LinuxQuestions.org
Download your favorite Linux distribution at LQ ISO.
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - Distributions > Slackware
User Name
Password
Slackware This Forum is for the discussion of Slackware Linux.

Notices


Reply
  Search this Thread
Old 03-21-2018, 05:36 AM   #61
hydrurga
LQ Guru
 
Registered: Nov 2008
Location: Pictland
Distribution: Linux Mint 19 MATE
Posts: 6,077
Blog Entries: 2

Rep: Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932

Quote:
Originally Posted by h2-1 View Post
hydrurga, sorry, I had to deal with some real bs, couldn't get back until now.

Re dates, pinxi isn't currently touching them, they are strings, and it prints whatever is given it. Dates are kind of a pain, the classic example being: 12/12/12 - where in the world does it come from? is it mm/dd/yy is it dd/mm/yy?

So guessing at string value dates that come from different sources has always been a bit risky. I also prefer the iso standard date format, and use it in my work always, because it's sortable, consistent, and not ambiguous, but it's slightly risky converting dates of unknown date time format to iso dates without knowing the format.

I've honestly never seen GiB for size, and having done tech a long time, that's saying something. I've seen G gB and GB, so I opted for the latter. That's one I think may be subjective.

But man df was helpful here, it looks like you are right:

so yes, I was wrong there, thank you for correcting me. I think I would prefer the less verbose form that df has, KMGTPE over the more verbose: KiB, hmm, though the latter looks nice, and might be educational.

I'm going to do some pinxi fixes then roll those all into inxi 2.9.02 to avoid too many master commits.
I would 100% agree with you - if the date string comes as is, with no indication as to which component is the months or the days (or even the years as you state), then you should leave it well alone. You might want to add that info though to the man file.

Regarding the prefixes, I too thought that e.g. GiB was a bit clumsy when I first started using it. I don't now though - those prefixes provide welcome unambiguity and I'm used to seeing them quite a lot in a computer context. It will take me a long time (if ever) before I start actually saying e.g. "Gibibytes". To me, GiB *is* the proper Gigabyte in a computing sense.

Keep up the great work and thanks for sharing the inxi beta with us here. It is definitely one of my favourite tools for obtaining information from posters we are trying to help here on LQ.
 
Old 03-21-2018, 09:14 AM   #62
rokytnji
LQ Veteran
 
Registered: Mar 2008
Location: Waaaaay out West Texas
Distribution: AntiX 17
Posts: 5,569
Blog Entries: 20

Rep: Reputation: 2650Reputation: 2650Reputation: 2650Reputation: 2650Reputation: 2650Reputation: 2650Reputation: 2650Reputation: 2650Reputation: 2650Reputation: 2650Reputation: 2650
I do not root login into a desktop session. Ever. Those readouts are in root terminal. Thanks for the concern though. I have hands on experience using smxi and sgfxi also. Which won't run in a desktop session. It requires running a root cli session.


All I can say is h2 has seen my posts on MX forum. My install steps on this chromebook are well documented there. My other update steps are well documented here. I have been very good at not holding anything back. Like some folks do on forums.

https://www.linuxquestions.org/quest...7/#post5830970
https://www.linuxquestions.org/quest...7/#post5831237
https://www.linuxquestions.org/quest...7/#post5832442
https://www.linuxquestions.org/quest...ml#post5832765
https://www.linuxquestions.org/quest...ml#post5832831
https://www.linuxquestions.org/quest...ml#post5833277
https://www.linuxquestions.org/quest...ml#post5833286
https://www.linuxquestions.org/quest...ml#post5833301

Plus not just on this forum.
https://forum.mxlinux.org/viewtopic....438686#p438686
https://forum.mxlinux.org/viewtopic....438762#p438762
I am not even ashamed when I make a mistake
https://forum.mxlinux.org/viewtopic....438852#p438852

So if h2 thinks this problem is peculiar to my install. And your concerns about me running my desktop session as a root user. PS. You can see when I type exit in a terminal before closing it. Any mistake I may have made will be in those links I posted.

