Re: Installing Debian 10.9 Buster on iMac G5 (powerpc)

[Thomas Schmitt]

Hi,

(there is at least one answer which was not Cc'ed to you as follow-up under
  https://lists.debian.org/debian-user/2021/05/msg00358.html
)

Pat Pathmanathan wrote:
> I finally managed to install the following image on my iMac G5 :
> [...] debian-10.0.0-ppc64-NETINST-1.iso [...]
> 'Starting Gnome manager'
> (or something to that effect). It then hangs with a black screen and a small
> blinking cursor on the top left corner.
> [...] Any suggestions?

Did you already try the "powerpc" architecture as alternative to "ppc64" ?
  
https://cdimage.debian.org/cdimage/ports/10.0/powerpc/iso-cd/debian-10.0-powerpc-NETINST-1.iso

If this doesn't work either, then i propose again to ask at
  debian-powe...@lists.debian.org
Subscribe at
  https://lists.debian.org/debian-powerpc/


Have a nice day :)

Thomas

Re: generate a rss.xml from a bunch of HTML files

[Andrei POPESCU]

On Lu, 10 mai 21, 01:44:32, Emanuel Berg wrote:
> Charles Curley wrote:
> 
> > Right. However, as I found out asking elsewhere, you can
> > include HTML in Markdown.
> 
> Hehehe, let's see, first write HTML, then include it in
> Markdown, then have the static site generator generate
> HTML... brilliant :)

Surely there must be some site generator with RSS support that takes 
"plain" HTML as input.

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser


signature.asc
Description: PGP signature

Re: generate a rss.xml from a bunch of HTML files

[Andrei POPESCU]

On Du, 09 mai 21, 20:15:38, Emanuel Berg wrote:
> Dan Ritter wrote:
> 
>  ... no one did it? :O
> >>>
> >>> In FLOSS this usually means nobody else needed it.
> >> 
> >> Impossible in this, basic case. The static generator guys who
> >> also did the RSS as mentioned already needed it, and did it,
> >> only not modular to fit this purpose (IIUC from reading here).
> >
> > They totally did. As a library, not a standalone.
> >
> > Because it turns out almost everybody needs consistency, and
> > that comes from dealing with the whole problem.
> 
> Yes, but ever heard of independent tools working together?

Except that an HTML to RSS converter is is basically useless by itself.

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser


signature.asc
Description: PGP signature

Re: generate a rss.xml from a bunch of HTML files

[Emanuel Berg]

davidson wrote:

>> How can I generate a rss.xml from a bunch of HTML files?
>>
>> Tho one would think this to be quite a simple tool of
>> parsing the HTML and outputting the RSS XML dialect,
>> I can't find any tool...
>
> XSLT is a language that is sort of made for describing this
> kind of transformation.
>
> My degree of XSLT-clue is quite low, but occasionally I find
> a small project pitched to my rudimentary ability, and try
> to level up a little.
>
> Whenever I do that, I find this debian package useful:
>
>   xsltproc - XSLT 1.0 command line processor
>XSLT is an XML language for defining transformations of XML files from
>XML to some other arbitrary format, such as XML, HTML, plain text, etc.
>using standard XSLT version 1.0 stylesheets.
>.
>This package contains a command line tool that facilitates XSLT
>transformations.
>   Homepage: http://xmlsoft.org/xslt/

Right, something like that!

> Sometimes I need this one too, to tweak HTML into something xsltproc
> can deal with:
>
>   tidy - HTML/XML syntax checker and reformatter
>Tidy corrects and cleans up HTML and XML documents by fixing
>markup errors and upgrading legacy code to modern standards.
>.
>This package contains a command line tool 'tidy'.
>   Homepage: http://www.html-tidy.org/

Yep, I know that, good tool.

> It sounds to me like you want to make a script that calls
> xsltproc to apply some XSLT transformation of your own
> devising. I think if I were in your place, I would study
> a few examples like this simple one...

But the transformation should actually not be my personal
devising, it should be the HTML definition and how that
translates to RSS...

>  "The XSLT used by html2rss-web"
>  html2rss-web/rss.xsl at master · gildesmarais/html2rss-web · GitHub
>  
> https://github.com/gildesmarais/html2rss-web/blob/master/public/rss.xsl#start-of-content

Yeah... but what tool is that? I'll check it out, for sure.

-- 
underground experts united
https://dataswamp.org/~incal

Re: generate a rss.xml from a bunch of HTML files

[Emanuel Berg]

Charles Curley wrote:

> Right. However, as I found out asking elsewhere, you can
> include HTML in Markdown.

Hehehe, let's see, first write HTML, then include it in
Markdown, then have the static site generator generate
HTML... brilliant :)

-- 
underground experts united
https://dataswamp.org/~incal

Re: generate a rss.xml from a bunch of HTML files

[Emanuel Berg]

The Wanderer wrote:

> One possible difference is that the ones I've looked at
> (admittedly nowhere near all of them) seem to expect the
> input to be in some other format, to be translated into HTML
> etc., rather than letting you write the HTML etc.
> directly and doing [whatever other things] with the result.
> For example, the package description for pelican (which you
> suggested earlier) says that it requires its input to be in
> Markdown or rST.
>
> In a case where you've already written the HTML et cetera
> and want to continue to do so, these generators don't seem
> to be applicable to the input form which you have available
> and want to continue to use.

Yes, that's exactly it, thank you!

But, I was thinking, aren't there tools to do the generator or
builder?

How does that work, you write a syntax grammar and feed it
to yacc?

-- 
underground experts united
https://dataswamp.org/~incal

Intel i810 graphics working as expected in openSUSE Tumbleweed, not in Stretch, Buster or Bullseye

[Felix Miata]

I un-mothballed a 20+ year old Dell GX110 with i810e graphics in an attempt to
assist a linuxquestions.org Stretch user stuck with 640x480 VESA in X, here:


I have Stretch, Buster & Bullseye all on my Dell. Basically, absence of /dev/fb0
and /dev/dri/* stuff legacy 80x25 text mode on the ttys, and VESA 640x480 in X. 
In
Debian, none of the required kernel modules for X graphics will autoload, but 
even
when manually loaded, /dev/fb0 and /dev/dri/* remain absent. The LQ user got
similar failure with Debian-based live AntiX.

