11:19
Revision:13053
Relative:URL: ^/trunk

Identification: JOSM/1.5 (13053 en) Linux Debian GNU/Linux 9.1 (stretch)
Memory Usage: 610 MB / 1820 MB (381 MB allocated, but free)
Java version: 1.8.0_151-8u151-b12-1~deb9u1-b12, Oracle Corporation, OpenJDK 64-Bit Server VM
Screen: :0.0 1680x1050, :0.1 904x1440
Maximum Screen Size: 1680x1440
Java package: openjdk-8-jre:amd64-8u151-b12-1~deb9u1
Java ATK Wrapper package: libatk-wrapper-java:all-0.33.3-13
Dataset consistency test: No problems found

Plugins:
+ DxfImport (1010)
+ ImportImagePlugin (33563)
+ PicLayer (33385)
+ ejml (32680)
+ geotools (33380)
+ jts (32699)
+ log4j (32699)
+ measurement (33088)

Last errors/warnings:
- W: No configuration settings found.  Using hardcoded default values for all pools.
- E: java.io.IOException: java.lang.NullPointerException. Cause: java.lang.NullPointerException
- E: Handled by bug report queue: java.io.IOException: java.lang.NullPointerException. Cause: java.lang.NullPointerException


=== REPORTED CRASH DATA ===
BugReportExceptionHandler#handleException:
No data collected.

Warning issued by: BugReportExceptionHandler#handleException

=== STACK TRACE ===
Thread: AWT-EventQueue-1 (17) of main
java.io.IOException: java.lang.NullPointerException
	at org.openstreetmap.josm.plugins.dxfimport.DxfImportTask.realRun(DxfImportTask.java:193)
	at org.openstreetmap.josm.gui.PleaseWaitRunnable.doRealRun(PleaseWaitRunnable.java:95)
	at org.openstreetmap.josm.gui.PleaseWaitRunnable.run(PleaseWaitRunnable.java:143)
	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
	at java.util.concurrent.FutureTask.run(FutureTask.java:266)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:748)
Caused by: java.lang.NullPointerException
	at org.openstreetmap.josm.plugins.dxfimport.DxfImportTask.realRun(DxfImportTask.java:182)
	... 7 more

[Don-vip]

The zip archive does not contain any .DXF file?

[jjaf.de]

Replying to Don-vip:

The zip archive does not contain any .DXF file?

Sorry, wrong URL. The Correct one is ​SKH 1000 DXF.

[Don-vip]

In console we can see this error:

[Fatal Error] :284:174: Attribute "stroke-width" was already specified for element "use".

[Don-vip]

Indeed the parser chokes on:

   <g id="Gelände-_#32_u._#32_Gewässertopographie" color="rgb(0,0,0)" stroke="currentColor" fill="none" stroke-width="0.10202025000004505">
    <line x1="546000" y1="5797400.499" x2="545977.414" y2="5797500" id="ID_126" stroke-width="0.18mm" stroke-dasharray="10, 0.05% "/>
    <line x1="545978.835" y1="5797500" x2="546000" y2="5797407.376" id="ID_127" stroke-width="0.18mm" stroke-dasharray="10, 0.05% "/>
    <use transform="translate(0 0)" id="ID_95" stroke-width="0.18mm" stroke-width="0.10202025000004505" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="#ID__#42_X0"/>
    <use transform="translate(0 0)" id="ID_124" stroke-width="0.18mm" stroke-width="0.10202025000004505" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="#ID__#42_X1"/>
   </g>

[jjaf.de]

So the Plugin converts the DXF to an interim SVG representation and this one includes superfluous stroke-width attributes in use elements?

It seems like the "0.18mm" may be derived from DXF values "18", but the "0.10202025000004505" seems to be a Plugin calculated value.

[Don-vip]

Yes, but not directly. The plugin uses the ​Kabeja library to transform to SVG, then the SVG is converted thanks to our embedded SVG engine: ​svgSalamander.