My post will probably be moot anyways. I can always start over by deleting pinxi using the locate command to find what to delete.

Code:
$ locate pinxi
/home/harry/.local/share/pinxi
/home/harry/.local/share/pinxi/location-main.txt
/home/harry/.local/share/pinxi/weather-32.4331-97.5390.txt
/home/harry/.local/share/pinxi/weather-79772.txt
/home/harry/.local/share/pinxi/weather-pecos-tx.txt
/usr/local/bin/pinxi
Then start over again from scratch. There is not a lot to remove if you notice. I still have inxi which works well for me.
Thanks again for the concern guys. I grok what you mean.

Edit: so as a experiment. I deleted the old install. Here is the reinstall. You can tell me where I messed up I guess.

Code:
harry@biker:~
$ sudo wget -O /usr/local/bin/pinxi https://github.com/smxi/inxi/raw/inxi-perl/pinxi
[sudo] password for harry: 
--2018-03-21 09:25:01--  https://github.com/smxi/inxi/raw/inxi-perl/pinxi
Resolving github.com (github.com)... 192.30.253.112, 192.30.253.113
Connecting to github.com (github.com)|192.30.253.112|:443... connected.
HTTP request sent, awaiting response... 302 Found
Location: https://raw.githubusercontent.com/smxi/inxi/inxi-perl/pinxi [following]
--2018-03-21 09:25:02--  https://raw.githubusercontent.com/smxi/inxi/inxi-perl/pinxi
Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 151.101.184.133
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|151.101.184.133|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 489470 (478K) [text/plain]
Saving to: ‘/usr/local/bin/pinxi’

/usr/local/bin/pinxi                      100%[=====================================================================================>] 478.00K  1.17MB/s    in 0.4s    

2018-03-21 09:25:02 (1.17 MB/s) - ‘/usr/local/bin/pinxi’ saved [489470/489470]

harry@biker:~
$ sudo chmod +x /usr/local/bin/pinxi
harry@biker:~
$ sudo pinxi -U
Starting pinxi self updater.
Using curl as downloader.
Currently running pinxi version number: 2.9.02
Current version patch number: 02-p
Current version release date: 2018-03-20
Updating pinxi in /usr/local/bin using inxi-perl branch as download source...
Successfully updated to inxi-perl branch version: 2.9.02
New inxi-perl branch version patch number: 02-p
New inxi-perl branch version release date: 2018-03-20
To run the new version, just start pinxi again.
----------------------------------------

Starting download of man page file now.
Skipping man download because branch version is being used.
harry@biker:~
$ pinxi -xxx -w
Error 45: Error opening file: /home/harry/.local/share/pinxi/location-main.txt 
Error: Permission denied
harry@biker:~
$
Code:
harry@biker:~
$  sudo pinxi -xxx -w
Weather:   Conditions: 50 F (10 C) - Clear Wind: From the SW at 5 MPH Humidity: 58% 
           Pressure: 30.22 in (1023 mb) Dew Point: 36 F (2 C) Location: Godley, TX, USA Altitude: 282.9 m 
           Time: Wed 21 Mar 2018 09:30:40 AM CDT (America/Chicago) Observation Time: March 21, 9:15 AM CDT
Code:
harry@biker:~
$ cd /home/harry/.local/share/pinxi
harry@biker:~/.local/share/pinxi
$ ls -l
total 8
-rw-r--r-- 1 root root  255 Mar 21 09:30 location-main.txt
-rw-r--r-- 1 root root 1091 Mar 21 09:30 weather-32.4331-97.5390.txt
harry@biker:~/.local/share/pinxi

Last edited by rokytnji; 03-21-2018 at 09:34 AM.
 