Welcome to openSUSE Tumbleweed 20210422 - Kernel 5.12.0-1.gde654dc-default
# lsmod | sort | egrep 'drm|dri|i91|intel|video' | grep -v snd
cec45056  2 i915,drm_kms_helper
drm   471040  2 i915,drm_kms_helper
drm_kms_helper225280  1 i915
fb_sys_fops16384  1 drm_kms_helper
i2c_algo_bit   16384  1 i915
i915 1847296  0
intel_pmc_bxt  16384  1 iTCO_wdt
intel_rng  16384  0
syscopyarea16384  1 drm_kms_helper
sysfillrect16384  1 drm_kms_helper
sysimgblt  16384  1 drm_kms_helper
video  49152  1 i915

Linux gx110.ij.net 5.10.0-6-686 #1 SMP Debian 5.10.28-1 (2021-04-09) i686
# lsmod | sort | egrep 'drm|dri|i91|intel|video' | grep -v snd
cec40960  2 i915,drm_kms_helper
drm   356352  2 i915,drm_kms_helper
drm_kms_helper163840  1 i915
fb_sys_fops16384  1 drm_kms_helper
i2c_algo_bit   16384  1 i915
i915 1744896  0
intel_pmc_bxt  16384  1 iTCO_wdt
intel_rng  16384  0
rng_core   16384  2 intel_rng
syscopyarea16384  1 drm_kms_helper
sysfillrect16384  1 drm_kms_helper
sysimgblt  16384  1 drm_kms_helper
video  45056  1 i915
# grep \(EE /var/log/Xorg.0.log
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[  1269.787] (EE) open /dev/dri/card0: No such file or directory
[  1269.787] (EE) open /dev/dri/card0: No such file or directory
[  1269.789] (EE) Unable to find a valid framebuffer device
[  1269.790] (EE) open /dev/fb0: No such file or directory

Web searches haven't directed me to any clues how to reconcile X working as
expected automagically in Tumbleweed with not working at all in Debian. Does
anyone here have recent experience with making i810 or i815 graphics hardware 
work
as expected, e.g. 1680x1050 native on an LCD display?
-- 
Evolution as taught in public schools is, like religion,
based on faith, not based on science.

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/

Re: PC fan getting very loud because of CPU load

[Celejar]

On Sat, 8 May 2021 17:18:35 -0400
Dan Ritter  wrote:

> Bret Busby wrote: 
> > 
> > I think this goes to the issue of client side processing, as opposed to
> > server side processing ( I believe, and, argue, that all processing involved
> > with web sites, should be server side, if the web sites are competently and
> > benignly written, and that client side processing, is malignant), and I
> > suggest that it could be worth viewing the source code of the web site(s)
> > responsible for the problem.
> 
> There are some reasonable cases for client-side computing. It's
> certainly badly overused.
> 
> > have a problem, could be in the use of plugins in firefox - some
> > particularly malicious web sites put up quite aggressive fights against ad
> > blocking and tracking blocking plugins, and try to burn out computers of
> > users who object to ads and being tracked and who object to websites trying
> > to steal the users' identities and personal information.
> 
> I've never encountered this -- please let me know some sample
> URLs whenever you have time. 

https://old.reddit.com/r/uBlockOrigin/comments/mu2qcm/indefinitely_increasing_blocked_elements/

I'm not sure that I would interpret this sort of thing as the website
putting up an aggressive fight or deliberately trying to burn out
anyone's computer, but it is an annoyance that one sometimes encounters.

Celejar

Re: generate a rss.xml from a bunch of HTML files

[Charles Curley]

On Sun, 09 May 2021 17:38:06 -0400
The Wanderer  wrote:

> One possible difference is that the ones I've looked at (admittedly
> nowhere near all of them) seem to expect the input to be in some other
> format, to be translated into HTML etc., rather than letting you write
> the HTML etc. directly and doing [whatever other things] with the
> result. For example, the package description for pelican (which you
> suggested earlier) says that it requires its input to be in Markdown
> or rST.

Right. However, as I found out asking elsewhere, you can include HTML
in Markdown. For example, Markdown doesn't by itself provide targets
for links. So in a glossary, I do this:

* EPROM: Erasable Programmable Read Only Memory,

Now I can make a link to that target:

... and burn to ROM (or, more likely, [EPROM](#eprom)).

Feel free to experiment with it. I suspect you might have to remove the
 tags and everything outside them, and then have your static site
generator provide those. That might end up being an advantage.

-- 
Does anybody read signatures any more?

https://charlescurley.com
https://charlescurley.com/blog/


pgplq6SYQ6QCR.pgp
Description: OpenPGP digital signature

Re: Installing Debian 10.9 Buster on iMac G5 (powerpc)

[Nicholas Geovanis]

On Sun, May 9, 2021, 1:18 PM Pat Pathmanathan 
wrote:

> I finally managed to install the following image on my iMac G5 :
> https://cdimage.debian.org/cdimage/ports/current/
>
> https://saimei.ftp.acc.umu.se/cdimage/ports/current/debian-10.0.0-ppc64-NETINST-1.iso
>
> The installation went fine. When I start the system, the grub menu loads
> allowing me to select Debian Linux to boot. On selecting the option it goes
> through the startup process and the last thing was 'Starting Gnome manager'
> (or something to that effect). It then hangs with a black screen and a
> small blinking cursor on the top left corner.
>
> It took many attempts to get here. Still no luck. Any suggestions?
>

Based on similar symptoms on Intel platforms, check brightness and contrast
on your monitor. It may simply be too low to see. Or as already suggested,
try other available sessions.

