[Bug-ddrescue] GNU ddrescue - possible bug with foreign language characters & filenames

[wilsonstreet]

Hi Antonio,

possible Precaution ;

I am using Ubuntu 10.04 (OS is in English) , which seems to
not have the most up to date version of GDD Rescue (ver
1.14) .

After recovering data from a 15GB partition,
to an image file, I then loop mounted the image file to be
able to read the data from it.

I proceeded to copy the folders and files, using Nautilus.

The partition was a windows one (I forget whether it was
NTFS or VFAT, because I had done one of both). It had come
from a system that had been running in a foreign language -
not English.

When copying the files, the folder and file names that had
unrecognizable characters in the seemed to have the
characters replaced by question marks.
This caused many files to appear to be named with the same
name, and there was no option in the Nautilus Copy function
to copy a duplicate named file (maybe by appending a number
to the name) - there was only options to Skip/Skip All/ or
Replace.

Some of this problem may well rest with Nautilus, or even
Ubuntu (when reading files named in another language set to
OS that is reading them).

Some of the problem may be the way dd_rescue handles some of
the data (like filenames or odd characters) , and the way
the resulting image behaves when mounted.
I would assume that the form of data if correct, is read
correctly by dd_rescue irrespective of whether the data
seems to make sense, but since dd_rescue is such an
excellent tool, it might be prudent to check that it
performs well when the data it is extracting is in a
different language to the one of the OS that it is being
used in.

Even if dd_rescue has not been part of this problem, it
would be helpful to provide some information on how to copy
data from images taken from OSes that used a different
language.

I hope this email proves to have been unnecessary.

Yours
Al