Old 03-21-2018, 01:47 PM   #63
h2-1
Member
 
Registered: Mar 2018
Distribution: Debian Testing
Posts: 75

Original Poster
Rep: Reputation: Disabled
kjhambrick, I think it's worth getting used to given that common language is using GB for 1000 and 1024.

rokytnji, I'm well aware of who you are, and have always appreciated your feedback. You misunderstood, I did not suggest you logged into your desktop login as root user, I suggested that because of the fact that when you ran a command that wrote files to the HOME, when you run a session in terminal as root, root was writing to your user /HOME/user, and not /root/, the most likely explanation is/was that the settings for XDG_DATA_DIR are being set in a way that sets both homes to the user home, thus, when you create a file as root in home, naturally user cannot write to it. This is roughly what you said happened, so there's not a lot of things that can cause that event to happen. Thus, when su for example is used to start a root terminal session, you are now root, and the HOME is /root, not /home/user

This is a reality, so to me, the question becomes, first, for my own curiosity, how is this happening, and second, where does it happen, is for example a first time su use of pinxi creating a config/data file/directory in /home/user that is owned by root? You can check this by checking the permissions on the directory itself, and the files in it. I'd like to know how it happened when it's not really supposed to ever be the case that root writes to user config/data files before I look into a possible fix, since I don't actually know or understand what I'd be fixing. Could this be an unknown XDG setting?

I tested this and verified that this is exactly what happens when no XDG data is set.

Obviously people aren't logging into their desktops in most cases as root, which is a discouraged practice, they are becoming root via:
sudo
su
su -

I tested the su, and it correctly writes to /root, if I su to root, it writes to /root, if not, it writes to /home/user. I also tested this for sudo, same thing. To be complete, I tested it for su -, same thing. So the environmental variable $HOME is not doing anything unexpected when run inside of Perl, it's using the correct $HOME, /home/user for user, and /root for su/sudo/su -, as I'd expect.

So basically, you want to see what's in these two values as user and root: $XDG_DATA_HOME $XDG_CONFIG_HOME as both regular user in terminal, and after su to root in terminal. If that's not the answer, then the case becomes much more interesting, and involves a variable I am totally unaware of re what can happen on a system.

Your example is interesting because it's something I had only the faintest awareness was possible, which is why I'm assuming a configuration setting which is setting the XDG variables to both use /home/user, which I believe is well within what you can do with those values, that is, you can set it to use one single directory globally I believe, though I never looked into them because I don't use them, had to learn about them a bit due to someone asking for inxi to use those as first default for the data / config paths.

When pinxi/inxi starts, it queries first the XDG variables to locate config and data, that's the 'new standard', which I don't see used much so far.

Last edited by h2-1; 03-21-2018 at 01:57 PM.
 
Old 03-21-2018, 02:56 PM   #64
hydrurga
LQ Guru
 
Registered: Nov 2008
Location: Pictland
Distribution: Linux Mint 19 MATE
Posts: 6,077
Blog Entries: 2

Rep: Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932
I've been working through the separate flags, comparing the output from my present (old!) 2.2.35 version with 2.9.02. Please find any differences that I've noted (only where I am not sure why there is a difference, where I have a comment to make, or it concerns an issue that wasn't covered in my previous post).

Of course, some of the differences may well be due to changes that have been made since 2.2.35, so please don't feel any necessity to comment on any of the following notes and differences unless there's something that you want to chase up. If this is of any use to you, then I will continue with the remaining flags.

-C
(old) Dual core Intel Core i5-3210M (-HT-MCP-)
(new) type: MT MCP
[changed from HT to MT]

(old) 1: 2494 MHz 2: 2494 MHz 3: 2494 MHz 4: 2494 MHz
(new) Core speeds: 1: 1198 2: 1198 3: 1197 4: 1197
[different values; perhaps add MHz?]

-G
(old) GLX Version: 3.0 Mesa 17.2.8
(new) version: 4.2 Mesa 17.2.8
[glxinfo shows OpenGL version string: 3.0 Mesa 17.2.8; OpenGL core profile version string: 4.2 (Core Profile) Mesa 17.2.8]

-o
Just a note that inxi lists a 400GiB encrypted partition of mine to be unmounted, although technically it has actually been mounted with VeraCrypt. However, I am not sure you would like to enter the realms of taking into account such mounts.

You might consider ordering unmounted drives by /dev device name.

inxi states "fs: root required" but this info is often obtainable using blkid without being root.
 
Old 03-21-2018, 03:56 PM   #65
h2-1
Member
 
Registered: Mar 2018
Distribution: Debian Testing
Posts: 75

Original Poster
Rep: Reputation: Disabled
hydrurga, good stuff. Unfortunately blkid on Debian shows nothing. However, looking at man blkid, it specifically suggest I not use it, and instead use lsblk -fs, and that provides PERFECT output, labels, uuid, and fs. So looks like lsblk -fs is the way to go.

I thought I was ordering by device name in unmounted, yes, it is ordered, there's a sort there. What do you see actly that does not appear to be sorting? It's sorting by the part after /dev/ that's just tacked back on for dev devices at output.

Code:
@unmounted = sort { $a->{'dev-base'} cmp $b->{'dev-base'} } @unmounted;
sorts on my systems.

I remember from a long time ago that some distros would give blkid output, and some wouldn't, I can't remember the details, but that was one reason I never used blkid in inxi. However, given their man several times says, don't use blkid, use lsblk, I'd say that achieves the purpose.

Great suggestion, I don't usually revisit old logic or assumptions in inxi since there's so much there, and I dislike ever having to use root for anything, so if it can be removed, all the better. I've always disliked using file, it isn't reliable, it can hang for a LONG time on some devices, so anything that is better is good.

Look for improvements there soon.

Last edited by h2-1; 03-21-2018 at 03:57 PM.
 
Old 03-21-2018, 04:01 PM   #66
hydrurga
LQ Guru
 
Registered: Nov 2008
Location: Pictland
Distribution: Linux Mint 19 MATE
Posts: 6,077
Blog Entries: 2

Rep: Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932
Nice one. Regarding the sort, it is currently a character sort, not a numerical value sort, so that my three unmounted partitions are listed in the order:

/dev/sda1
/dev/sda14
/dev/sda2
 
Old 03-21-2018, 05:15 PM   #67
hydrurga
LQ Guru
 
Registered: Nov 2008
Location: Pictland
Distribution: Linux Mint 19 MATE
Posts: 6,077
Blog Entries: 2

Rep: Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932
Continuing on down the parameters (same versions as my previous post):

-p
It is difficult to determine what order the partitions are being listed in. As only mounted partitions are listed, I would have thought that ordering by the mountpoint path would be the most natural, otherwise the device name in /dev.

The filesystem for my two mounted NTFS partitions is listed as "fuseblk". Very possibly so, but in reality they are NTFS and perhaps should be listed as such given that the ext4 partitions are listed as "ext4".

-t cm10
(old) Memory: MB / % used - Used/Total: 2395.6/5853.8MB - top 10 active
1: mem: 438.93MB (7.4%) command: firefox pid: 23214

(new) Memory MiB/% used - Command - pid - top: 10
1: mem: 419.2 MiB command: firefox pid: 23214

[The memory section says "MiB/% used" but no percentages are given]

-w
I find it quite surprising that, given that most of the world works in degrees Celsius, the temperature is given in Fahrenheit with the Celsius temperature in brackets. To an American I imagine it looks fine, but certainly to me as a Brit, it looks anachronistic (and we're only half-way, stuck in limbo between SI units and "imperial" units).

-W
For interest, I used -W Paris (without a country name). The output was 76 F (25 C) - Thunderstorm Time: Wed 21 Mar 2018 21:50:35 WAT - obviously not the Paris in France. This is such a small part of inxi, I know, and I feel petty for mentioning this, but you would have thought that the major world city names, at least the capitals, would be tied to those cities. Not wanting to annoy Americans any more after my temperature comment, but when I asked for the weather in Athens, you can guess which Athens it gave me, and it wasn't the one in Greece.