Pat
>
>
>

Re: generate a rss.xml from a bunch of HTML files

[Emanuel Berg]

Dan Ritter wrote:

 ... no one did it? :O
>>>
>>> In FLOSS this usually means nobody else needed it.
>> 
>> Impossible in this, basic case. The static generator guys who
>> also did the RSS as mentioned already needed it, and did it,
>> only not modular to fit this purpose (IIUC from reading here).
>
> They totally did. As a library, not a standalone.
>
> Because it turns out almost everybody needs consistency, and
> that comes from dealing with the whole problem.

Yes, but ever heard of independent tools working together?
E.g., this zsh

  hits=$(cut -d ' ' -f 2,8 $log | grep $p | sort | cut -d ' ' -f 1 | uniq -c | 
wc -l)

to make up a script or little program, e.g.

  https://dataswamp.org/~incal/blog/meta/piles

that produces the desired result:

  https://dataswamp.org/~incal/blog/hits.txt

Anyway that's what I would like here, if anyone else wants
something else somewhere else by all means don't make it
standalone, put it in a library... that we can't access (?) to
get a standalone either, maybe, it is too consistent with
everything else it is impossible to make out...

>>> Why do you need it? Maybe we can suggest other means to
>>> achieve your (true) goal.
>> 
>> True goal!
>> 
>> I have a blog [1], just a bunch of HTML5/CSS files,
>> absolutely nothing advanced, and I'd like an RSS file [2]
>> which is generated from the HTML files (not the CSS, so
>> even simpler actually) so I for example can submit it here
>> [2] and read it with Gnus :)
>> 
>> So one needs a parser to parse the HTML, dispose of
>> unnecessary stuff, walk the tree (ha) and output it as an
>> RSS file.
>
> That's what all those static site generators do.
>
> As a bonus, they usually offer templating (so the structure
> of pages looks similar to each other) and shared CSS (so the
> visuals are decoupled from the structure, and can be changed
> without going in to every page to repeat tweaks.)

I don't have to do that, as you see pages look similar and
there is just one CSS file:

  https://dataswamp.org/~incal/blog/index.html
  https://dataswamp.org/~incal/blog/global.css
  
> It sounds like you've written about a quarter of a static site
> generator already. 

Heh, no, what do you mean?

> You could continue down that path, or just install Pelican
> and be happy in about a day.

I would if it would do what I want namely get an RSS file and
only that. (Maybe it can even do that, I don't know.)

> Don't become that person who gets angry at wheels because
> wheels need axles and bearings when all you ever needed was
> a couple of good round logs.

Who is angry, I'm not angry and especially not at wheels:

  https://dataswamp.org/~incal/work-photos/stand.jpg
  https://dataswamp.org/~incal/#bike

But no one has or have heard of a CLI parser or shell tool...?

-- 
underground experts united
https://dataswamp.org/~incal

Re: generate a rss.xml from a bunch of HTML files

[The Wanderer]

On 2021-05-09 at 15:36, Dan Ritter wrote:

> Emanuel Berg wrote:

>>> It sounds like you've written about a quarter of a static site 
>>> generator already.
>> 
>> Heh, no, what do you mean?
>> 
>>> You could continue down that path, or just install Pelican and be
>>> happy in about a day.
>> 
>> I would if it would do what I want namely get an RSS file and only
>> that. (Maybe it can even do that, I don't know.)
> 
> If only you read the documentation, you would know.
> 
>> But no one has or have heard of a CLI parser or shell tool...?
> 
> That is what a static site generator is.
> 
> It's a command-line tool that takes a directory full of content 
> files, a set of templates, a CSS file or 3, and spits out a web site
> ready to be served by your favorite web server, including the thing
> you asked for: an RSS or ATOM feed.

One possible difference is that the ones I've looked at (admittedly
nowhere near all of them) seem to expect the input to be in some other
format, to be translated into HTML etc., rather than letting you write
the HTML etc. directly and doing [whatever other things] with the
result. For example, the package description for pelican (which you
suggested earlier) says that it requires its input to be in Markdown or
rST.

In a case where you've already written the HTML et cetera and want to
continue to do so, these generators don't seem to be applicable to the
input form which you have available and want to continue to use.


For myself, if I were to write a static Website, I would probably do it
by literally writing the raw HTML (and probably not even using CSS, as
it goes against the "don't define how the output should look" principle
which I understand to have been part of HTML as it was originally
designed); the HTML, et cetera, output by any such generator would
probably not be in line with my preferences.

Then again, I'm not sure a static site of that type would really be
suitable for having an RSS-type feed of, since by definition it would be
static and not receiving updates such as might go into such a feed.

-- 
   The Wanderer

The reasonable man adapts himself to the world; the unreasonable one
persists in trying to adapt the world to himself. Therefore all
progress depends on the unreasonable man. -- George Bernard Shaw



signature.asc
Description: OpenPGP digital signature

Re: generate a rss.xml from a bunch of HTML files

[davidson]

On Sun, 9 May 2021 Emanuel Berg wrote:

How can I generate a rss.xml from a bunch of HTML files?

Tho one would think this to be quite a simple tool of parsing
the HTML and outputting the RSS XML dialect, I can't find any
tool...


XSLT is a language that is sort of made for describing this kind of
transformation.

My degree of XSLT-clue is quite low, but occasionally I find a small
project pitched to my rudimentary ability, and try to level up a
little.

Whenever I do that, I find this debian package useful:

  xsltproc - XSLT 1.0 command line processor
   XSLT is an XML language for defining transformations of XML files from
   XML to some other arbitrary format, such as XML, HTML, plain text, etc.
   using standard XSLT version 1.0 stylesheets.
   .
   This package contains a command line tool that facilitates XSLT
   transformations.
  Homepage: http://xmlsoft.org/xslt/

Sometimes I need this one too, to tweak HTML into something xsltproc
can deal with:

  tidy - HTML/XML syntax checker and reformatter
   Tidy corrects and cleans up HTML and XML documents by fixing
   markup errors and upgrading legacy code to modern standards.
   .
   This package contains a command line tool 'tidy'.
  Homepage: http://www.html-tidy.org/

If I wanted to do what you want to do, those are the tools I'd use.

