Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.os.linux.misc > #16668 > unrolled thread

Which files/S contain the virus?

Back to article view | Back to comp.os.linux.misc

Contents

  Which files/S contain the virus? no.top.post@gmail.com - 2016-01-05 07:18 +0000
    Re: Which files/S contain the virus? Rich <rich@example.invalid> - 2016-01-05 11:06 +0000
      Re: Which files/S contain the virus? Joe Beanfish <joebeanfish@nospam.duh> - 2016-01-05 14:40 +0000
        Re: Which files/S contain the virus? Teemu Likonen <tlikonen@iki.fi> - 2016-01-05 17:48 +0200
        Re: Which files/S contain the virus? Unknown <dog@gmail.com> - 2016-01-11 17:03 +0000
    Re: Which files/S contain the virus? William Unruh <unruh@invalid.ca> - 2016-01-05 17:36 +0000
      Re: Which files/S contain the virus? Keith Keller <kkeller-usenet@wombat.san-francisco.ca.us> - 2016-01-05 10:51 -0800

#16668 — Which files/S contain the virus?

I've previously charactertized the new-fad of replacing the single-quote
by multiple bytes, as a VIRUS, because a destructive fad is a virus.
E2 80 ?? seems popular.

Some kiddie wrote <ASCII is no good any more, if we want to write in 
Chinese and Japanese> ?

My current problem is to find text-files which contain xE2 = d226.
`sed` can clean them, but I want to see them first.

  find ./ -type f -exec sed '/\d226/!d' {} \;
  find ./ -type f -exec sed '/\xe2/!d' {} \;

prints the lines containing a hex(E2) char,
which I can then paste into `mc` or some grepper, to get the fileName.

It's a pity that *nix is so ad hoc ?
AFAICS the character-classes listed for `grep` and `sed` are very different.
`sed` gives me the ability to specify any char/byte;
but `grep` doesn't cater for all 256.

What am I missing?

== TIA.

[toc] | [next] | [standalone]

#16669

In alt.os.linux.slackware no.top.post@gmail.com wrote:
> I've previously charactertized the new-fad of replacing the single-quote
> by multiple bytes, as a VIRUS, because a destructive fad is a virus.
> E2 80 ?? seems popular.

> Some kiddie wrote <ASCII is no good any more, if we want to write in 
> Chinese and Japanese> ?

> My current problem is to find text-files which contain xE2 = d226.
> `sed` can clean them, but I want to see them first.

>   find ./ -type f -exec sed '/\d226/!d' {} \;
>   find ./ -type f -exec sed '/\xe2/!d' {} \;

> prints the lines containing a hex(E2) char,
> which I can then paste into `mc` or some grepper, to get the fileName.

> It's a pity that *nix is so ad hoc ?
> AFAICS the character-classes listed for `grep` and `sed` are very different.
> `sed` gives me the ability to specify any char/byte;
> but `grep` doesn't cater for all 256.

> What am I missing?

That Chris Glur is an idiot.

  $ echo -e "\xe2\x80" > magic-quote
  $ xxd magic-quote
  0000000: e280 0a
  $ cp magic-quote magic-quote2
  $ grep $'\xe2' magic-quote*
  magic-quote:â€
  magic-quote2:â€

Grep works just fine with any character you give it to search (other
than maybe an ASCII null).  But first, you have to type the characters
into the shell in such a way that Bash understands your meaning.  You
can not directly type a \xe2 character (well, not easily) from the
keyboard.  So you have to tell Bash "this next thing is a special
character".  That is what the $'\xe2' piece does.  It tells Bash "this
next thing is a special character that can't be directly typed on the
keyboard".

The problem, as always, was your failure to understand.

[toc] | [prev] | [next] | [standalone]

#16676

On Tue, 05 Jan 2016 11:06:19 +0000, Rich wrote:
> In alt.os.linux.slackware no.top.post@gmail.com wrote:\
[annoying trolling snipped]
>> My current problem is to find text-files which contain xE2 = d226.
>> `sed` can clean them, but I want to see them first.
> 
>>   find ./ -type f -exec sed '/\d226/!d' {} \;
>>   find ./ -type f -exec sed '/\xe2/!d' {} \;
> 
>> prints the lines containing a hex(E2) char,
>> which I can then paste into `mc` or some grepper, to get the fileName.
> 
>> It's a pity that *nix is so ad hoc ?
>> AFAICS the character-classes listed for `grep` and `sed` are very different.
>> `sed` gives me the ability to specify any char/byte;
>> but `grep` doesn't cater for all 256.
> 
>> What am I missing?
> 
> That Chris Glur is an idiot.
> 
>   $ echo -e "\xe2\x80" > magic-quote
>   $ xxd magic-quote
>   0000000: e280 0a
>   $ cp magic-quote magic-quote2
>   $ grep $'\xe2' magic-quote*
>   magic-quote:â€
>   magic-quote2:â€
> 
> Grep works just fine with any character you give it to search (other
> than maybe an ASCII null).  But first, you have to type the characters
> into the shell in such a way that Bash understands your meaning.  You
> can not directly type a \xe2 character (well, not easily) from the
> keyboard.  So you have to tell Bash "this next thing is a special
> character".  That is what the $'\xe2' piece does.  It tells Bash "this
> next thing is a special character that can't be directly typed on the
> keyboard".
> 
> The problem, as always, was your failure to understand.

The real issue is that grep doesn't have it's own functional
syntax for representing any character (not sure when gnu broke this).
Your method requires a particular shell trick rather than standalone
grep or when running from a shell that doesn't know that trick.
If one resorts to perl regexp with -P there is some function but
still not complete:

$ echo -e "abc\xe2\x80" > magic-quote
$ echo -e "\xe2\x80" > magic-quote2
$ xxd magic-quote
00000000: 6162 63e2 800a                           abc...
$ xxd magic-quote2
00000000: e280 0a                                  ...
$ grep -P '\x62' magic-quote*
Binary file magic-quote matches
$ grep -P '\142' magic-quote*
Binary file magic-quote matches
$ grep -P '\xe2' magic-quote*
  [no match]
$ grep -P '\342' magic-quote*
  [no match]
$ grep --version
grep (GNU grep) 2.21
Copyright (C) 2014 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>.
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.

Written by Mike Haertel and others, see <http://git.sv.gnu.org/cgit/grep.git/tree/AUTHORS>.

[toc] | [prev] | [next] | [standalone]

#16679

[Multipart message — attachments visible in raw view] — view raw

Joe Beanfish [2016-01-05 14:40:05Z] wrote:

> The real issue is that grep doesn't have it's own functional syntax
> for representing any character (not sure when gnu broke this). Your
> method requires a particular shell trick [...]

That's maybe the "not reinventing the wheel" thing. One could say that
not every program needs to implement an input method for arbitrary
characters. Using shell's features or whatever input method user prefers
is fine.

-- 
/// Teemu Likonen   - .-..   <https://github.com/tlikonen> //
// PGP: 4E10 55DC 84E9 DFF6 13D7 8557 719D 69D3 2453 9450 ///

[toc] | [prev] | [next] | [standalone]

#16716

On Tue, 05 Jan 2016 14:40:05 +0000, Joe Beanfish wrote:

> On Tue, 05 Jan 2016 11:06:19 +0000, Rich wrote:
>> In alt.os.linux.slackware no.top.post@gmail.com wrote:\
> [annoying trolling snipped]
>>> My current problem is to find text-files which contain xE2 = d226.
>>> `sed` can clean them, but I want to see them first.
>> 
>>>   find ./ -type f -exec sed '/\d226/!d' {} \; find ./ -type f -exec
>>>   sed '/\xe2/!d' {} \;
>> 
>>> prints the lines containing a hex(E2) char, which I can then paste
>>> into `mc` or some grepper, to get the fileName.
>> 
>>> It's a pity that *nix is so ad hoc ?
>>> AFAICS the character-classes listed for `grep` and `sed` are very
>>> different. `sed` gives me the ability to specify any char/byte; but
>>> `grep` doesn't cater for all 256.
>> 
>>> What am I missing?
>> 
>> That Chris Glur is an idiot.
>> 
>>   $ echo -e "\xe2\x80" > magic-quote
>>   $ xxd magic-quote
>>   0000000: e280 0a
>>   $ cp magic-quote magic-quote2
>>   $ grep $'\xe2' magic-quote*
>>   magic-quote:â€
>>   magic-quote2:â€
>> 
>> Grep works just fine with any character you give it to search (other
>> than maybe an ASCII null).  But first, you have to type the characters
>> into the shell in such a way that Bash understands your meaning.  You
>> can not directly type a \xe2 character (well, not easily) from the
>> keyboard.  So you have to tell Bash "this next thing is a special
>> character".  That is what the $'\xe2' piece does.  It tells Bash "this
>> next thing is a special character that can't be directly typed on the
>> keyboard".
>> 
>> The problem, as always, was your failure to understand.
> 
> The real issue is that grep doesn't have it's own functional syntax for
> representing any character (not sure when gnu broke this). Your method
> requires a particular shell trick rather than standalone grep or when
> running from a shell that doesn't know that trick. If one resorts to
> perl regexp with -P there is some function but still not complete:
> 
> $ echo -e "abc\xe2\x80" > magic-quote $ echo -e "\xe2\x80" >
> magic-quote2
> $ xxd magic-quote
> 00000000: 6162 63e2 800a                           abc... $ xxd
> magic-quote2
> 00000000: e280 0a                                  ... $ grep -P '\x62'
> magic-quote*
> Binary file magic-quote matches
> $ grep -P '\142' magic-quote*
> Binary file magic-quote matches
> $ grep -P '\xe2' magic-quote*
>   [no match]
> $ grep -P '\342' magic-quote*
>   [no match]
> $ grep --version
> grep (GNU grep) 2.21
> Copyright (C) 2014 Free Software Foundation, Inc. License GPLv3+: GNU
> GPL version 3 or later <http://gnu.org/licenses/gpl.html>. This is free
> software: you are free to change and redistribute it. There is NO
> WARRANTY, to the extent permitted by law.
> 
> Written by Mike Haertel and others, see
> <http://git.sv.gnu.org/cgit/grep.git/tree/AUTHORS>.
=====
Yes, and my version seems to find more than yours,
eg. for the "[no match]"  :---