-m (as sudo)
Array-1 and Devicex memory is still shown as GB rather than GiB.

I'll take another break there.
 
Old 03-21-2018, 06:29 PM   #68
h2-1
Member
 
Registered: Mar 2018
Distribution: Debian Testing
Posts: 75

Original Poster
Rep: Reputation: Disabled
hydrurga, re the sorts, basically, given how few people this would impact, I'd say it's acceptable. There are so many possible device names out there that to try to make some rule that would make smarter sorts is just not worth it in my opinion, most users will have < 10 partitions on say, sda, so I think covering most users is ok. At this point, I'd remind myself the perfect is the enemy of the quite decent.

I'll get to the other stuff, getting a load of good feedback and suggestions here and on other forums.

re -p fs, pinxi is reporting what it is told. It's much more difficult to engage in depth not trusting of the data, and to try to dig in, so it reports what it has been told re fs and doesn't try to overthink it, since that requires orders of magnitude more code to achieve.

Last edited by h2-1; 03-21-2018 at 06:32 PM.
 
Old 03-21-2018, 06:32 PM   #69
hydrurga
LQ Guru
 
Registered: Nov 2008
Location: Pictland
Distribution: Linux Mint 19 MATE
Posts: 6,077
Blog Entries: 2

Rep: Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932
Continuing on...

-xB
model: COMPAL PABAS0241231
[There are two spaces in the middle of the output string. Of course, this might be a fixed width string.]

-wx
Please see previous note on SI vs. imperial measurements, in this case KPH v MPH and mb vs in. Except of course that mb isn't an SI unit, although widely accepted for meteorological uses.

-xxxA
[A should only be able to accept -x and -xx, but the above option still works. There are probably more like this.]

-V
I'm running the beta from a directory on my Desktop directory using e.g. ./inxi -V

Program Location: /home/hydrurga/Desktop/inxi-master/./
[the last /./ is superfluous]

Website: https://github.com/smxi/inxi or http://smxi.org/
[Given that http://smxi.org/ redirects to https://smxi.org, the latter might as well be used]

Forums: http://techpatterns.com/forums/forum-33.html
[as above (redirection is to https form)]

There is no redirection by gnu.org, but equally their website is available on https.

Generally, I don't understand why some of the -x, -xx and -xxx options exist. Sometimes they add only one or two data to the output. Is it the case for some that they are only designated as -x[etc.] data because they might not be available on most machines?

Another break...

Last edited by hydrurga; 03-21-2018 at 06:33 PM.
 
Old 03-21-2018, 06:38 PM   #70
h2-1
Member
 
Registered: Mar 2018
Distribution: Debian Testing
Posts: 75

Original Poster
Rep: Reputation: Disabled
just a few quick ones. -x has nothing to do with -A, -x means, show extra data level 1, -xx means show extra data level 2, and -xxx means show extra data level 3. Or, in long form, --extra 3

These extra switches have zero connection with the actual options, nor should they. what they do is trigger internal tests on data generation and output. the tests are integer based, that is, if $extra > 0, do this. if $extra > 1, do this.

So there's no connection at all between -A and -x. You can type if you like, -xxxxxxxx which will set $extra to 8, but of course, there are only currently 3 levels used in data generation and output. That would be absurdly code intensive to try to maintain, and time intensive, the user is free to type in as many or as few x as they way, and the programming happily does its > comparisons, and the stuff all 'just works', that was actually a messy method I cleaned up totally from binxi/bash inxi, where I used explicit booleans to trip levels 1, 2, and 3. Those were a pain to type, ugly to look at in the code, so when I did pinxi, I decided to use integers, and allow 0 to anything counts. Since the tests internally simply test the value of the integer, it doesn't care how high or low it is, unless it's below the test value, or above it.