My couple of round logs, so to speak ;)


tt-rss maybe, but when I install it it tries to setup a MySQL
database which fails.

I don't know why, but it seems too involved anyway, there
isn't a webpile2rss tool like this or something:

 $ webpile2rss *.html > rss.xml # sweet


It sounds to me like you want to make a script that calls xsltproc to
apply some XSLT transformation of your own devising. I think if I were
in your place, I would study a few examples like this simple one...

 "The XSLT used by html2rss-web"
 html2rss-web/rss.xsl at master · gildesmarais/html2rss-web · GitHub
 
https://github.com/gildesmarais/html2rss-web/blob/master/public/rss.xsl#start-of-content

...and this somewhat more complex-looking one...

 "W3C RSS 1.0 News Feed Creation How-To"
 https://www.w3.org/2001/10/glance/doc/howto

...and model my efforts on something in between.

Maybe you could find more appropriate models yourself, since you know
what you're looking for better than I do. But those are the two that
caught my eye first.

Anyways, good luck with your project.

--
Ce qui est important est rarement urgent
et ce qui est urgent est rarement important
-- Dwight David Eisenhower

Re: generate a rss.xml from a bunch of HTML files

[Emanuel Berg]

> I have a blog, just a bunch of HTML5/CSS files, absolutely
> nothing advanced, and I'd like an RSS file which is
> generated from the HTML files (not the CSS, so even simpler
> actually) so I for example can submit it [to Gwene] and read
> it with Gnus

Speaking of Emacs (Emacs Gnus), in GNU ELPA there is webfeeder
[1] - actually I'd prefer a shell tool for this, but that's
just me - but after trying to get it to work for a long time,
I now give up. I don't know if that is also just me or if it
actually doesn't work?

The simple-enough example and boilerplate function looks like
this

  (webfeeder-build
  "atom.xml"
  "./public"
  "https://example.org/";
  '("post1.html" "post2.html" "post3.html")
  :title "My homepage"
  :description "A collection of articles in Atom")

so if anyone wants to try it should be simple. Tell me if you
have more success.

[1] https://gitlab.com/Ambrevar/emacs-webfeeder

-- 
underground experts united
https://dataswamp.org/~incal

Re: Installing Debian 10.9 Buster on iMac G5 (powerpc)

[Felix Miata]

Pat Pathmanathan composed on 2021-05-09 19:02 (UTC+0100):

> I finally managed to install the following image on my iMac G5 :
> https://cdimage.debian.org/cdimage/ports/current/
> https://saimei.ftp.acc.umu.se/cdimage/ports/current/debian-10.0.0-ppc64-NETINST-1.iso

> The installation went fine. When I start the system, the grub menu loads
> allowing me to select Debian Linux to boot. On selecting the option it goes
> through the startup process and the last thing was 'Starting Gnome manager'
> (or something to that effect). It then hangs with a black screen and a
> small blinking cursor on the top left corner.
> 
> It took many attempts to get here. Still no luck. Any suggestions?

Is it truly hung, or can you reach a login prompt via Ctrl-Alt-F3 or remote?

Did you select autologin at installation time?
-- 
Evolution as taught in public schools is, like religion,
based on faith, not based on science.

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/

Re: generate a rss.xml from a bunch of HTML files

[Dan Ritter]

Emanuel Berg wrote: 
> > It sounds like you've written about a quarter of a static site
> > generator already. 
> 
> Heh, no, what do you mean?
> 
> > You could continue down that path, or just install Pelican
> > and be happy in about a day.
> 
> I would if it would do what I want namely get an RSS file and
> only that. (Maybe it can even do that, I don't know.)

If only you read the documentation, you would know.

> But no one has or have heard of a CLI parser or shell tool...?

That is what a static site generator is.

It's a command-line tool that takes a directory full of content
files, a set of templates, a CSS file or 3, and spits out a
web site ready to be served by your favorite web server,
including the thing you asked for: an RSS or ATOM feed.

-dsr-

Re: Can't set Compose key, Bullseye, XFCE

[Greg Wooledge]

On Sun, May 09, 2021 at 12:34:16PM -0600, Charles Curley wrote:
> On Sun, 9 May 2021 18:13:39 +0100
> Keith Edmunds  wrote:
> 
> > I have set the compose key (settings, keyboard, Layout tab) to Right
> > Ctrl, but it has no effect. Tried a couple of other keys too.
> > 
> > Same setup worked on Buster without problem (this is a new install,
> > not an upgrade). Am I missing something obvious, or is this a bug in
> > Bullseye?
> 
> I see the same thing on both Buster and Bullseye. I also tried another
> key (left Windows) on Bullseye, and it also failed.

For whatever it's worth, mine is working fine on bullseye.

keysym Super_R = Multi_key

But to be fair, I have not rebooted nor restarted X11 in some time
(uptime 44 days).

Re: Can't set Compose key, Bullseye, XFCE

[Charles Curley]

On Sun, 9 May 2021 18:13:39 +0100
Keith Edmunds  wrote:

> I have set the compose key (settings, keyboard, Layout tab) to Right
> Ctrl, but it has no effect. Tried a couple of other keys too.
> 
> Same setup worked on Buster without problem (this is a new install,
> not an upgrade). Am I missing something obvious, or is this a bug in
> Bullseye?

I see the same thing on both Buster and Bullseye. I also tried another
key (left Windows) on Bullseye, and it also failed.

-- 
Does anybody read signatures any more?

https://charlescurley.com
https://charlescurley.com/blog/

Networkmanager hell !

[Bhasker C V]

Hi,

 I recently compiled the wl driver for my BCM4331 and the driver loads 
fine.


