BUG: PQ creates invalid GPX/XML

[+RCH65]

The PQ generator creates an invalid XML if there are Special characters (below X'20' ?) somewhere in the cache description. I don't know how they get into the listing, probably the owner used Copy&Paste. Example: GC4N77D

 

This results in the following invalid data:

 

<name>GC4N77D</name>

<desc>déjà vu's „das war der Anfang“ #1 by dejavue7, Traditional Cache (1.5/3)</desc>

<url>http://www.geocaching.com/seek/cache_details.aspx?guid=0e532090-104c-4e07-a501-c05ad8538a07</url>

<urlname>déjà vu's „das war der Anfang“ #1</urlname>

..snip...

 

Many (Garmin) GPSr stop importing caches at this point with the result, that you have just a partial Import.

Even a full-blown XML Reader on a Windows system refuses to import this file.

 

PS: After removing the red marked data, everything works as expected...

Edited January 27, 2014 by RCH65

[+ecanderson]

Granted, characters below 20h probably have no business in a cache listing, but the bug belongs to Garmin. Lack of robust parsing code is the real problem.

[team tisri]

Granted, characters below 20h probably have no business in a cache listing, but the bug belongs to Garmin. Lack of robust parsing code is the real problem.

 

If a full blown XML parser has the same trouble it sounds like it's a wider issue than just Garmin units.

[team tisri]

The PQ generator creates an invalid XML if there are Special characters (below X'20' ?) somewhere in the cache description. I don't know how they get into the listing, probably the owner used Copy&Paste. Example: GC4N77D

 

This results in the following invalid data:

 

<name>GC4N77D</name>

<desc>déjà vu's „das war der Anfang“ #1 by dejavue7, Traditional Cache (1.5/3)</desc>

<url>http://www.geocaching.com/seek/cache_details.aspx?guid=0e532090-104c-4e07-a501-c05ad8538a07</url>

<urlname>déjà vu's „das war der Anfang“ #1</urlname>

..snip...

 

Many (Garmin) GPSr stop importing caches at this point with the result, that you have just a partial Import.

Even a full-blown XML Reader on a Windows system refuses to import this file.

 

PS: After removing the red marked data, everything works as expected...

 

Thanks for the info in red, I'd seen the error but never been sufficiently interested to dig into it any deeper. Maybe a simple code parser could try and open the file as an XML document and, if it failed, open it as a text file and do an automated removal of anything from through followed by opening the remaining text as an XML document.