bug-xorriso
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Writing to ISO file again with different Paras creates old version


From: Thomas Schmitt
Subject: Re: Writing to ISO file again with different Paras creates old version
Date: Fri, 28 Feb 2020 09:29:47 +0100

Hi,

sean loony wrote:
> 2. Now I change test1 to test2:
> -dev /somewhere/my.iso  -update_l /foo/bar  test2  /foo/bar/boo
> But I get an ISO with test1 and the files from /foo/bar/boo

I cannot reproduce this here:

  xorriso-1.4.8 \
     -abort_on FATAL -for_backup -disk_dev_ino on \
     -dev /dvdbuffer/test.iso -disk_pattern on \
     -update_l /home/thomas test1 /home/thomas/conf_bd -- \
     -commit

  xorriso-1.4.8 \
     -abort_on FATAL -for_backup -disk_dev_ino on \
     -dev /dvdbuffer/test.iso -disk_pattern on \
     -update_l /home/thomas test2 /home/thomas/conf_bd -- \
     -commit

  xorriso -indev /dvdbuffer/test.iso -find -- 2>&1 | less

reports

  '/'
  '/test1'
  '/test1/conf_bd'
  '/test1/conf_bd/ASKME_level_0'
  ... more files in /test1 ...
  '/test1/conf_bd/specific_arguments.20040202.224656'
  '/test2'
  '/test2/conf_bd'
  '/test2/conf_bd/ASKME_level_0'
  ... same files but under /test2 ...
  '/test2/conf_bd/specific_arguments.20040202.224656'

This is the expectable result of writing two sessions by two xorriso runs,
using -dev in the second run. (The first run on non-existing test.iso could
have used -outdev as well.)

If i delete /dvdbuffer/test.iso between the two write runs, i only get
/test2 from the second write run.


> I figured out if I waited a few minutes with step 2 it works.

I have no idea what could have caused this observation.
Changes in /foo/bar during the waiting time could influence what gets
copied from disk in the second run. But they cannot plausibly influence
under what directory the changed files get put in the ISO.


> Does xorriso use some old tempfiles here or whats wrong?

The only information carried from run to run is inside the ISO filesystem
which the second run loaded.
But this only explains the presence of /test1, not the absence of /test2.


I lack of proposals except the request to check again and to watch out for
both /test1 and /test2 in the ISO after the second run.


Have a nice day :)

Thomas



reply via email to

[Prev in Thread] Current Thread [Next in Thread]