Does the automatic setInstallLocation even work?

[Se7enLC]

Can somebody using BB 0.2 run this test and tell me what it says?

pm getInstallLocation

If that doesn't work, run this:

cd /system/bin
sh pm getInstallLocation

Please let me know if you had to use the second method and what the output is.

[actngrezy]

Can somebody using BB 0.2 run this test and tell me what it says?

pm getInstallLocation

If that doesn't work, run this:

cd /system/bin
sh pm getInstallLocation

Please let me know if you had to use the second method and what the output is.

Give me n gr if none does it by time I'm home I will

[temchik]

mine returns 2. by the way, even if I set it manually to 1 a reboot reverts it back to 2...

[Se7enLC]

mine returns 2. by the way, even if I set it manually to 1 a reboot reverts it back to 2...

Huh - must be something weird about my installation.

I'm looking directly at the installation file for BB 0.2 (FRF57V02.zip) at:
/system/bin/pm
/system/etc/install-recovery.sh

And I see NO WAY that pm should be able to execute from the commandline OR from /system/bin if those two files properly install. pm, even if marked executable, cannot execute unless you add #!/system/bin/sh as the first line.

Can somebody post the contents of their /system/bin/pm file and verify that it runs when they type "pm"? Can you also type "busybox which pm" to see that it is, in fact, executing the /system/bin/pm script?

[temchik]

here's mine:

# busybox which pm
busybox which pm
/system/bin/pm
# ls -l /system/bin/pm
ls -l /system/bin/pm
-rwxr-xr-x root     shell         191 2008-08-01 08:00 pm
# cat /system/bin/pm
cat /system/bin/pm
# Script to start "pm" on the device, which has a very rudimentary
# shell.
#
base=/system
export CLASSPATH=$base/framework/pm.jar
exec app_process $base/bin com.android.commands.pm.Pm "$@"

# pm getInstallLocation
pm getInstallLocation
2[external]

[Se7enLC]

Huh, that's exactly what I have.

Looks like it's a bug (intentional feature?) in the default android shell that makes it just assume that any executable text file is a shell script without needing a shebang line! I switched back to the default shell, and it worked fine again.

To anybody else having this problem - you need to modify /system/bin/pm to include that header line, and you're all set.

http://en.wikipedia.org/wiki/Shebang_%28Unix%29

[michael_campbell]

Huh, that's exactly what I have.

Looks like it's a bug (intentional feature?) in the default android shell that makes it just assume that any executable text file is a shell script without needing a shebang line! I switched back to the default shell, and it worked fine again.

To anybody else having this problem - you need to modify /system/bin/pm to include that header line, and you're all set.

Shebang (Unix) - Wikipedia, the free encyclopedia

This is fairly standard unix behavior.

bash-3.2$ ls -la foo
ls: cannot access foo: No such file or directory
bash-3.2$ echo "echo I am foo\!" > foo
bash-3.2$ chmod +x foo
bash-3.2$ ./foo
I am foo!
bash-3.2$

[Se7enLC]

This is fairly standard unix behavior.

bash-3.2$ ls -la foo
ls: cannot access foo: No such file or directory
bash-3.2$ echo "echo I am foo\!" > foo
bash-3.2$ chmod +x foo
bash-3.2$ ./foo
I am foo!
bash-3.2$

Well I stand corrected! I guess the bug lies with the busybox ash shell, then, for not exhibiting that behavior. I've never actually tried to make a shell script without a shebang line. Learn something new every day...

[michael_campbell]

Well I stand corrected! I guess the bug lies with the busybox ash shell, then, for not exhibiting that behavior. I've never actually tried to make a shell script without a shebang line. Learn something new every day...

The only reason I knew was that I often forget the shebang on perl/ruby/python scripts, and then gasp in awe at the error messages that (ba)sh emits when trying to run the damn things!