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

Re: [PATCH 1/5] Install hfsutils in-target for NewWorld Power Macs (adapted from d-i/yaboot-installer).



> On Nov 16, 2017, at 9:52 AM, Frank Scheiner <frank.scheiner@web.de> wrote:
> 
>> run "dch" which should not alter the timestamp at the bottom. Please
>> recreate the changelog entry with just "dch" so the timestamp doesn't
>> change.
> 
> I suspected that a tool can be used for this, though I didn't know which one. But if `dch` doesn't change the timestamp, than the last addition will have the timestamp of the original one.

Which is exactly what we want. The final timestamp will be made by the person who uploads the package to Debian unstable. And that is always done by a member of the debian-installer team. Just look at previous commits and look how they touched the debian/changelog.

> 
>>> +case $ARCH in
>>> +    powerpc/powermac_newworld|ppc64/powermac_newworld)
>>> +    if ! apt-install hfsutils; then
>>> +        info "Calling 'apt-install hfsutils' failed"
>>> +        # Hm, unable to install hfsutils into /target/, what should we do?
>>> +        db_input critical grub-installer/apt_install_hfsutils_failed || [ $? -eq 30 ]
>>> +        if ! db_go; then
>>> +            db_progress STOP
>>> +            exit 10 # back up to menu
>>> +        fi
>>> +        db_get grub-installer/apt_install_hfsutils_failed
>>> +        if [ "$RET" != true ]; then
>>> +            db_progress STOP
>>> +            exit 1
>>> +        fi
>>> +    fi
>>> +    ;;
>>> +esac
>>> +
>> This should be moved to the grub-installer script at the appropriate
>> position. I would suggest placing it into a new case statement after
>> the case statement to wipe the prep partition which starts around line
>> 400. So, please put it after that one and, again, match with
>> $ARCH:$grub_package instead of just $ARCH.
> 
> Will change the match, but the whole installation part is needed earlier, before nw_format_offs() - which needs hfsutils - will get called in the match on [1] when PATCH 3/5 is applied.

Ok, I‘ll look into this again once I‘m in the office.

Reply to: