Summary: ls [A-B]* I expect this to give me all files starting with either
A or B. Instead I get all files starting with A,B,a,b

The bash expansion seems to ignore capitals.
(similary for ls [a-b]* )

How do I get bash to respect the case in its expansion?

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On Tue, 23 Jun 2020 21:35:54 -0400, William Unruh <unruh@invalid.ca> wrote:

The bash expansion seems to ignore capitals.
(similary for ls [a-b]* )
How do I get bash to respect the case in its expansion?

LC_ALL=C ls [a-b]*
That's from the Pattern Matching section of man bash.
Note that if a directory starts with a or b, it's full contents will be listed.

Regards, Dave Hodgins


--
Change dwhodgins@nomail.afraid.org to davidwhodgins@teksavvy.com for
email replies.

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On Wed, 24 Jun 2020 01:35:54 -0000 (UTC), William Unruh wrote:

The bash expansion seems to ignore capitals.
(similary for ls [a-b]* )

How do I get bash to respect the case in its expansion?

I would try [a-zA-Z]


--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On 2020-06-24, David W. Hodgins <dwhodgins@nomail.afraid.org> wrote:

On Tue, 23 Jun 2020 21:35:54 -0400, William Unruh <unruh@invalid.ca> wrote:

The bash expansion seems to ignore capitals.
(similary for ls [a-b]* )
How do I get bash to respect the case in its expansion?

LC_ALL=C ls [a-b]*
That's from the Pattern Matching section of man bash.
Note that if a directory starts with a or b, it's full contents will be

listed.

Thanks.
My LC_ALL is en_CA.UTF-8. But how in the world does
en_CA.UTF-8 make a, b occur between A and B? Certainly it distingushes
between a and A and ls [a]* -d does not list files starting with A.

And how do , say the French, list files which say start with or contain an accented
character
Weird.

Regards, Dave Hodgins

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On 24.06.2020 at 15:48, William Unruh scribbled:

On 2020-06-24, David W. Hodgins <dwhodgins@nomail.afraid.org> wrote:

On Tue, 23 Jun 2020 21:35:54 -0400, William Unruh
<unruh@invalid.ca> wrote:=20

The bash expansion seems to ignore capitals.
(similary for ls [a-b]* )
How do I get bash to respect the case in its expansion? =20

LC_ALL=3DC ls [a-b]*
That's from the Pattern Matching section of man bash.
Note that if a directory starts with a or b, it's full contents
will be listed. =20

=20
Thanks.
My LC_ALL is en_CA.UTF-8. But how in the world does
en_CA.UTF-8 make a, b occur between A and B? Certainly it distingushes between a and A and ls [a]* -d does not list files starting with A.

It's actually LC_COLLATE which does that. Insofar I know, every locale
mixes upper- and lowercase for sorting, and only LC_COLLATE=3DC sorts
uppercase first and lowercase after that =E2=80=94 or vice versa; I don't remember.

By the way, LC_ALL, when set, overrides all other LC_* variables. So
for instance, if you have LC_TIME set to "en_CA.UTF-8" and LC_ALL to
"C", then all of your locale settings are effectively set to "C", even
though the variables themselves are not altered.


--=20
With respect,
=3D Aragorn =3D


--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Strider (2:250/1@fidonet)

On 2020-06-24, Aragorn <thorongil@telenet.be> wrote:

On 24.06.2020 at 15:48, William Unruh scribbled:

On 2020-06-24, David W. Hodgins <dwhodgins@nomail.afraid.org> wrote:

On Tue, 23 Jun 2020 21:35:54 -0400, William Unruh
<unruh@invalid.ca> wrote:

The bash expansion seems to ignore capitals.
(similary for ls [a-b]* )
How do I get bash to respect the case in its expansion?

LC_ALL=C ls [a-b]*
That's from the Pattern Matching section of man bash.
Note that if a directory starts with a or b, it's full contents
will be listed.

Thanks.
My LC_ALL is en_CA.UTF-8. But how in the world does
en_CA.UTF-8 make a, b occur between A and B? Certainly it distingushes
between a and A and ls [a]* -d does not list files starting with A.

It's actually LC_COLLATE which does that. Insofar I know, every locale
mixes upper- and lowercase for sorting, and only LC_COLLATE=C sorts
uppercase first and lowercase after that — or vice versa; I don't
remember.

But this is different. It is not only sorting them mixed up but is also
finding them is caseless. [a]* finds only files that start with a, but
[A-B]* finds files which start with A, a, B, b. That is independent of
sorting.

By the way, LC_ALL, when set, overrides all other LC_* variables. So
for instance, if you have LC_TIME set to "en_CA.UTF-8" and LC_ALL to
"C", then all of your locale settings are effectively set to "C", even
though the variables themselves are not altered.

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On Wed, 24 Jun 2020 19:05:48 -0400, William Unruh <unruh@invalid.ca> wrote:

But this is different. It is not only sorting them mixed up but is also finding them is caseless. [a]* finds only files that start with a, but [A-B]* finds files which start with A, a, B, b. That is independent of sorting.

https://en.wikipedia.org/wiki/Unicode_collation_algorithm
It ignores case and accents when sorting so that a, A, à, À, etc are all treated
as being the same.

When you have [a]*, the square brackets are ignored since it isn't a range being selected.

Regards, Dave Hodgins

--
Change dwhodgins@nomail.afraid.org to davidwhodgins@teksavvy.com for
email replies.

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On 2020-06-25, David W. Hodgins <dwhodgins@nomail.afraid.org> wrote:

On Wed, 24 Jun 2020 19:05:48 -0400, William Unruh <unruh@invalid.ca> wrote:

But this is different. It is not only sorting them mixed up but is also
finding them is caseless. [a]* finds only files that start with a, but
[A-B]* finds files which start with A, a, B, b. That is independent of
sorting.

https://en.wikipedia.org/wiki/Unicode_collation_algorithm
It ignores case and accents when sorting so that a, A, à, À, etc are all

treated

as being the same.

But as I said, it is not sorting. It is selecting.
Since a* is different from A* the selection algorithm clearly treats A
and a differently. Also, doing
LC_COLLATE=C ls [a-b]* still gives me all cases, while LC_ALL=c ls [a-b]*
does not. Thus it must be some other LC that determines it.

When you have [a]*, the square brackets are ignored since it isn't a range being selected.

Regards, Dave Hodgins

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On Thu, 25 Jun 2020 05:06:17 -0000 (UTC), William Unruh wrote:

But as I said, it is not sorting. It is selecting.
Since a* is different from A* the selection algorithm clearly treats A
and a differently.

Sound about right. [a-z] selects only lowercase.

See up thread for my suggestion.

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On 2020-06-25, Bit Twister <BitTwister@mouse-potato.com> wrote:

On Thu, 25 Jun 2020 05:06:17 -0000 (UTC), William Unruh wrote:

But as I said, it is not sorting. It is selecting.
Since a* is different from A* the selection algorithm clearly treats A
and a differently.

Sound about right. [a-z] selects only lowercase.

See up thread for my suggestion.

Nope

ls -d [a-b]*

.......
BillUnruhPortfolioReview23Mar20-altered.pdf
Bill.zip
bin/
binarypulsar-Sci
----

env|grep LC_

LC_ALL=en_CA.UTF-8
LC_SOURCED=1
LC_CTYPE=C

LC_COLLATE=C env|grep LC_

LC_COLLATE=C
LC_ALL=en_CA.UTF-8
LC_SOURCED=1
LC_CTYPE=C

LC_COLLATE=C ls -d [a-b]*

.......
BillUnruhPortfolioReview23Mar20-altered.pdf
Bill.zip
bin/
binarypulsar-Sci
----




--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On Thu, 25 Jun 2020 14:32:12 -0000 (UTC), William Unruh wrote:

On 2020-06-25, Bit Twister <BitTwister@mouse-potato.com> wrote:

On Thu, 25 Jun 2020 05:06:17 -0000 (UTC), William Unruh wrote:

But as I said, it is not sorting. It is selecting.
Since a* is different from A* the selection algorithm clearly treats A
and a differently.

Sound about right. [a-z] selects only lowercase.

See up thread for my suggestion.

Nope

ls -d [a-b]*

One more time [a-b] indicates lowercase only.