Re imperial measurements, this is actually something that -w/-W supports internally but it's not currently implemented, time is a factor, all this takes time. Some stuff I left til later since the goal of release is to hit 3.0 with all fixes that seem to be best for pre 3.0. Since I'm using 00 counter, that means there's now 96 possible releases to go before 3.0.

I'm also tracking these issues here: https://github.com/smxi/inxi/issues/134

and adding stuff that I will forget, which is all of it if I don't note it there as I go along.

Last edited by h2-1; 03-21-2018 at 06:41 PM.
 
Old 03-21-2018, 06:42 PM   #71
hydrurga
LQ Guru
 
Registered: Nov 2008
Location: Pictland
Distribution: Linux Mint 19 MATE
Posts: 6,077
Blog Entries: 2

Rep: Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932
Quote:
Originally Posted by h2-1 View Post
hydrurga, re the sorts, basically, given how few people this would impact, I'd say it's acceptable. There are so many possible device names out there that to try to make some rule that would make smarter sorts is just not worth it in my opinion, most users will have < 10 partitions on say, sda, so I think covering most users is ok. At this point, I'd remind myself the perfect is the enemy of the quite decent.

I'll get to the other stuff, getting a load of good feedback and suggestions here and on other forums.

re -p fs, pinxi is reporting what it is told. It's much more difficult to engage in depth not trusting of the data, and to try to dig in, so it reports what it has been told re fs and doesn't try to overthink it, since that requires orders of magnitude more code to achieve.
Fair enough. I've always been an outlier and my system with its 17 partitions obviously is too.

There is no hurry whatsoever. My feedback is just there for you to ignore or use as you see fit, in your own time.

Re: -p. Perhaps the filesystem info should be sourced from elsewhere then? After all, the partition's filesystem info can be obtained with lsblk -fs. In this case, I would think that the fact that a partition is ntfs is more useful information than it is fuseblk (a generic term that may not mean anything to those less-experienced users you talked about before).
 
Old 03-21-2018, 06:44 PM   #72
hydrurga
LQ Guru
 
Registered: Nov 2008
Location: Pictland
Distribution: Linux Mint 19 MATE
Posts: 6,077
Blog Entries: 2

Rep: Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932Reputation: 1932
Quote:
Originally Posted by h2-1 View Post
just a few quick ones. -x has nothing to do with -A, -x means, show extra data level 1, -xx means show extra data level 2, and -xxx means show extra data level 3. Or, in long form, --extra 3

These extra switches have zero connection with the actual options, nor should they. what they do is trigger internal tests on data generation and output. the tests are integer based, that is, if $extra > 0, do this. if $extra > 1, do this.

So there's no connection at all between -A and -x. You can type if you like, -xxxxxxxx which will set $extra to 8, but of course, there are only currently 3 levels used in data generation and output. That would be absurdly code intensive to try to maintain, and time intensive, the user is free to type in as many or as few x as they way, and the programming happily does its > comparisons, and the stuff all 'just works', that was actually a messy method I cleaned up totally from binxi/bash inxi, where I used explicit booleans to trip levels 1, 2, and 3. Those were a pain to type, ugly to look at in the code, so when I did pinxi, I decided to use integers, and allow 0 to anything counts. Since the tests internally simply test the value of the integer, it doesn't care how high or low it is, unless it's below the test value, or above it.

Re imperial measurements, this is actually something that -w/-W supports internally but it's not currently implemented, time is a factor, all this takes time. Some stuff I left til later since the goal of release is to hit 3.0 with all fixes that seem to be best for pre 3.0. Since I'm using 00 counter, that means there's now 96 possible releases to go before 3.0.

I'm also tracking these issues here: https://github.com/smxi/inxi/issues/134

and adding stuff that I will forget, which is all of it if I don't note it there as I go along.
Thumbs up. You're doing a great job.
 
Old 03-21-2018, 06:50 PM   #73
h2-1
Member
 
Registered: Mar 2018
Distribution: Debian Testing
Posts: 75

Original Poster
Rep: Reputation: Disabled
A bit more, re -x, the reason -x options exist is because inxi was developed with terseness in mind.

That's why default -b is very short. It was meant for irc use.

If you resist the urge to overthink the -x stuff, it's actually very simple.

-x means extra. The extra values are listed in man and --help.

the first level of -x supplies extra data that many users would find useful.
the second level shows data some users might find useful
the 3rd level, when I did it right, shows data that only specialists would find useful, or that is such garbage quality that whiie inxi can get it, it considers it junk or so arcane that only the most insistent of users would find it of value.

Since the more data you confront the human brain with, the more likely it is to go into max out overload state, at which point the mental processes simply cease to see what is in front of it any more, and to blur out reality.

To see this do:
inxi
inxi -b
inxi -F
inxi -v 1
inxi -v 2
inxi -v 3
inxi -v 4
inxi -v 5
inxi -v 6
inxi -v 7
inxi -v 8

One of the biggest uses inxi has is forum support, so what I see happen is that the various fora figure out what they want to see, and then list that command on their how to post issues page. Common is -Fzx or -zF, for example.

That shows roughly what most people need to see about a system, but doesn't show stuff people don't like to see, like busids and chip ids because those are only interesting if you are looking more deeply into a hardware issue.

Remember, inxi targets a very wide user base, from very high level heavy duty sys admins to new users asking for help on Mint forums, so the options allow the users to create roughly the output they want to see.

I've seen some forums request the by line data for example, using the individual line flags.

For me, except for now testing inxi, where I use -v 8 because it prints everything, and will expose the bugs, and inxi, because it prints almost nothing, and uses some different methods for shortening its output, and -b, which also uses different methods on some lines, to keep them short, what I normally use is -bxxx because that shows me everything I usually want to see on hardware except sensors, and if I want sensors, I use -Fxxx. But that's just me, others only want simpler forms.

I hope this roughly answers it, but the bottom line is, inxi was designed from its first day for user support, so the switches reflected the levels of information support people would want to see, which is up to them.

Last edited by h2-1; 03-21-2018 at 06:53 PM.
 
1 members found this post helpful.
Old 03-21-2018, 06:56 PM   #74
h2-1
Member
 
Registered: Mar 2018
Distribution: Debian Testing
Posts: 75

Original Poster
Rep: Reputation: Disabled
hydrurga, my only surprise in this process has been the totally amazing and great feedback I'm getting, so many great ideas, observations, etc, so I don't want to in anyway discourage any closer inspection, which are finding things I did not see all along, finding new things inxi/pinxi can show, filling out missing stuff, like lsblk -fs soon to be implemented, next patch release, but it takes time, all of it.
 
Old 03-21-2018, 06:59 PM   #75
h2-1
Member
 
Registered: Mar 2018
Distribution: Debian Testing
Posts: 75

Original Poster
Rep: Reputation: Disabled
"I've always been an outlier"

that's what makes your feedback so especially valuable. I've been so fortunate so far to have access to several major outlier users, one tested the stuff on some major hardcore server setups, the kind we can only dream of even seeing, let alone run, so every outlier data exposes bad assumption I made, or failed to make at all, and each one is helping inxi get better.

Don't get me wrong, when I see /dev/sda14 my hacker innards squirm and want to fix it, lol

Actually, I'll go back later and update that issue thread on github with a lot of these.

Again, anything you see, I welcome, even if I get frustrated, I want inxi to support the outliers because that means it's going to be rock solid for the core.
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off



Similar Threads
Thread Thread Starter Forum Replies Last Post
Firefox: are we its informal beta-testers? FeyFre Slackware 29 11-21-2011 02:15 AM
Beta Testers Needed msound General 18 07-28-2006 10:22 AM
looking for beta testers grizzly General 5 03-20-2004 11:24 AM

LinuxQuestions.org > Forums > Linux Forums > Linux - Distributions > Slackware

All times are GMT -5. The time now is 07:58 AM.

Main Menu
Advertisement
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration