Bionic32-light - 27 April, 2019

Message

nic007 · #91 Post by **nic007** » Mon 06 May 2019, 12:19

Smithy wrote:OK, I have done the developers job for them just once.
A convoluted affair, that seemed to get there eventually..
Precise Light Update PET: 121k as opposed to 8mb (don't know why,xz maybe)

https://www.dropbox.com/s/aixnaymvlpa2d ... o.pet?dl=1

But isn't this one a bit of a community project. I don't like this attitude of "I have done the developers job for them just once" of yours.

jrb · #92 Post by **jrb** » Mon 06 May 2019, 12:49

peebee wrote:
Smithy wrote:So we end up with a pile of+1, +2, +3 ISOs, taking up gigs of data and just languishing there.
May I respectfully suggest that developers/ builders offer a PET in addition to a DELTA update of OS.
OR make a DELTA that will update the Puppy SFS.

DELTA was sort of OK in the Micko days of experimentation with thin Slacko, but soon becomes tiresome in the catchup game.
Unless you're an avid collector of redundant isos - just delete any intermediate isos - keep just the base and the most recent - simples!

There is a utility to make a pet (mainly intended for full installs)...........
http://www.murga-linux.com/puppy/viewtopic.php?t=115088

A Puppy SFS only delta would be about the same size.......but would cause more problems for the "full install brigade"............

Thanks for explaining this Peebee. If I might point out, in the extra time you spend waiting for an ISO to download you could download the .delta and build the new ISO. Its quick, easy and completely painless.

jrb wrote:Put the .delta file in the same directory as upupbb32-light-19.03-uefi+0.iso. Click once on the .delta file and then click "Generate" and its a done deal. You will have the new ISO. 175Mb

This reminds me of the early days in Puppyland when people kept insisting that you needed a Full install in linux, or that running as Root was the end of the world.

Smithy wrote:OK, I have done the developers job for them just once.
A convoluted affair, that seemed to get there eventually..
Precise Light Update PET: 121k as opposed to 8mb (don't know why,xz maybe)

https://www.dropbox.com/s/aixnaymvlpa2d ... o.pet?dl=1

This is Puppy, we are all developers

I had better point out before anyone uses this .pet that this is the Bionic32-light thread, NOT!!! the Precise-light thread .Please only intstall this .pet in full installs or CD installs of Precise-light.

Smithy · #93 Post by **Smithy** » Mon 06 May 2019, 16:37

Yes, not even got round to trying your Bionic Light yet JRB

And don't forget the thousands who probably never post and run Puppy
from a usb stick in a "frugal" as it is called manner. Delta is a no no, Pet is a Yes.

musher0 · #94 Post by **musher0** » Mon 06 May 2019, 18:40

Smithy?

You're such a great guy when you're not grouchy!

Take it easy, man.

I used to hate deltas, I even was afraid of them. But a delta is not an ugly centipede,
not a rat, not a cockroach: it's just a computer file.

Take a little time to study how to use them, and you'll find them practical.

Don't worry, nobody's asking you to love them or even like them!

BFN.

musher0 · #95 Post by **musher0** » Tue 07 May 2019, 07:40

Hello all.

I very much like the approach of the cdrskin people to disc burning. IMO, it's much
more intuitive than that of the cdrecord utilities, even if cdrskin is a CLI app.

To burn a Puppy or DOG iso, the command is simply:

Code: Select all

cdrskin -v dev=/dev/sr0 speed=4 fs=8m driveropts=burnfree padsize=300k -eject tazpuppy-5.0-beta-28.iso

or

Code: Select all

cdrskin -v dev=/dev/sr0 blank=as_needed -eject DevuanDog-2019-03-29_32-bit.iso

.
Plus their README doc provides good examples of most situations, so you're not
figuring out the action you wish to do, from scratch.

So... voilà ! Compiled for and on this upupbb32-light. The number at the end of
the file name is the ldd version for this pup. No guarantees, but I suspect this
compilation of cdrskin will work on any Pup with an ldd version equal to or higher
than 2.27. cdrskin-1.50 is the latest version as of this writing.

To reassemble a split archive, you know the drill, don't you?

Code: Select all

cat xa*.pet > cdrskin-150glb227.pet

and then you click as usual on the restored pet archive to install it.

It's a bit like the delta thing, Smithy!

BFN.

jrb · #96 Post by **jrb** » Thu 09 May 2019, 01:51

I have just posted Portable Browser Installer which will download the latest Firefox, Palemoon and/or Seamonkey, install them to a drive of your choice, remember that drive even if the name changes (usb order of insertion), place an entry on the Internet menu and ask if you want to make it your default browser.

Running this way will save use of ram as the browser files are not uploaded to ram like Puppy.

Works very well in Bionic32-light for me. I may issue an update soon with this replacing the Firefox-portable default browser.

Cheers, J

musher0 · #97 Post by **musher0** » Thu 09 May 2019, 10:42

Hello all.

If any of you experience the same non-US keyboard layout problem I had, I believe
I have figured it out,after much research and try-out of a number of keyboard-
related utilities.

I cannot say with absolute certainty that urxvt is the first upstream cause
-- maybe it is the way it was compiled?
-- Is it a glitch in version 9.22?
-- Is it the udev rules?
-- Is it some other language interface? ( E.g. i18n; MLterm comes with its own
i18n support. )
-- Is it the absence, in Puppy, of the traditional "kbd utils" found in other distros?
Whatever the ultimate cause may be, I now know that giving my CLI editors MLterm
support solved the problem.

The "ML" in MLterm stand for "multi-lingual". It was designed initially for Asian
languages, but if it can do that, it can accommodate "simpler" locales such as
fr_CA.UTF-8 and the ca keyboard layout.

MLterm has been around for a good while, but it is still maintained. As a matter of
fact, the devs have published a new version earlier this year.

MLterm lets me use the same ca keyboard layout in CLI as in GTK ( geany, for
ex., or any browser ), which urxvt seemed incapable of doing in the upupBB32's,
both jrb's and peebee's.

I will continue monitoring MLterm on my upupbb32-light for a day or two. If I get no
bugs, I'll make available a pet archive of the latest version, so other users of a
"latin" keyboard may benefit from this find.

BFN.

musher0 · #98 Post by **musher0** » Sun 12 May 2019, 18:34

Hello all.

Please find attached a corrective for pmount mis-labeled labels.

I didn't edit anything. I just plucked the pmount and the required
function from jrb's Precise Light, and it worked.

Screenshots of correct pmount attached.

IHTH

futwerk · #99 Post by **futwerk** » Thu 16 May 2019, 01:50

new backgrounds

jrb · #100 Post by **jrb** » Thu 16 May 2019, 01:55

futwerk wrote:new backgrounds

Wow, more beautiful artwork. Thanks again futwerk.

Is it just me or is bionic6.jpg a bit erotic?

musher0 · #101 Post by **musher0** » Thu 16 May 2019, 23:49

musher0 wrote:Hello all.

If any of you experience the same non-US keyboard layout problem I had, I believe
I have figured it out,after much research and try-out of a number of keyboard-
related utilities.

I cannot say with absolute certainty that urxvt is the first upstream cause
-- maybe it is the way it was compiled?
-- Is it a glitch in version 9.22?
-- Is it the udev rules?
-- Is it some other language interface? ( E.g. i18n; MLterm comes with its own
i18n support. )
-- Is it the absence, in Puppy, of the traditional "kbd utils" found in other distros?
Whatever the ultimate cause may be, I now know that giving my CLI editors MLterm
support solved the problem.

The "ML" in MLterm stand for "multi-lingual". It was designed initially for Asian
languages, but if it can do that, it can accommodate "simpler" locales such as
fr_CA.UTF-8 and the ca keyboard layout.

MLterm has been around for a good while, but it is still maintained. As a matter of
fact, the devs have published a new version earlier this year.

MLterm lets me use the same ca keyboard layout in CLI as in GTK ( geany, for
ex., or any browser ), which urxvt seemed incapable of doing in the upupBB32's,
both jrb's and peebee's.

I will continue monitoring MLterm on my upupbb32-light for a day or two. If I get no
bugs, I'll make available a pet archive of the latest version, so other users of a
"latin" keyboard may benefit from this find.

BFN.

Hello all.

Tests conclusive. The only thing:
-- I'd recommend avoiding the transparency settings in mlterm.
---- Once, doing a long mksquasfs session in an mlterm console with transparency
on, the characters became blurry to the point they were unreadable.

Aside from that, it's a great console, with superior handling of accented characters
in languages using the Latin alphabet.

Please find attached. To reassemble the split archive:

Code: Select all

cat xa?-mlterm*.pet > mlterm-3.8.8_exec.pet

Then click on mlterm-3.8.8_exec.pet to install as usual.

I have also attached a simple mlterm configuration directory to get you started.
There are also in this zip file some docs explaining how to configure MLterm. IHTH.

Important note --
This mlterm was compiled on and for jrb's upupbb32-light and tested on same.
It may not work on other Pups.

Any constructive feedback appreciated. TIA.

Enjoy!

musher0 · #102 Post by **musher0** » Sat 18 May 2019, 05:57

Hello again.

I'm not sure if I should be eating humble pie, or beating my chest saying "mea
culpa" three times...

What do you think is meant by "this configure option adds all non-mutichoice
options?" Anybody knows what "non-multichoice" really means? This is what Marc
Lehmann, author of rxvt-unicode, says about the "--enable-everything option" on
the configuration line of rxvt-unicode:

--enable-everything
Add (or remove) support for all non-multichoice options listed in
"./configure --help", except for "--enable-assert" and
"--enable-256-color".

Not terribly clear to me even after reading the listing provided by

Code: Select all

./configure --help=short | more

Source: Lehman's README.configure file in the rxvt-unicode v. 9.22 source package.

On a hunch, I recompiled the same rxvt-unicode v. 9.22 with only the --prefix=/usr
setting, and by magic, the expected behavior for the ca keyboard was back in urxvt
and any app supported by it (e.g. the joe editor).

Proof in attached scrot.

So... If you are using an alternative keyboard configuratrion and the accents in your
language are not at the place they used to be in urxvt, please replace my previous
compile of rxvt-unicode ( presented a few pages back ) with this one. ( Please see
attached files below. )

Please accept my apologies for any inconvenience.

I guess this problem is now solved, technically, but not scientifically.
By that I mean, we have a solution, although I'm not certain about which
parameter exactly in < --enable-everything > caused it previously. If you
know, please chime in: not knowing the ultimate cause is still bugging me.

In any case, in spite of what appears to have been a useless run-around,
we end up with two very good, up-to-date, console utilities and some
add'l learning in our grey matter.

BFN.

jrb · #103 Post by **jrb** » Wed 29 May 2019, 03:20

I'm heading out on a road trip for the next two or three weeks. Will not be able to offer support until I get back.

Cheers, J

Sage · #104 Post by **Sage** » Thu 04 Jul 2019, 12:36

Your "two or three weeks" are up! Good, worthwhile project, so look forward to more goodies...

sheldonisaac · #105 Post by **sheldonisaac** » Thu 04 Jul 2019, 14:47

jrb (in part) wrote:[.. here is Update two.

How can I get the interfaces to be called the usual eth0 and wlan0?

Does it have to do with systemd (whatever that is)?
Can we get rid of that?

Thanks again,
Sheldon

6502coder · #106 Post by **6502coder** » Thu 04 Jul 2019, 16:53

I read somewhere that adding "net.ifnames=0" to the kernel line in the boot stanza will do the trick. I can't find that article but here's a similar discussion:

https://askubuntu.com/questions/689070/ ... ev-changes

sheldonisaac · #107 Post by **sheldonisaac** » Thu 04 Jul 2019, 16:58

6502coder wrote:I read somewhere that adding "net.ifnames=0" to the kernel line in the boot stanza will do the trick. I can't find that article but here's a similar discussion:

https://askubuntu.com/questions/689070/ ... ev-changes

Thanks, will try that.

peebee · #108 Post by **peebee** » Thu 04 Jul 2019, 19:18

see:
http://www.murga-linux.com/puppy/viewto ... 998#989998

rockedge · #109 Post by **rockedge** » Thu 04 Jul 2019, 20:30

I read somewhere that adding "net.ifnames=0" to the kernel line in the boot stanza will do the trick.

it will. I use it on all my Bionic puppy versions and now with FirstRib aka WeePup

like this :

Code: Select all

title firstrib (Void Linux Flavour)
  root (hd0,0)
  kernel /firstrib/vmlinuz  net.ifnames=0
  initrd /firstrib/initramfs.gz

Code: Select all

title Bionic64 (sda1/Bionic64)
  uuid 01cec708-a02c-4fce-8656-50f9700d951f
  kernel /Bionic64/vmlinuz   psubdir=Bionic64 pmedia=atahd pfix=fsck net.ifnames=0
  initrd /Bionic64/initrd.gz

this will force the network device count to start at eth0
_

kschewe · #110 Post by **kschewe** » Wed 28 Aug 2019, 03:51

I have a question about getting sound to work on this version.
I am using an ancient laptop and so far It runs very smooth.
I was looking in the startup modules to see if there was one for opl3sa. I have tahr 6 arcade installed and I had the same issue. I was able to find the module in the startup and voila after a reboot it worked. This is probably the most modern OS this old laptop will see.
thanks

(old)Puppy Linux Discussion Forum

(old)Puppy Linux Discussion Forum

Bionic32-light - 27 April, 2019

Re: Bionic32-light