cupx files (ie 2 zip concatenated files) are not extracted correctly

[scls19fr]

Hello,

not sure if that's a misunderstanding of how to use binwalk (I'm quite beginer in that kind of tools) but it doesn't extract cupx files correctly.

Steps to reproduce:

• Download Big_sample_file.cupx or Small_sample_file.cupx from https://tophatsoaring.atlassian.net/browse/TH-760

• Run binwalk --dd='.*' small_sample_file.cupx -M

binwalk creates 4 files

$ tree _small_sample_file.cupx.extracted/
_small_sample_file.cupx.extracted/
├── 0
├── 1ABB2
├── 1ABC8
└── 1AD2D

It should in fact simply find 2 zip files (which have been simply concatenated into one).

Reference bug https://github.com/XCSoar/XCSoar/issues/795

Cupx file format description can be found at https://downloads.naviter.com/docs/SeeYou_CUPX_file_format.pdf

Foremost extracts file correctly https://github.com/XCSoar/XCSoar/issues/795#issuecomment-1008918231 but I'd prefer using binwalk as it can easily be used as Python module.

Kind regards

[devttys0]

This is likely due to binwalk identifying each ZIP entry as a separate ZIP file. Binwalk v3 is much better at preventing this type of thing, and successfully reports and extracts only the two ZIP files in the cupx files you linked to: