[X] My Assistant

Hurray, the web just got a little worse again.

(click for full size)
[files.catbox.moe] (IMG:[files.catbox.moe] https://files.catbox.moe/h00eo8.png)

what did google groups offer exactly? never quite got into it

Google is just ramping up their information control for the election.

what did google groups offer exactly? never quite got into it

Access to some of usenet for free.

Shell script to take a link to a google drive 'preview' page and download the file from it; it also bypasses the annoying "too big to be checked by antivirus scan" page.

It might fail/mess up with small files that don't trigger the antivirus warning page; I've not found a small one to check this on yet. Please let me know if anyone tries this and finds such a problem.
It only will work properly on those small files if crafting a URL properly will force a download page like that to display from which I can fetch a UUID. So likely not.

CODE

#! /usr/bin/env sh
# when given a link to the 'antivirus scan can't be done' page, this downloads
# the file from the page. It also tries to download from preview pages,
# although this may or may not work properly with files that are small
# enough for scanning. Untested in this case.
# Accepts multiple URL's if desired.

# mktemp and wget are the only non-posix things here, I think.
# should be relatively easy to substitute curl or something for wget, and I
# think mktemp could be replaced by some static path if you don't think it'll
# run in parallel. Or make your own random file generation script/program.

#https://drive.usercontent.google.com/download
#?id=[redacted]
#&export=download
#&authuser=0
#&confirm=t
#&uuid=[redacted]
#&at=[redacted]

# 'at' parameter ('access token?') seems to be absent unless all cookies are
# correctly fed in, but does not seem to be necessary for a DL to work.

getkeyval() {
  sed 's/^.*\"'"$1"'\" value=\"//;s/\".*//'
  echo
}

cleanupfunc() # catch ^C and similar and delete temp files before exiting
{
  if [ -e "$TEMPFILE" ]; then
    rm -f "$TEMPFILE"
  fi
  exit 0
}
trap cleanupfunc HUP INT QUIT ABRT TERM
# remove trap for ^C and other signals that would kill the program
trap - HUP INT QUIT ABRT TERM
# run clean up (delete temp files)

TEMPFILE="$(mktemp -t 'urlgen_gdrive_XXXXXX.png')"
# echo "temp file ""$TEMPFILE"
# takes one or more 'virus scan warning' pages/preview pages and downloads
while [ "$#" -gt 0 ]; do

  echo "$1" | grep -q drive.google.com/file/d/
  if [ "$?" -eq 0 ]; then
    # get av warning link from main dl/preview page url
    driveid="$(echo "$1" | sed 's!^.*drive.google.com/file/d/!!;s!/.*!!')"
    avurl='https://drive.usercontent.google.com/download?id='"$driveid"'&export=download&authuser=0'
  else
    # we were given an "AV check warning" url directly
    avurl="$1"
  fi

  wget --quiet --user-agent='Mozilla/5.0 (X11; Linux x86_64; rv:116.0) Gecko/20100101 Firefox/116.0' "$avurl" -O "$TEMPFILE"
  driveid='?id='"$(getkeyval "id" < "$TEMPFILE")"
  driveexport='&export='"$(getkeyval "export" < "$TEMPFILE")"
  driveauthuser='&authuser='"$(getkeyval "authuser" < "$TEMPFILE")"
  driveconfirm='&confirm='"$(getkeyval "confirm" < "$TEMPFILE")"
  driveuuid='&uuid='"$(getkeyval "uuid" < "$TEMPFILE")"
  #  driveat='&at='"$(getkeyval "at" < "$TEMPFILE")"
  dlurl='https://drive.usercontent.google.com/download'"$driveid""$driveexport""$driveauthuser""$driveconfirm""$driveuuid"

  echo "$dlurl"
  wget --content-disposition --user-agent='Mozilla/5.0 (X11; Linux x86_64; rv:116.0) Gecko/20100101 Firefox/116.0' "$dlurl"
  shift 1
done

cleanupfunc

This post has been edited by dragontamer8740: Jan 20 2024, 11:29

Can a DC 20v 7.5amp power supply work for a 19v 7.8amp device?

The volts are within 10%, if the amps are higher it would not be an issue as the device would just draw what it needs but with the volts higher it can be an issue depending on how well the device is designed?

Can a DC 20v 7.5amp power supply work for a 19v 7.8amp device?

The volts are within 10%, if the amps are higher it would not be an issue as the device would just draw what it needs but with the volts higher it can be an issue depending on how well the device is designed?

assuming its some device that is just going to use it for charging and expects some brick power supply, probably fine 9/10 times. voltages are regulated way better with cheap ics so internal voltage wont be wrong any time soon. and on the power supply side things are always some degree of overrated for the application. I've thrown a lot of things on the 24v rail in cabinets anywhere from 20 to 12v original power supply rated, with minimal issues.
regardless of the extremes, most cheap power supplies are easily a volt off in certain scenarios.

without any warranties or liability implied, I'd say probably a solid 95% chance it'll be fine

Agreed with cate-chan.

Can a DC 20v 7.5amp power supply work for a 19v 7.8amp device?

The volts are within 10%, if the amps are higher it would not be an issue as the device would just draw what it needs but with the volts higher it can be an issue depending on how well the device is designed?

Probably. I'd say it's within the margin of error for basically anything.

The voltage will be fine; the current may be an issue if you're planning to run stuff at max load on it but I wouldn't expect it to matter.

In a world where things like capacitors frequently have a +-20% tolerance range, it's likely fine.

Usually stuff is made with a little overhead for unexpected draw, too.

If you look at it in terms of heat dissipation, you're looking at 150 watts vs. 148.2 watts. I don't think 1.8 watts of difference is going to be the difference between failure and proper operation.

Most stuff has internal voltage regulators if it's really that mission-critical, which work to normalize the power they're being fed. All that might happen with most designs is that those may run a little warmer.

I've powered an NES off of 14V DC (it asks for 9V) before. Stuff with 7805 regulators can handle a lot as long as they can dissipate the heat. Of course, if this is remotely modern (I'm guessing a laptop?), it will use some switching regulator instead. Even there, though, I've run my ipod dock off more than the 12V it wants for a year or two. And that certainly is not using a linear regulator (massively inefficient; I'd feel the heat if it were).

I also have used 3.8V batteries to replace 3.7V ones in game boys and they've lasted years like that.

==

edit: one of my thinkpads (the T510 I got used a couple months ago for about $80) seems to have a problem with its ethernet NIC. It's maxing out at just a bit under 100Mbits/sec instead of closer to 1G over my LAN. I already tried using different cables and I checked the pins on the ethernet connector. They looked fine. I even tried using the ethernet port on the docking station (uses the same NIC, I think, but via a differently positioned connector) and it also had trouble. So I just bought an Expresscard gigabit ethernet controller; hopefully that'll actually fix it. Maybe in the future if I get a hot air station I can try to fix the NIC itself.

Funnily, the comparatively enormous 15" T510 only has a 34mm expresscard slot, where my 12" X201 from the same year with the same processor has a 54mm one. Guessing the X201 only kept a 54mm slot in that year because it shared a chassis almost entirely with the earlier X200 model, but the T510 did not share as much with the T500.

This post has been edited by dragontamer8740: Jan 27 2024, 16:18

Everyone should probably update OpenSSH to the 9.6 series due to a vulnerability that was found a little while ago.

I just finally got around to it today, and noticed that X11 forwarding windows were taking a lot longer to start.

There's a setting in 9.6 called 'ObscureKeystrokeTiming' that accidentally messed with X forwarding as well.

Either set it to 'no' in your ssh_config file or on command line (ssh -o ObscureKeystrokeTiming=no) to make it a lot snappier.

This post has been edited by dragontamer8740: Jan 28 2024, 20:52

assuming its some device that is just going to use it for charging and expects some brick power supply, probably fine 9/10 times. voltages are regulated way better with cheap ics so internal voltage wont be wrong any time soon. and on the power supply side things are always some degree of overrated for the application. I've thrown a lot of things on the 24v rail in cabinets anywhere from 20 to 12v original power supply rated, with minimal issues.
regardless of the extremes, most cheap power supplies are easily a volt off in certain scenarios.

without any warranties or liability implied, I'd say probably a solid 95% chance it'll be fine

I have done it will beefier hardware, just never a laptop before. Though when people test laptop power supplies they seem to have large (5-10%) variance anyways.

Agreed with cate-chan.
Probably. I'd say it's within the margin of error for basically anything.

The voltage will be fine; the current may be an issue if you're planning to run stuff at max load on it but I wouldn't expect it to matter.

In a world where things like capacitors frequently have a +-20% tolerance range, it's likely fine.

Usually stuff is made with a little overhead for unexpected draw, too.

The power supply I was looking at is advertised at 20v but was tested as 19.5 volts.

oh wow they actually fixed the fanbox importer. guess I was too hasty to call kemono party dead

I want to have as many screens as I wish.
Tech guys please update the hardware and AR/VR technology
can't wait to see my dream come true (IMG:[invalid] style_emoticons/default/smile.gif)

I have done it will beefier hardware, just never a laptop before. Though when people test laptop power supplies they seem to have large (5-10%) variance anyways.
The power supply I was looking at is advertised at 20v but was tested as 19.5 volts.

Your laptop will be just fine, if that's what it is.

Everyone should probably update OpenSSH to the 9.6 series due to a vulnerability that was found a little while ago.

I just finally got around to it today, and noticed that X11 forwarding windows were taking a lot longer to start.

There's a setting in 9.6 called 'ObscureKeystrokeTiming' that accidentally messed with X forwarding as well.

Either set it to 'no' in your ssh_config file or on command line (ssh -o ObscureKeystrokeTiming=no) to make it a lot snappier.

its only a problem if you have X11 forwarding enabled?

I don't even know where all my linux machines are, let alone updating them. I somehow manage to break stuff when I try to update them.

its only a problem if you have X11 forwarding enabled?

No, but it's only significantly noticeable when it's enabled.

Made a typo while typing Google.com in the URL bar. Ended up loading a site called Goole.com instead, complete with a search bar (by Mojeek search engine), and an ad. It's logo is probably a reference to the real life English town, Goole.

Out of the loop ever since I moved to OnlyOffice for my personal needs. Saw that LibreOffice jumped to v24. Last I remembered it was still v7 last year.

Edit:

Just seen this. Make sense.

Starting from 2024, TDF will adopt calendar based-release numbering, so the next major release will be LibreOffice 24.2 in February 2024.

This post has been edited by elda88: Feb 3 2024, 07:00

SSDs have been steadily regaining price after the catastrophic (for the manufacturers) price drop in 2023. I'm not in real need of SSDs right now, would rather get more HDDs but well, I knew the trend couldn't have kept going forever.

The .NET portability we were promised ~10 years ago is going just fine.
[postimg.cc] (IMG:[i.postimg.cc] https://i.postimg.cc/rFX3JWTZ/what-an-improvement.png)
Ah, yes. Of course. That windows 10 VM that I obviously already have. Because everyone has one of those, right?
What a worthwhile change and improvement.
[postimg.cc] (IMG:[i.postimg.cc] https://i.postimg.cc/TYRtn0F0/do-you-guys-not-have-phones.png)

Just when I'd thought they couldn't be any more of an insufferable cunt. This program stopped being buildable in Mono years ago already, so I'm not entirely surprised. And when this was reported they claimed that it was because they'd 'moved to .NET core,' despite using Winforms and .NET core supposedly being the true answer to making portable C# programs.

At this point I'd like to find an "editor" for Pokémon saves that just exports data structures it knows about to an XML file or similar.

With how little I use such save editors though, it might be best to just figure out the addresses for different things from one of the game decompilation efforts and use a hex editor for stuff like fixing the realtime clock setting in the GBA games.

This post has been edited by dragontamer8740: Feb 6 2024, 17:12

At this point I'd like to find an "editor" for Pokémon saves that just exports data structures it knows about to an XML file or similar.

With how little I use such save editors though, it might be best to just figure out the addresses for different things from one of the game decompilation efforts and use a hex editor for stuff like fixing the realtime clock setting in the GBA games.

if its opensource and makes limited use of all the c# windowsisms, the majority of the non ui code should be able to smashed into actaully portable dotnet core code. but sure might be fun/good to just start over with something need specific

This post has been edited by cate_chan: Feb 6 2024, 19:46