-> grep -P '\x62' magic-quote* == magic-quote:abc€€
-> grep -P '\142' magic-quote* == magic-quote:abc€€
-> grep -P '\xe2' magic-quote* == 
magic-quote:abc€€
magic-quote2:€€
-> grep -P '\342' magic-quote* ==
magic-quote:abc€€
magic-quote2:€€

-> man grep  == ...
 -P, --perl-regexp
  Interpret PATTERN as a Perl regular expression.This is highly
  experimental and grep -P may warn of unimplemented features.

No thank you !

=============== OK, I had a lie-down to relax with TTS, and
decided: so what if I object to 100 functions, each with
100 options = *nix ad-hokery and that I hate perl; I'll put it
in my library, as: FindQtVirusD
# Show files with apostrophe virus in DIR
find $1 -type f -exec grep -Pl '\xe2' {} \;
==== It's already put to work.

Thanks,
  take the rest of the day off. 
And ad-homenen: don't forget to take your meds.

[toc] | [prev] | [next] | [standalone]

#16681

On 2016-01-05, no.top.post@gmail.com <no.top.post@gmail.com> wrote:
> I've previously charactertized the new-fad of replacing the single-quote
> by multiple bytes, as a VIRUS, because a destructive fad is a virus.
> E2 80 ?? seems popular.

Please do not debase the word "virus" to mean "I do not like it". It has
a perfectly good meaning at present which differentiates certain things
from others, and to change the meaning to so nebulous a thing as
"something I do not like" is a disservice to the language and to clear
communication. 

>
> Some kiddie wrote <ASCII is no good any more, if we want to write in 
> Chinese and Japanese> ?

Not a kiddie, and the desire to service people who use different
languages than you happen to is laudible goal. But I do agree that
having random UTF stuck into text is a pain in the butt.

>
> My current problem is to find text-files which contain xE2 = d226.
> `sed` can clean them, but I want to see them first.
>
>   find ./ -type f -exec sed '/\d226/!d' {} \;
>   find ./ -type f -exec sed '/\xe2/!d' {} \;
>
> prints the lines containing a hex(E2) char,
> which I can then paste into `mc` or some grepper, to get the fileName.

egrep should will give the line and the filename. 

>
> It's a pity that *nix is so ad hoc ?

???

> AFAICS the character-classes listed for `grep` and `sed` are very different.
> `sed` gives me the ability to specify any char/byte;
> but `grep` doesn't cater for all 256.

egrep does AFAIK.

>
> What am I missing?
>
>== TIA.
>
>
>
>
>
>
>
>
>
>
>
>
>

[toc] | [prev] | [next] | [standalone]

#16686

["Followup-To:" header set to alt.os.linux.slackware.]
On 2016-01-05, William Unruh <unruh@invalid.ca> wrote:
>
> to change the meaning to so nebulous a thing as
> "something I do not like" is a disservice to the language and to clear
> communication. 

Indeed, impeding clear communication is ''Chris Glur's'' primary goal.

--keith


-- 
kkeller-usenet@wombat.san-francisco.ca.us
(try just my userid to email me)
AOLSFAQ=http://www.therockgarden.ca/aolsfaq.txt
see X- headers for PGP signature information

[toc] | [prev] | [standalone]

Back to top | Article view | comp.os.linux.misc

csiph-web