I can suggest reading the following page very carefully. https://www.regular-expressions.info/posixbrackets.html


--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

Further experiments.

LC_ALL='' LC_COLLATE=C ls -d [a-b]*
still picks out both lower case and capitals, but now orders them with
capitals first. So it is doing what is needed on sorting, but not on
selecting.
What LC_ variable is determining the selection criterion of the bash
globbing?
Is there anywhere that tells one what the various LC_ are? Has anyone
looked at the code for bash to see what it uses? Maybe it uses LC_ALL?


What determines what bash uses to select the case on globbing?


On 2020-06-25, William Unruh <unruh@invalid.ca> wrote:

On 2020-06-25, Bit Twister <BitTwister@mouse-potato.com> wrote:

On Thu, 25 Jun 2020 05:06:17 -0000 (UTC), William Unruh wrote:

But as I said, it is not sorting. It is selecting.
Since a* is different from A* the selection algorithm clearly treats A
and a differently.

Sound about right. [a-z] selects only lowercase.

See up thread for my suggestion.

Nope

ls -d [a-b]*

......
BillUnruhPortfolioReview23Mar20-altered.pdf
Bill.zip
bin/
binarypulsar-Sci
----

env|grep LC_

LC_ALL=en_CA.UTF-8
LC_SOURCED=1
LC_CTYPE=C

LC_COLLATE=C env|grep LC_

LC_COLLATE=C
LC_ALL=en_CA.UTF-8
LC_SOURCED=1
LC_CTYPE=C

LC_COLLATE=C ls -d [a-b]*

......
BillUnruhPortfolioReview23Mar20-altered.pdf
Bill.zip
bin/
binarypulsar-Sci
----

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On 2020-06-25, Bit Twister <BitTwister@mouse-potato.com> wrote:

On Thu, 25 Jun 2020 14:32:12 -0000 (UTC), William Unruh wrote:

On 2020-06-25, Bit Twister <BitTwister@mouse-potato.com> wrote:

On Thu, 25 Jun 2020 05:06:17 -0000 (UTC), William Unruh wrote:

But as I said, it is not sorting. It is selecting.
Since a* is different from A* the selection algorithm clearly treats A >>>> and a differently.

Sound about right. [a-z] selects only lowercase.

See up thread for my suggestion.

Nope

ls -d [a-b]*

One more time [a-b] indicates lowercase only.

One more time. I gave you the evidence. It does not. I have now run the experiment many times. It choses both
lower and upper case if LC_ALL is en_CA.UTF-8

I can suggest reading the following page very carefully. https://www.regular-expressions.info/posixbrackets.html

globbing is not a regular expression.

I do experiments in this case, not theory.
What is your value of LC_ALL?
What do you get if you do
ls [a-b]* -d
?
(assuming that you have a mixture of lower and upper case file names)

Reding the bash man page again, I noticed the globasciiranges option.

Do you have the shell option globasciranges set?
shopt globasciiranges
is that on or off? Mine was off.

I have now put
shopt -s globasciiranges
into .bashrc, and the ranges behave properly, as expected, even with
LC_ALL not being C.




--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On 25.06.2020 at 14:32, William Unruh scribbled:

On 2020-06-25, Bit Twister <BitTwister@mouse-potato.com> wrote:

On Thu, 25 Jun 2020 05:06:17 -0000 (UTC), William Unruh wrote:

But as I said, it is not sorting. It is selecting.
Since a* is different from A* the selection algorithm clearly
treats A and a differently.

Sound about right. [a-z] selects only lowercase.

See up thread for my suggestion.

Nope

ls -d [a-b]*

......
BillUnruhPortfolioReview23Mar20-altered.pdf
Bill.zip
bin/
binarypulsar-Sci
----

env|grep LC_

LC_ALL=en_CA.UTF-8
LC_SOURCED=1
LC_CTYPE=C

LC_COLLATE=C env|grep LC_

LC_COLLATE=C
LC_ALL=en_CA.UTF-8
LC_SOURCED=1
LC_CTYPE=C

LC_COLLATE=C ls -d [a-b]*

......
BillUnruhPortfolioReview23Mar20-altered.pdf
Bill.zip
bin/
binarypulsar-Sci
----

As I wrote already, Bill, LC_ALL overides all other LC_* settings.