Problem is: Kabeja is dead (last update in 2010), and several people tried to improve it differently, on their own (myself included):

• ​fuzziness: Nov 2010
  • ​skenny: Jun/Sep 2014
    • ​wfr: Oct 2015
      • ​Ticketfly: April/May 2016
        • ​accu-rate: Jan/Oct 2017

• ​Antea: Nov 2010 -> Dec 2016
  • ​myself: Sep 2016

I have not enough time right now to dig into Kabeja code. If you find what's wrong, I could release a forked version which fixes this problem. Not sure however where things are going between Antea's line of development and fuziness' one...

[jjaf.de]

Replying to Don-vip:

Yes, but not directly. The plugin uses the ​Kabeja library to transform to SVG, then the SVG is converted thanks to our embedded SVG engine: ​svgSalamander.

So i believe one can improve robustness by following ​Postel's law by making kabeja output unambiguous and make svgSalamanders Fatal Error a Warning only.

Problem is: Kabeja is dead (last update in 2010), and several people tried to improve it differently, on their own (myself included):

• ​fuzziness: Nov 2010
  • ​skenny: Jun/Sep 2014
    • ​wfr: Oct 2015
      • ​Ticketfly: April/May 2016
        • ​accu-rate: Jan/Oct 2017

• ​Antea: Nov 2010 -> Dec 2016
  • ​myself: Sep 2016

I have not enough time right now to dig into Kabeja code. If you find what's wrong, I could release a forked version which fixes this problem. Not sure however where things are going between Antea's line of development and fuziness' one...

From ​Dxf-Import source i believe ​kabeja-0.4 is the used ​source.

By reading the context of grep --recursive --ignore-case 'stroke-width ./kabeja-0.4/ i believe kabeja-0.4/blocks/svg/src/org/kabeja/svg/generators/SVGViewportGenerator.java (svgContext.put) could be the problem:

public void toSAX(ContentHandler handler, Map svgContext, DXFEntity entity,
        TransformContext transformContext) throws SAXException {
[...]
                SVGUtils.addAttribute(attr,
                    SVGConstants.SVG_ATTRIBUTE_STROKE_WITDH,
                    SVGUtils.formatNumberAttribute(width));
                svgContext.put(SVGContext.STROKE_WIDTH, new Double(width));

I would delete the line 175 with svgContext.put as it seems the value width is pushed in two different ways to probably the same context. This may be a remnant of an API change since it is the only occurrence in the whole project here and the code around it uses SVGUtils.addAttribute only.

I did not execute any code to verify!

But would't that unconditional double stroke-width mean that DxfImport could not import anything!? Am i experiencing a bug in a rarely used Plugin again - like the last one we had contact ;-) ?

BTW: ATTRIBUTE_STROKE_WITDH is a typo with 22 occurrences, but maybe this is fixed in the forks already.

[Don-vip]

Replying to jjaf.de:

By reading the context of grep --recursive --ignore-case 'stroke-width ./kabeja-0.4/ i believe kabeja-0.4/blocks/svg/src/org/kabeja/svg/generators/SVGViewportGenerator.java (svgContext.put) could be the problem:

public void toSAX(ContentHandler handler, Map svgContext, DXFEntity entity,
        TransformContext transformContext) throws SAXException {
[...]
                SVGUtils.addAttribute(attr,
                    SVGConstants.SVG_ATTRIBUTE_STROKE_WITDH,
                    SVGUtils.formatNumberAttribute(width));
                svgContext.put(SVGContext.STROKE_WIDTH, new Double(width));

I would delete the line 175 with svgContext.put as it seems the value width is pushed in two different ways to probably the same context. This may be a remnant of an API change since it is the only occurrence in the whole project here and the code around it uses SVGUtils.addAttribute only.

I did not execute any code to verify!

But would't that unconditional double stroke-width mean that DxfImport could not import anything!? Am i experiencing a bug in a rarely used Plugin again - like the last one we had contact ;-) ?

BTW: ATTRIBUTE_STROKE_WITDH is a typo with 22 occurrences, but maybe this is fixed in the forks already.

Thanks for searching, I'm giving it a try :)

[Don-vip]

It was painful, but I have an improvement with ​https://github.com/JOSM/Dxf-Import/releases/tag/v1012. Can you please test?

[Don-vip]

Ticket #14988 has been marked as a duplicate of this ticket.