 Networkmanager lists wifi network but does not connect ! (nm-applet 
shows a single dot and then keeps revolving around until it fails)


However,

 1. I stop Networkmanager and wpa_supplicant

 2. I generate a simple config using wpa_passphrase

 3. connect via wpa_supplicant -Dwext -cconfig -iwlp4s0

  and it works !

I have tried to google through as much as possible to understand why 
wpa_supplicant does not connect under NetworkManager but works fine when 
run stand-alone !


I would appreciate if someone can tell me what is happening. I can send 
the logs if needed.


Is there a way to debug networkmanager ?

All that I can do by running network-manager from commandline in 
foreground is


wpa_supplicant[28827]: wlp4s0: CTRL-EVENT-SSID-TEMP-DISABLED id=0 
ssid="myssid" auth_failures=2 duration=20 reason=CONN_FAILED


Before someone says that this is password issue, i am pretty sure I am 
typing the password correct and have confirmed it by looking into the 
system-connections folder for that connection.


If I use another USB adapter, networkmanager is able to connecto the 
same wifi.


Please ! I am going mad here ! There is no good info anywhere to debug 
wpa_supplicant working under networkmanager (OR) to dump the config 
generated by networkmanager



--
Bhasker C V
Secure Mails: http://keys.gnupg.net/pks/lookup?op=get&search=0x4D05FEEC54E47413
Registered Linux User: #306349

Can't set Compose key, Bullseye, XFCE

[Keith Edmunds]

I have set the compose key (settings, keyboard, Layout tab) to Right Ctrl,
but it has no effect. Tried a couple of other keys too.

Same setup worked on Buster without problem (this is a new install, not an
upgrade). Am I missing something obvious, or is this a bug in Bullseye?

Re: generate a rss.xml from a bunch of HTML files

[Emanuel Berg]

Andrei POPESCU wrote:

>> ... no one did it? :O
>
> In FLOSS this usually means nobody else needed it.

Impossible in this, basic case. The static generator guys who
also did the RSS as mentioned already needed it, and did it,
only not modular to fit this purpose (IIUC from reading here).

> Why do you need it? Maybe we can suggest other means to
> achieve your (true) goal.

True goal!

I have a blog [1], just a bunch of HTML5/CSS files, absolutely
nothing advanced, and I'd like an RSS file [2] which is
generated from the HTML files (not the CSS, so even simpler
actually) so I for example can submit it here [2] and read it
with Gnus :)

So one needs a parser to parse the HTML, dispose of
unnecessary stuff, walk the tree (ha) and output it as
an RSS file.

Then I can have a Makefile to update the RSS file if and when
I do changes and add more posts.

[1] https://dataswamp.org/~incal/blog
[2] http://gwene.org

-- 
underground experts united
https://dataswamp.org/~incal

Re: generate a rss.xml from a bunch of HTML files

[Dan Ritter]

Emanuel Berg wrote: 
> Andrei POPESCU wrote:
> 
> >> ... no one did it? :O
> >
> > In FLOSS this usually means nobody else needed it.
> 
> Impossible in this, basic case. The static generator guys who
> also did the RSS as mentioned already needed it, and did it,
> only not modular to fit this purpose (IIUC from reading here).

They totally did. As a library, not a standalone.

Because it turns out almost everybody needs consistency, and
that comes from dealing with the whole problem.

 
> > Why do you need it? Maybe we can suggest other means to
> > achieve your (true) goal.
> 
> True goal!
> 
> I have a blog [1], just a bunch of HTML5/CSS files, absolutely
> nothing advanced, and I'd like an RSS file [2] which is
> generated from the HTML files (not the CSS, so even simpler
> actually) so I for example can submit it here [2] and read it
> with Gnus :)
> 
> So one needs a parser to parse the HTML, dispose of
> unnecessary stuff, walk the tree (ha) and output it as
> an RSS file.

That's what all those static site generators do.

As a bonus, they usually offer templating (so the structure of
pages looks similar to each other) and shared CSS (so the
visuals are decoupled from the structure, and can be changed
without going in to every page to repeat tweaks.)

It sounds like you've written about a quarter of a static site
generator already. 

You could continue down that path, or just install Pelican and
be happy in about a day.

Don't become that person who gets angry at wheels because wheels
need axles and bearings when all you ever needed was a couple of
good round logs. 

-dsr-

Re: generate a rss.xml from a bunch of HTML files

[Emanuel Berg]

Dan Ritter wrote:

> hugo, jekyll, lektor, nanoc, staticsite, pelican

... unless they can be told by way of options perhaps to do
just the HTML-to-RSS part?

-- 
underground experts united
https://dataswamp.org/~incal

Re: generate a rss.xml from a bunch of HTML files

[Emanuel Berg]

Dan Ritter wrote:

>> I don't know why, but it seems too involved anyway, there
>> isn't a webpile2rss tool like this or something:
>> 
>>   $ webpile2rss *.html > rss.xml # sweet
>
> There isn't one packaged in Debian, but there are libraries
> packaged which would allow you to build one.

... no one did it? :O

> You might be happy with a static site generator, which takes
> a directory full of documents and arranges them into
> a website, usually including an RSS or ATOM feed.

Well, I already have that, well, whatever it means, I want
only the RSS...

-- 
underground experts united
https://dataswamp.org/~incal

generate a rss.xml from a bunch of HTML files

[Emanuel Berg]

How can I generate a rss.xml from a bunch of HTML files?

Tho one would think this to be quite a simple tool of parsing
the HTML and outputting the RSS XML dialect, I can't find any
tool...

tt-rss maybe, but when I install it it tries to setup a MySQL
database which fails.

I don't know why, but it seems too involved anyway, there
isn't a webpile2rss tool like this or something:

  $ webpile2rss *.html > rss.xml # sweet

?

-- 
underground experts united
https://dataswamp.org/~incal

Re: repeated system mail, /etc/.pwd.lock ?