You have your LC_ALL set to "en_CA.UTF-8". By consequence, your
"LC_COLLATE=C" has no effect.


--
With respect,
= Aragorn =


--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Strider (2:250/1@fidonet)

While that may be true, I can also do
LC_ALL='' LC_COLLATE=C ls [a-b]* -d
Now those capital B filenames are still selected, but they now are
listed at the beginning of the list of files, instead of interspersed
with the lower case b file.

Ie, the LC_ALL does not determine bash glob per se.
(except of course LC_ALL=C does).
shopt -s globasciiasciiranges
does work independent of the setting of LC_ALL.
LC_COLLATE does not determine the selection process for those ranges.
LC_ALL does, but I cannot see any specific LC_XXX which does control it.
Very strange.



On 2020-06-25, Aragorn <thorongil@telenet.be> wrote:

On 25.06.2020 at 14:32, William Unruh scribbled:

On 2020-06-25, Bit Twister <BitTwister@mouse-potato.com> wrote:

On Thu, 25 Jun 2020 05:06:17 -0000 (UTC), William Unruh wrote:

But as I said, it is not sorting. It is selecting.
Since a* is different from A* the selection algorithm clearly
treats A and a differently.

Sound about right. [a-z] selects only lowercase.

See up thread for my suggestion.

Nope

ls -d [a-b]*

......
BillUnruhPortfolioReview23Mar20-altered.pdf
Bill.zip
bin/
binarypulsar-Sci
----

env|grep LC_

LC_ALL=en_CA.UTF-8
LC_SOURCED=1
LC_CTYPE=C

LC_COLLATE=C env|grep LC_

LC_COLLATE=C
LC_ALL=en_CA.UTF-8
LC_SOURCED=1
LC_CTYPE=C

LC_COLLATE=C ls -d [a-b]*

......
BillUnruhPortfolioReview23Mar20-altered.pdf
Bill.zip
bin/
binarypulsar-Sci
----

As I wrote already, Bill, LC_ALL overides all other LC_* settings.

You have your LC_ALL set to "en_CA.UTF-8". By consequence, your "LC_COLLATE=C" has no effect.

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On Thu, 25 Jun 2020 16:43:23 -0000 (UTC), William Unruh wrote:

On 2020-06-25, Bit Twister <BitTwister@mouse-potato.com> wrote:

On Thu, 25 Jun 2020 14:32:12 -0000 (UTC), William Unruh wrote:

On 2020-06-25, Bit Twister <BitTwister@mouse-potato.com> wrote:

On Thu, 25 Jun 2020 05:06:17 -0000 (UTC), William Unruh wrote:

But as I said, it is not sorting. It is selecting.
Since a* is different from A* the selection algorithm clearly treats A >>>>> and a differently.

Sound about right. [a-z] selects only lowercase.

See up thread for my suggestion.

Nope

ls -d [a-b]*

One more time [a-b] indicates lowercase only.

One more time. I gave you the evidence. It does not.

And yet, it is working as you directed it to do so.
NOT like you wanted. You tell it you want Uppercase and see what happens.

I have now run the experiment many times. It choses both
lower and upper case if LC_ALL is en_CA.UTF-8

I can suggest reading the following page very carefully.
https://www.regular-expressions.info/posixbrackets.html

globbing is not a regular expression.

I do experiments in this case, not theory.

Screw theory. I am trying to get you to read how your brackets are working. Apparently you are not going to do so. :(

What is your value of LC_ALL?

I always try to work with defaults wherever possible.
$ env | grep LC_
LC_SOURCED=1

What do you get if you do
ls [a-b]* -d
?
(assuming that you have a mixture of lower and upper case file names)

As indicated I get any file name with a lower case letter in the a to b range. Adding uppercase A and B to the range gets me the files with uppercase A to B range.

$ ls
a1.a a.a a.A A.a b.b B.B

$ ls [a-b]*
a1.a a.a a.A A.a b.b

$ ls [a-bA-B]*
a1.a a.a a.A A.a b.b B.B

Reding the bash man page again, I noticed the globasciiranges option.

Do you have the shell option globasciranges set?
shopt globasciiranges
is that on or off? Mine was off.

Same here.

I have now put
shopt -s globasciiranges
into .bashrc, and the ranges behave properly, as expected, even with
LC_ALL not being C.

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On 25.06.2020 at 17:22, William Unruh scribbled:

While that may be true, I can also do
LC_ALL='' LC_COLLATE=C ls [a-b]* -d
Now those capital B filenames are still selected, but they now are
listed at the beginning of the list of files, instead of interspersed
with the lower case b file.

Ie, the LC_ALL does not determine bash glob per se.
(except of course LC_ALL=C does).
shopt -s globasciiasciiranges
does work independent of the setting of LC_ALL.
LC_COLLATE does not determine the selection process for those ranges.
LC_ALL does, but I cannot see any specific LC_XXX which does control
it. Very strange.

But, you forgot the number-one rule... The shell always globs BEFORE
processing anything else. So the "*" in your command was already
globbed BEFORE the "[a-b]" was.


--
With respect,
= Aragorn =


--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Strider (2:250/1@fidonet)

On 2020-06-25, Aragorn <thorongil@telenet.be> wrote:

On 25.06.2020 at 17:22, William Unruh scribbled:

While that may be true, I can also do
LC_ALL='' LC_COLLATE=C ls [a-b]* -d
Now those capital B filenames are still selected, but they now are
listed at the beginning of the list of files, instead of interspersed
with the lower case b file.

Ie, the LC_ALL does not determine bash glob per se.
(except of course LC_ALL=C does).
shopt -s globasciiasciiranges
does work independent of the setting of LC_ALL.
LC_COLLATE does not determine the selection process for those ranges.
LC_ALL does, but I cannot see any specific LC_XXX which does control
it. Very strange.

But, you forgot the number-one rule... The shell always globs BEFORE processing anything else. So the "*" in your command was already
globbed BEFORE the "[a-b]" was.

Which is why I said bash was doing it. [a-b] is also glob (and if * were
done before [a-b] then it would have no idea what to grab, because that
is the first character). It is the shell that globs the [a-b] as well,
as
echo [a-b]*
shows




--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On 2020-06-25, Bit Twister <BitTwister@mouse-potato.com> wrote:

On Thu, 25 Jun 2020 16:43:23 -0000 (UTC), William Unruh wrote:

On 2020-06-25, Bit Twister <BitTwister@mouse-potato.com> wrote:

On Thu, 25 Jun 2020 14:32:12 -0000 (UTC), William Unruh wrote:

On 2020-06-25, Bit Twister <BitTwister@mouse-potato.com> wrote:

On Thu, 25 Jun 2020 05:06:17 -0000 (UTC), William Unruh wrote:

But as I said, it is not sorting. It is selecting.
Since a* is different from A* the selection algorithm clearly treats A >>>>>> and a differently.

Sound about right. [a-z] selects only lowercase.

See up thread for my suggestion.

Nope

ls -d [a-b]*

One more time [a-b] indicates lowercase only.

One more time. I gave you the evidence. It does not.

And yet, it is working as you directed it to do so.
NOT like you wanted. You tell it you want Uppercase and see what happens.

I have now run the experiment many times. It choses both
lower and upper case if LC_ALL is en_CA.UTF-8

I can suggest reading the following page very carefully.
https://www.regular-expressions.info/posixbrackets.html

globbing is not a regular expression.

I do experiments in this case, not theory.

Screw theory. I am trying to get you to read how your brackets are working. Apparently you are not going to do so. :(

What is your value of LC_ALL?

I always try to work with defaults wherever possible.
$ env | grep LC_
LC_SOURCED=1

What do you get if you do
ls [a-b]* -d
?
(assuming that you have a mixture of lower and upper case file names)

As indicated I get any file name with a lower case letter in the a to b

range.

Adding uppercase A and B to the range gets me the files with uppercase A to

B range.

I do not. As I indicated I get the same as if I did [abB]*
If I do LC_ALL=C I get what you get. But then since so many things send
out utf these days it is nice to see them instead of splodges on the
page.

However on another machine I have
env|grep LC_
LC_SOURCED=1
LC_CTYPE=C

and I get the equiv to
ls -d [aBb]
If I unset LC_CTYPE
I still get the same.

So I am NOT getting either what I asked for nor am I getting anything consistant.

$ ls
a1.a a.a a.A A.a b.b B.B

$ ls [a-b]*
a1.a a.a a.A A.a b.b

$ ls [a-bA-B]*
a1.a a.a a.A A.a b.b B.B

Reding the bash man page again, I noticed the globasciiranges option.

Do you have the shell option globasciranges set?
shopt globasciiranges
is that on or off? Mine was off.

Same here.

So, I have

I have now put
shopt -s globasciiranges
into .bashrc, and the ranges behave properly, as expected, even with
LC_ALL not being C.

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On Thu, 25 Jun 2020 21:47:04 -0000 (UTC), William Unruh wrote:

I do not. As I indicated I get the same as if I did [abB]*

Just be aware that [abB]* is not the same thing as [a-bB]*.

$ ls [a-bB]*
a1.a a.a a.A A.a a.B b.b B.B B.c

$ ls [abB]*
a1.a a.a a.A a.B b.b B.B B.c

$ ls
1a.a a1.a a.a a.A A.a a.B b.b B.B B.c c.a c.b c.d da.ad

$ env | grep utf
LANG=en_US.utf8
LANGUAGE=en_US.UTF-8:en_US:en
LESSCHARSET=utf-8
XTERM_LOCALE=en_US.utf8

$ echo $TERM
xterm

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On 26/6/20 10:25 am, Bit Twister wrote:

$ env | grep utf
LANG=en_US.utf8
LANGUAGE=en_US.UTF-8:en_US:en
LESSCHARSET=utf-8
XTERM_LOCALE=en_US.utf8

$ echo $TERM
xterm

Curiously I require "-i" to be stated.
Do you have it set globally?

[faeychild@unimatrix ~]$ env | grep -i utf
LC_MEASUREMENT=en_AU.UTF-8
LANG=en_AU.UTF-8
LC_TIME=en_AU.UTF-8


[faeychild@unimatrix ~]$ env | grep utf
[faeychild@unimatrix ~]$



--
faeychild
Running plasmashell 5.15.4 on 5.6.14-desktop-2.mga7 kernel.
Mageia release 7 (Official) for x86_64 installed via Mageia-7-x86_64-DVD.iso


--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On Sat, 27 Jun 2020 08:31:48 +1000, faeychild wrote:

On 26/6/20 10:25 am, Bit Twister wrote:

$ env | grep utf
LANG=en_US.utf8
LANGUAGE=en_US.UTF-8:en_US:en
LESSCHARSET=utf-8
XTERM_LOCALE=en_US.utf8

$ echo $TERM
xterm

Curiously I require "-i" to be stated.
Do you have it set globally?

Sorry about that. Yes I do. Same for locate.

$ type grep
grep is aliased to `grep --color --ignore-case'

$ type locate
locate is aliased to `locate --ignore-case'

[faeychild@unimatrix ~]$ env | grep -i utf
LC_MEASUREMENT=en_AU.UTF-8
LANG=en_AU.UTF-8
LC_TIME=en_AU.UTF-8

[faeychild@unimatrix ~]$ env | grep utf
[faeychild@unimatrix ~]$

--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)

On 27/6/20 9:34 am, Bit Twister wrote:

Sorry about that. Yes I do. Same for locate.

$ type grep
grep is aliased to `grep --color --ignore-case'

$ type locate
locate is aliased to `locate --ignore-case'

Ahh

[faeychild@unimatrix ~]$ type grep
grep is aliased to `grep --color'


[faeychild@unimatrix ~]$ type locate
locate is /usr/bin/locate



tiny but profound differences

regards


--
faeychild
Running plasmashell 5.15.4 on 5.6.14-desktop-2.mga7 kernel.
Mageia release 7 (Official) for x86_64 installed via Mageia-7-x86_64-DVD.iso


--- MBSE BBS v1.0.7.17 (GNU/Linux-x86_64)
* Origin: A noiseless patient Spider (2:250/1@fidonet)