The Ftape Tools Manual - Installation
There are some DejaGnu
tests available for some programs. Some of
them require a spare tape cartridge and a tape drive installed,
all data on this cartridge will be lost! Those tests are
disabled by default, you need to set a special configuration
option to enable them. See Running tests after compilation of the programs.
All the tests can be executed by running make
with the
check
target, i.e. by running
make check
from either the top level directory of the distribution or from one of its subdirectories.
If any one of the tests fails, then please first check whether your tape drive and the ftape driver is installed correctly. If this is the case, contact Claus-Justus Heine.
The easiest way to send me the test results is the following. Change to the directory of the program with the failing test cases. Then do a
make RUNTESTFLAGS="--mail heine@math1.rwth-aachen.de" check
which will email me a summary of the test run. Of course, for this to work you need an active Internet connection, or an otherwise properly configured email delivery system.
Currently, only three of the programs come with DejaGnu
test-suites, and there is a tiny testsuite which checks for general
ftape features respectively bugs.
swapout
|
Non-destructive tests, simply check whether the options parsing works. See Swapout. |
vtblc
|
Destructive tests. Require a spare cartridge and an installed floppy
tape drive. Checks have to be enabled using the
|
ftmt
|
Destructive tests. Require a spare cartridge and an installed floppy
tape drive. Checks have to be enabled using the
|
signal handling
|
This is a non-destructive test, but nevertheless it has to be enabled
using the The test requires to insert a tape cartridge into your tape drive, but it does nothing but to retension that cartridge. In case the test fails, please first try to run the test program
manually. The
|
End Of Data Handling
|
This is a non-destructive test, but nevertheless it has to be enabled
using the The following example shows how the ftape driver should behave when trying to read past the area which contains already recorded data:
In other words, the expected behaviour is to return two times a zero
byte count and |
Writing tests for the Tcl/Tk front-end ftape-tool
would probably a
little bit hard, testing the ftformat
program would consume much
time. Formatting an average Travan-3
cartridge takes about 8
hours.
A DejaGnu
test for the listtape.pl
program could easily be
written, but this hasn't been done yet.
![]() |
Use these buttons to jump to the top menu |