[Emanuel Berg]

David Wright wrote:

> FYI:
>
> I installed iwatch, and that immediately generated two messages from
> /etc/.etckeeper. Then I upgraded:
>
>   apt apt-doc apt-utils bind9-host curl dnsutils exim4
> exim4-base exim4-config exim4-daemon-light
>   firefox-esr firefox-esr-l10n-en-gb gstreamer1.0-gl
> gstreamer1.0-libav gstreamer1.0-plugins-bad
>   gstreamer1.0-plugins-base gstreamer1.0-plugins-good
> gstreamer1.0-pulseaudio gstreamer1.0-x
>   libapt-inst2.0 libapt-pkg5.0 libbind9-161 libcurl3-gnutls
> libcurl4 libdns-export1104 libdns1104
>   libgstreamer-gl1.0-0 libgstreamer-plugins-bad1.0-0
> libgstreamer-plugins-base1.0-0 libirs161
>   libisc-export1100 libisc1100 libisccc161 libisccfg163
> libjs-underscore libldb1 liblwres161
>   libopenjp2-7 openjdk-11-jre openjdk-11-jre-headless
> wpasupplicant xserver-common
>   xserver-xorg-core xserver-xorg-legacy
> 44 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
>
> and got 387 more messages.
>
> I then added one new user, which generated 97 more, where
> /etc/.pwd.lock was the subject of four of them.
>
> Purging iwatch then generated a final three.
>
> So the OP's 2757 is no surprise with the
> default configuration.

Really, I was 100% it was something _I_ did...

-- 
underground experts united
https://dataswamp.org/~incal

Re: generate a rss.xml from a bunch of HTML files

[Andrei POPESCU]

On Du, 09 mai 21, 14:19:44, Emanuel Berg wrote:
> Dan Ritter wrote:
> 
> >> I don't know why, but it seems too involved anyway, there
> >> isn't a webpile2rss tool like this or something:
> >> 
> >>   $ webpile2rss *.html > rss.xml # sweet
> >
> > There isn't one packaged in Debian, but there are libraries
> > packaged which would allow you to build one.
> 
> ... no one did it? :O

In FLOSS this usually means nobody else needed it.

Why do you need it? Maybe we can suggest other means to achieve your 
(true) goal.

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser


signature.asc
Description: PGP signature

Re: generate a rss.xml from a bunch of HTML files

[Dan Ritter]

Emanuel Berg wrote: 
> How can I generate a rss.xml from a bunch of HTML files?
> 
> Tho one would think this to be quite a simple tool of parsing
> the HTML and outputting the RSS XML dialect, I can't find any
> tool...
> 
> tt-rss maybe, but when I install it it tries to setup a MySQL
> database which fails.

That's a web-based RSS reader, not a generator.

> I don't know why, but it seems too involved anyway, there
> isn't a webpile2rss tool like this or something:
> 
>   $ webpile2rss *.html > rss.xml # sweet

There isn't one packaged in Debian, but there are libraries
packaged which would allow you to build one.

You might be happy with a static site generator, which takes a
directory full of documents and arranges them into a website,
usually including an RSS or ATOM feed.

Packaged in Debian:

hugo, jekyll, lektor, nanoc, staticsite, pelican

-dsr-

Re: PC fan getting very loud

[Tixy]

On Sun, 2021-05-09 at 12:13 +0200, Thomas Schmitt wrote:
> Hi,
> 
> Alexander V. Makartsev wrote:
> > > Put this mini-PC under any, not just Firefox, multi-threaded workload
> > > and you will have the same outcome.
> 
> deloptes wrote:
> > What would be a good test to find out if this theory is true?
> 
> Did you already compare the Firefoxes about the number of threads they run ?
> (top -H or top interactive command "H" come to my mind.)
> 
> I remember the request in this mail thread that you should post an URL
> which triggers the problem for you. It might help to get more measurements
> and opinions.
> (A known browser hog is Gitlab. Just looking at
>   https://salsa.debian.org/optical-media-team
> brings my Xeon CPU to full speed for about a minute

Doesn't for me, about 50% of a CPU for a couple of seconds. There
again, I've got a pretty fast CPU (an i7-9700) which I guess makes all
the difference.

-- 
Tixy

Re: PC fan getting very loud

[Thomas Schmitt]

Hi,

Alexander V. Makartsev wrote:
> > Put this mini-PC under any, not just Firefox, multi-threaded workload
> > and you will have the same outcome.

deloptes wrote:
> What would be a good test to find out if this theory is true?

Did you already compare the Firefoxes about the number of threads they run ?
(top -H or top interactive command "H" come to my mind.)

I remember the request in this mail thread that you should post an URL
which triggers the problem for you. It might help to get more measurements
and opinions.
(A known browser hog is Gitlab. Just looking at
  https://salsa.debian.org/optical-media-team
brings my Xeon CPU to full speed for about a minute and keeps it slightly
excited until i go to a boring page like
  
https://salsa.debian.org/optical-media-team/libisofs/-/raw/89c99327282f926c5ba40e49a67ac051cd1fe24a/debian/changelog
and hide the browser window behind my many xterms.)


Have a nice day :)

Thomas

Re: PC fan getting very loud

[Alexander V. Makartsev]

On 09.05.2021 12:49, deloptes wrote:

Alexander V. Makartsev wrote:


On 09.05.2021 02:30, deloptes wrote:

Charles Curley wrote:


Fujitsu ESPRIMO Q520 when opening some sh*tty web sitesin firefox the
fan gets extremly noisy.

I have a similar problem with some of my older laptops. I have switched
to Vivaldi (based on chromium), and that has reduced the problem
considerably. Firefox is a resource hog.

Q520 is not a week/old machine it is i5 cpu


I think you have unrealistic expectations.
This unit has 4th gen 2 core CPU rated at 35W TDP. [1] Which is not as
efficient as modern CPUs and don't have the ability to step-down to
low-frequency (under 1GHz) modes.
Turbo Boost doesn't count in the TDP rating and it makes this mini-PC to
generate even more heat.
That amount of heat has to be dissipated with super small heatsink and
cooling fan. [2]
Keep in mind, that fan also has to cool down every other part of this
mini-PC, like ICH, RAM, HDD and power circuit components.
No wonder it struggles to keep up and produce so much noise.
Put this mini-PC under any, not just Firefox, multi-threaded workload
and you will have the same outcome.


Thank you - I've been thinking of this, but somehow I am not convinced. See
the CPU frequency (800).
What would be a good test to find out if this theory is true? Should I test
by opening the case. If theory is true with proper cooling it wouldn't run
crazy. Correct?
You can monitor and log temperatures and other characteristics with 
'psensor'.
TCase of i5-4570T is only 66.35°C and that means CPU will have to keep 
the temperature at that point or lower.
I'd expect an idle temperature of this CPU (without any load) with this 
cooling solution will be already around 40-45°C if not higher.
By simple launching of any application will make the CPU load to spike 
to 100% and heat to build up.
I'm not sure if disabling CPU vulnerabilities mitigations 
(meltdown\spectre) would improve the performance of the CPU, allowing it 
to go into idle state sooner and this way make it cooler without 
resorting to throttling.
Opening the case won't help much and will probably be more harmful if 
the cooling solution was designed to make the air flow through the case 
and this way cooldown the other components of this unit.
I doubt you can do anything to improve cooling efficiency of this 
mini-PC, because its cooling solution is not standard and you can't just 
replace it with a more efficient one, like you can with standard PC.
Basically, a small sized cooler fan means it will have higher RPM 
(3000+) and high RPM means it will produce a lot of loud noise.



C700

vendor_id   : GenuineIntel
cpu family  : 6
model   : 42
model name  : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
stepping: 7
microcode   : 0x2f
cpu MHz : 1720.697
cache size  : 6144 KB

Q520

vendor_id   : GenuineIntel
cpu family  : 6
model   : 60
model name  : Intel(R) Core(TM) i5-4570T CPU @ 2.90GHz
stepping: 3
microcode   : 0x28
cpu MHz : 800.998
cache size  : 4096 KB



--
With kindest regards, Alexander.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄

Re: why btdownloadcurses can't open it

[davidson]

On Sat, 8 May 2021 Long Wind wrote:

when i open attached torrent with  btdownloadcurses,


Which debian package is this "btdownloadcurses" command from? As far
as I know, it could be from either bittorrent or bittornado.


it says:
got bad file info - path ~最新最快影片每日更新.url disallowed for security reasons


My very uninformed guess is that your bittorrent client reads the
.torrent file and sees that (according to information in the .torrent
file) it is supposed to store some downloaded content in a file with
that name.

But the bittorrent client doesn't like that name. It doesn't want to
create a file with that name. Maybe it doesn't like the name because
it doesn't like path components that begin with a '~' character.

What I have said above is only a guess.


they recommend using bitcomet, which supports Windows/android/macos,
not linux 


Bitcomet (as far as I know) is not packaged in debian. As didier
pointed out, Debian has a lot of other bittorrent clients packaged.

Find one of those that looks like it might suit you, and try it
out. That is what I would do.


why btdownloadcurses can't open it?


It seems to me that it *did* open the bittorrent file you've
attached. It opened it, parsed the information inside, found some
information it didn't like, and reported that to you.

A bittorrent file is not content. Instead, it is information *about*
content (some collection of files) that is stored (hopefully) on other
people's devices. In other words, it's just meta-data that can help
you obtain the data you are interested in. In a bittorrent file is
information that a bittorrent client can use, to find those other
devices, and then ask some of them to send you pieces of that content.

--
Ce qui est important est rarement urgent
et ce qui est urgent est rarement important
-- Dwight David Eisenhower

Re: Social-media antipathy (was Re: How i can optimize my operating system?)

[Will Mengarini]

* Andrei POPESCU  [21-03/12=Fr 17:51 +0200]:
> When Facebook was in its infancy (at least in my country)
> they were spamming my e-mail inviting me to join [...]

I wanted to know what "my country" designated, so ...

- `g Andrei POPESCU` ---
Andrei Popescu is a Romanian lawyer and a judge at the General Court
of the European Union.  He graduated in law from the University of
Bucharest in 1971 and obtained his doctorate in 1980.  -- Wikipedia

Andrei P - VP of Software Engineering - Android - Google Inc ...
uk.linkedin.com/andreip

Dr Andrei Popescu | Computer Science | The University of Sheffield

Andrei Popescu-Belis - Professor of computer science at HEIG-VD
iict-space.heig-vd.ch/apu

Andrei Popescu.  Carnegie Master.  As a passionate musician, [...]

Dr Ion-Andrei Popescu.  Ortho and trauma surgeon [...]

Andrei Popescu (born 1985-02-20=We) is a Romanian professional footballer

Andrei Popescu.  Singapore.  Co-Founder of COSS.IO & SCX Holdings [...]

Andrei Popescu(. Moscow, INR. )

Andrei Popescu (37) is a tennis player from Romania.

Andrei Popescu.  Andrei Popescu is an actor,
known for Professor Thompson (2018).

Dr Andrei Popescu is an Emergency Medicine Specialist in Los Angeles CA.

Dr Andrei Popescu, MD is a board certified internist in Wooster, Ohio.


You, sir, have led a rich, varied, well-traveled, and colorful life!

-- 
 Will Mengarini  
 Free software: the Source will be with you, always.
perl -le"print unpack '%C*',MENGARINI"

Re: PC fan getting very loud

[deloptes]

Alexander V. Makartsev wrote:

> On 09.05.2021 02:30, deloptes wrote:
>> Charles Curley wrote:
>>
 Fujitsu ESPRIMO Q520 when opening some sh*tty web sitesin firefox the
 fan gets extremly noisy.
>>> I have a similar problem with some of my older laptops. I have switched
>>> to Vivaldi (based on chromium), and that has reduced the problem
>>> considerably. Firefox is a resource hog.
>> Q520 is not a week/old machine it is i5 cpu
>>
> I think you have unrealistic expectations.
> This unit has 4th gen 2 core CPU rated at 35W TDP. [1] Which is not as
> efficient as modern CPUs and don't have the ability to step-down to
> low-frequency (under 1GHz) modes.
> Turbo Boost doesn't count in the TDP rating and it makes this mini-PC to
> generate even more heat.
> That amount of heat has to be dissipated with super small heatsink and
> cooling fan. [2]
> Keep in mind, that fan also has to cool down every other part of this
> mini-PC, like ICH, RAM, HDD and power circuit components.
> No wonder it struggles to keep up and produce so much noise.
> Put this mini-PC under any, not just Firefox, multi-threaded workload
> and you will have the same outcome.
> 

Thank you - I've been thinking of this, but somehow I am not convinced. See
the CPU frequency (800).
What would be a good test to find out if this theory is true? Should I test
by opening the case. If theory is true with proper cooling it wouldn't run
crazy. Correct?


> 
> [1]
>
https://ark.intel.com/content/www/us/en/ark/products/75045/intel-core-i5-4570t-processor-4m-cache-up-to-3-60-ghz.html
> [2] https://www.youtube.com/watch?v=TD9y0bkHKjQ
> 



C700

vendor_id   : GenuineIntel
cpu family  : 6
model   : 42
model name  : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
stepping: 7
microcode   : 0x2f
cpu MHz : 1720.697
cache size  : 6144 KB

Q520

vendor_id   : GenuineIntel
cpu family  : 6
model   : 60
model name  : Intel(R) Core(TM) i5-4570T CPU @ 2.90GHz
stepping: 3
microcode   : 0x28
cpu MHz : 800.998
cache size  : 4096 KB

Re: why btdownloadcurses can't open it

[didier gaumet]

Hello,

please do not attach a (possibly corrupted) file to your post

the error you get is reported (rightly or wrongly) as bittornado 
specific and the software you use seems based on bittornado

http://support.proaudiotorrents.org/knowledgebase.php?article=7
http://manpages.ubuntu.com/manpages/bionic/man1/btdownloadcurses.bittornado.1.html

I would suggest trying a bittorrent client already packaged by Debian
https://packages.debian.org/search?suite=stable§ion=all&arch=any&searchon=all&keywords=bittorrent

Re: PC fan getting very loud because of CPU load

[deloptes]

Bret Busby wrote:

> I think this goes to an issue, over which, I tend to get into heated
> arguments, including with my wife, who is a software developer, who
> develops web sites that I believe to be responsible.
> 
> In the original post, from memory, was stated that it happened with some
> web sites, and, not with others, and, this involves an issue of system
> load, due to particular web sites.
> 

Yes but it does not seem to be relevant, because I installed  same firefox
on both PCs and open same number of windows with same content.
On the C700 no CPU load and on the Q520 the described behavior just few
seconds after opening the first link.
So it must be something with the Q520 specifically.
I now will install also the same kernel on both machines to see if it makes
a difference.

> I think this goes to the issue of client side processing, as opposed to
> server side processing ( I believe, and, argue, that all processing
> involved with web sites, should be server side, if the web sites are
> competently and benignly written, and that client side processing, is
> malignant), and I suggest that it could be worth viewing the source code
> of the web site(s) responsible for the problem.
> 
> I sometimes encounter web sites that are so badly and heavily client
> side weighted, that it can take five to ten minutes, to get a response
> from  a key press, in a form, and, other web sites whizz through stuff,
> on the same computer, in the same web browser.
> 
> So, I tend to have more than one web browser running at the same time,
> with firefox script enabled, and, the other(s) script disabled.. On my
> other system, I have firefox running, script enabled, and, Seamonkey
> running, script disabled.
> 
> Another issue, depending upon the nature of the web site(s) with which
> you have a problem, could be in the use of plugins in firefox - some
> particularly malicious web sites put up quite aggressive fights against
> ad blocking and tracking blocking plugins, and try to burn out computers
> of users who object to ads and being tracked and who object to websites
> trying to steal the users' identities and personal information.
> 

Yes, this is true - I have to inspect this. On the C700 there is ad-blocker.
on the Q520 I am not sure.

> So, if you can find a trusted web site, I suggest temporarily disabling
> all plugins, and, monitoring the effect, if any, on your system load - I
> would not do this, with the web site(s) responsible for this problem - I
> think the system load problem, could be your computer defending itself
> against an attack.
> 
> I also wonder whether you notice any unexpected massive data traffic.
> 

no

> I have, from time to time, noticed unexpected sustained downloads, using
> up tens of gigabytes of my quota.
> 
> Unfortunately, insofar as I am aware, Linux does not have any packages
> that indicate what websites are responsible for Internet traffic; if I
> notice sustained downloading of over half a megabyte per second, all
> that I can do, is turn off the networking, for a couple of hours, and,
> check to see whether it resumes the unsolicited traffic.

This implies installing a device between the modem and the PC, but at the
end I may also do this. It will just take time

thank you for your long post and opinion

Re: repeated system mail, /etc/.pwd.lock ?

[Emanuel Berg]

Jonathan Dowland wrote:

> On Wed, May 05, 2021 at 03:51:16PM +0200, Emanuel Berg wrote:
>
>> I learned something as well, how to delete mails. First see
>> how many mails there are, say there are 756, then type
>> t 1-756 RET and then hold down q :)
>
> That's a slow way. "T iwatch ; d" or "T ~b iwatch ; d" would
> be faster (using ; bound to tag-prefix: apply next function
> to tagged messages)

d * even faster perhaps...

-- 
underground experts united
https://dataswamp.org/~incal