gmane.comp.shells.zsh.user

Autocomplete for umount with space

David Lee — 2012-05-23T10:10:55

Hi ZSH users,

Autocompletion for umount lists mount points clipped at the first space. For example, if I have "/Volumes/Media Drive" and "/Volumes/Media Disk", autocomplete for amount will only list a single "/Volumes/Media".

Is there a fix or workaround for this?

--David

globbing in assignment

Han Pingtian — 2012-05-23T06:18:59

Hello,

I just notice filename globbing doesn't occur in my script, like 

v=*

but it works in the interactive shell. Althoug I can make it work
by "setop glob_assign" or use "v=(*)", but it looks like the behavior's
diversity is a little strange.

Any thoughts?

Thanks.

completion oddity

Danek Duvall — 2012-05-21T21:40:23

If I have a completion function like this:

    _k () { _arguments --r1-word --r2-word }

and I press TAB at the end of the following commandline:

    % k --r

then I get

    % k --r-word

Another tab beeps and lists the choices, a third tab beeps and chooses
--r1-word, and thereon tabs simply cycle between the two.  That's more or
less expected.

If I define _k instead as

    _k () { _arguments --r1-word --really-r1-word }

then the second tab simply completes --r1-word, which seems wrong to me.

This is all with "zsh -f", with nothing but

    autoload -Uz compinit
    compinit -i
    compdef _k k

run before attempting the completion.

A co-worker says he noticed this behavior change (from better to worse)
sometime around the switch in Solaris from 4.3.10 to 4.3.12, but I can
reproduce it on our builds that had 4.3.9, so I'm guessing it's been around
for a while.

Thanks,
Danek

Extracting the 4th word of the first line in a file - is there a more elegant solution?

Ronald Fischer — 2012-05-18T13:16:54

Within a script, I have need to store the 4th word in the first line of
a file.

Of course this is trivial to implement. However, I'm curious whether
there exists a better way to implement than what I found.

My current solution goes like this:

# $1 is the filename
line=$(head -n 1 $1)
field=${lin[(w)4]}

This is OK, but this needs an auxiliary variable 'line'. 


I could also do it like this:

field=$(head -n 1 $1|cut  -f 4 -d ' ')

No aux variable, but needs a pipe


So I wonder, whether there is a elegant "zsh" solution, which solves
this in a more elegant way?

Ronald

How do I find shortest match?

TJ Luoma — 2012-05-16T19:18:31

I have a folder which has a lot of txt files, and in that folder 
are a lot of duplicate files. Most of the duplicates are 
numbered like this:

10-6- Make a universal 10-6-7 Snow Leopard installer-1.txt
10-6- Make a universal 10-6-7 Snow Leopard installer-2.txt
10-6- Make a universal 10-6-7 Snow Leopard installer-3.txt
10-6- Make a universal 10-6-7 Snow Leopard installer-4.txt
10-6- Make a universal 10-6-7 Snow Leopard installer.txt

Not not all of them. For example, I might have another identical 
file named

     todo-make-snowleopardinstaller.txt

What I want to do is go through the entire folder and find all 
duplicate files (files with identical md5sum).

Then I want to keep ONLY the one with the shortest filename.

Here's what I have so far

#!/bin/zsh

DIR=/Users/luomat/Dropbox/txt/

     # to avoid 'arg list too long'
     # note that 'gmd5sum' prints the sum
     # and then two spaces, and then the filename
ALL=$(find $DIR -type f -print0 | xargs -0 gmd5sum)

     # these are all the MD5 sums which occur MORE than one time
     # (which we get by removing any results with only one result
SUMS=($(echo $ALL | awk '{print $1}' | sort  | uniq -c |\
         egrep -v '^   1 ' | awk '{print $2}'))



for SUM in $SUMS
do

     # for each unique MD5 sum, do this:


     # get a list of all of the matching filenames MINUS the
     # sum itself
     MATCHES=($(echo "$ALL" | egrep "^$SUM" | sed "s${SUM}  ##g"))

     # ???

done


I don't know what to do in the ??? to compare the filenames and 
choose the shortest one.

Any ideas?

Or is there a better way to do this?

Thanks

TjL

global aliases substituting within a path

Ronald of Steiermark — 2012-05-16T11:41:18

I have a set of directory structures like this:

aaa/foo/bar/baz/xxx
bbb/foo/bar/baz/yyy
ccc/foo/bar/baz/zzz
etc.

I'm looking for a way to make typing easier on the command line, in
order to not have to type foo/bar/baz all the time.

Of course I can achieve this by setting a shell variable in my .zshrc:

X=foo/bar/baz

Then I can do for instance

  ls aaa/$X/xxx

Now I recently learned about global aliases, which permit alias
substitution to be done within the command line, and I thought that I
maybe could use this. Here was my (failed) attempt:

alias -g X=foo/bar/baz

# Does NOT work at hoped
ls aaa/X/xxx

X is not substituted, because it is not a word on its own (not
surrounded by spaces).

My question: For my problem, do I have to stick with my original
solution (shell variable), or is it a way to do it with aliases, or is
there maybe an even more clever way to achieve my goal?

Ronald

global aliases substituting within a path

Ronald Fischer — 2012-05-16T11:43:13

I have a set of directory structures like this:

aaa/foo/bar/baz/xxx
bbb/foo/bar/baz/yyy
ccc/foo/bar/baz/zzz
etc.

I'm looking for a way to make typing easier on the command line, in
order to not have to type foo/bar/baz all the time.

Of course I can achieve this by setting a shell variable in my .zshrc:

X=foo/bar/baz

Then I can do for instance

  ls aaa/$X/xxx

Now I recently learned about global aliases, which permit alias
substitution to be done within the command line, and I thought that I
maybe could use this. Here was my (failed) attempt:

alias -g X=foo/bar/baz

# Does NOT work at hoped
ls aaa/X/xxx

X is not substituted, because it is not a word on its own (not
surrounded by spaces).

My question: For my problem, do I have to stick with my original
solution (shell variable), or is it a way to do it with aliases, or is
there maybe an even more clever way to achieve my goal?

Ronald

changing case to Title Case

TJ Luoma — 2012-05-14T19:30:49

$ foo="ONE TWO thRee FoUR"
$ print -r -- "${(C)foo}"

will output this:

One Two Three Four

but this:

$ foo="ONE TWO thRee FoUR's"
$ print -r -- "${(C)foo}"



will give this:

One Two Three Four'S

Note the S is capitalized

$ foo="ONE TWO thRee FoUR's.pdf"
$ print -r -- "${(C)foo}"     

will give this:

One Two Three Four'S.Pdf

Note that the P in PDF and the S after ' are capitalized.

Is there any way to avoid that?

TjL

FNG specify a count

zzapper — 2012-05-14T15:53:58

Hi


This is a PHP regexp  /^FLOW_([0-9]{3}).csv$/

I can simulate this in a brute fashion


Is it possible to use a count 3 in ZSH FNG?

Tip to clean up your unused scripts

zzapper — 2012-05-11T15:52:42

Hi

Create a directory "backup" move  all your scripts there.

Copy the ones you 'know' you use back. Then only copy them back as you need 
them. You'll find that 80% remain in the backup folder!


Sorry kind of obvious but it has been useful to me :)

logical NOT

Mark van Dijk — 2012-05-10T11:59:34

Before I explain what I need let me explain why I need it.

Just now I was reading about MAC addresses, their binary
representation, and how MAC addresses are transmitted across a network.
This document mentions the address AC:DE:48:00:00:00. Converting that
to binary is easy:

for i (${(s.:.):-AC:DE:48:00:00:80}) echo $(( [#2] hex=0x$i ))

Here we can see 0xAC == 2#10101100.

From what I understand this is transmitted over a network from right to
left, so the receiver will receive 00110101 in that order. In this
sequence the first transmitted bit represents the individual/group
address bit and the second one represents the universally/locally
administered address bit, the reason for me reading this.

Then I wondered about how to let zsh "reverse" a binary number and came
up with the logical NOT: simply flip each bit. I've tried a lot of ways
but they all fail. What confuses me most is this:

% echo $(( [#2] ~0xAC ))
-2#10101101

I'd appreciate thoughts on this, and to hear if it makes any sense at
all :)

Thanks,
Mark.

how to use character range [] in substitution pattern :gs

reckoner — 2012-05-09T14:01:33

I'm trying to use the the [] character range to substitute for embedded 
parenthesis as in the following:

% echo ${x:gs/[()]/_/}

But This isn't working for some reason. It's not complaining about 
syntax, so I'm pretty sure that I'm using the character range [] 
incorrectly somehow.

Any help appreciated.

zargs: unexpected, non-xargs behaviour

Stephen Blott — 2012-05-07T08:41:41

Hi,

Consider this script:

# ----------------------------------------------------

   autoload -U zargs

   data=(foo bar)

   echo $data | xargs -n 1 echo XX
   zargs -n 1 -- $data -- =echo XX

# ----------------------------------------------------

I expected these to be equivalent, but here's the output:

   $ zsh < zargs_test.sh
   XX foo                                                                         
   XX bar                                                                         
   zargs: argument list too long                                                  
   zsh: exit 1     zsh < zargs_test.sh

To get zargs to work as intended (separate calls to echo for each
argument), I have use "-n 2".

So, at least for this example, when the documentation says "This function
works like GNU xargs, ...", that doesn't appear to be the case.

Am I missing something?

Steve

a 'require' function for zsh scripts and interactive functions

TJ Luoma — 2012-05-03T13:56:06

I'm trying to come up with a function which will allow me to 'require' that a given command is found in $PATH, so I can put a line at the top of a script like this:

#!/bin/zsh

require gmv lynx wget

...

or in another function

foo () {

require bar

bar  
}

and know that the script will check to see if those are there before executing.

I have defined 'require' like this

require () {

for UTIL in $< at >
do

if (( $+commands[$UTIL] ))
then
:
else
msg "No $UTIL found"

if [[ "$SHLVL" == "1" ]]
then
return 1
else
exit 1
fi
fi

done
}



The SHLVL is intended to keep my login shell from exiting if a function doesn't find a required command.

The : in the if/else/fi is because I wasn't sure how else to do a "not" for (( $+commands[$UTIL] ))

(`msg` is just a fancy way of doing `echo` which uses `growlnotify` on Mac. http://luo.ma/msg)

When I finished creating `require`, I found myself wondering if I had just reinvented a wheel that zsh already implemented some other way, so I thought I'd ask. I also thought there might be other ways to improve this if zsh didn't already have something built-in.

TjL

Command not found handler for non-searched commands?

Benjamin R. Haskell — 2012-04-29T16:22:16

I use autocd quite a bit.  And often the first thing I want to do when 
starting a new project is to create a directory and cd into it.  I tried 
creating the following command_not_found_handler:

command_not_found_handler () {
   local dir create
   echo HEREIAM >&2
   (( $# == 1 )) || return 1
   [[ $1 == */* ]] || return 1
   dir=$1
   read -q "create?Create $dir [y/N]? " || return 1
   mkdir -p $dir || return 1
   cd $dir
}

But, it doesn't get called when I need it:

## when I don't need it:
$ fake-command-here
HEREIAM
zsh: command not found: fake-command-here

## when I do need it:
$ ~/tmp/one-off-project
zsh: no such file or directory: /home/bhaskell/tmp/one-off-project

I see from the description of how commands are found that it won't get 
called when there's a slash in the command.  Is there a way to force it? 
Otherwise, is there another, easy way to accomplish my goal?

compinit causes completion to fail?

TJ Luoma — 2012-04-29T05:21:20

I split my .zshenv and .zshrc into several smaller files which are sourced separately, and now I've broken completion :-/

If I include these two lines in my zshenv file[1]:

autoload -U compinit
compinit



then completion fails with this message:

_complete:117: command not found: _normal

and a ~/.zcompdump file is created [2]

but if I comment out those two lines:

#autoload -U compinit
#compinit



then completion seems to work OK.

HOWEVER, if I remove my ~/.zshenv and all related ~/.z* files, and then use 'autoload -U compinit' and 'compinit' then it works fine, and the 'zcompdump' file looks better [4], which further solidifies my belief that the problem here is something I've done in my .zshenv files… but what?


Did I fail to load something before turning on compinit, or did I set something too early? Bindkeys, perhaps?


If you look at the 'zcompdump' file [2], specifically at line 117, it says:

bindkey '^[g" get-lin'   

that looks wrong to me. In fact, ALL of the bindkey lines look like they have lost the last character.

This started immediately after I split my ~/.zshenv file into several different files. You can see them all here [3]. This is using 4.3.11 on Mac OS X 10.7.3.  

I have no idea what I've done here, but I'd appreciate any help.


TjL


[1]  
https://www.dropbox.com/sh/7t1jagj5clkz0ci/_nzmp1eO9b/interactive/completion.sh





[2]
http://dl.dropbox.com/u/18414/tmp/zcompdump.txt

[3]
https://www.dropbox.com/sh/7t1jagj5clkz0ci/28xgQ85Mye

[4]
http://dl.dropbox.com/u/18414/tmp/zcompdump2.txt

do you use separate .zshenv and .zshrc files?

TJ Luoma — 2012-04-28T19:07:32

I keep vacillating between wanting to separate my config files into two separate files or keep them in one.

If I understand ZSH correctly, I could use one file (.zshenv) and put the settings which *would* have been in .zshrc into a block like this:

if [[ -o login ]]
then
# do stuff which would have been in .zshrc here
fi

The reason for doing this is that I tend to forget which file has whatever thing it is that I want to tweak/edit/change/add, and so I end up having to open one, search for what I'm looking for, realize it's in the other file, and then open the other file.

ISTM that it would be easier to just keep everything in one file and separate the login stuff using the `if` statement above.

Are there any drawbacks to that method?

can strftime show 'p.m.' instead of 'PM'?

TJ Luoma — 2012-04-28T18:55:25

Before I begin, I should say that I realize this may (seem to) be extremely picayune, but it consistently annoys me. Judge me as you will :-)

`man strftime` says this:

     %p    is replaced by national representation of either "ante meridiem" (a.m.)  or "post meridiem" (p.m.)  as appropriate.

     %F    is equivalent to ``%Y-%m-%d''.

     %r    is equivalent to ``%I:%M:%S %p''.

However when I do this in zsh

$ strftime "%F %r" "$EPOCHSECONDS"

I get this:

2012-04-28 02:50:24 PM

Ideally I would like "PM" to be "p.m." but I'd probably settle for "pm"

I tried using '%P' instead of '%p' (thinking that might invert the case) but that just gave me a literal 'P' instead.

I realize that I could use:

strftime "%F %r" "$EPOCHSECONDS" | tr '[A-Z]' '[a-z]'

or even

strftime "%F %r" "$EPOCHSECONDS" | sed 's#AM#a.m#g; s#pPM#p.m.#g'

but I wondered if there was a better (more efficient) way.

Thanks

TjL

${(s::)VAR} vs "${(s::)VAR}"

Kynn Jones — 2012-04-26T20:48:00

Hi!

I was surprised by the following (note that the first output consists of
3 lines while the second output consists of 4; the only difference between
the two cases is that the first one iterates over ${(s::)VAR} while the
second does so over its double-quoted variant, "${(s::)VAR}"):

% (VAR=123; for c ( ${(s::)VAR} ) echo ">$c<")
% (VAR=123; for c ( "${(s::)VAR}" ) echo ">$c<")

Even after-the-fact I cannot explain this difference.  I would appreciate
a description of what's going on (and that, I hope, will lead to the
sought-after explanation).

Thanks!

kj

rake completion with colons in their names

Benoit Daloze — 2012-04-21T12:55:49

Hello,

I've observed rake completion does not work fully with nested rake
tasks (having a namespace and being in the form "namespace:task").

This has already been reported in
zsh-workers 22600: http://www.zsh.org/mla/workers/2006/msg00475.html
and Debian #379942: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=379942

The generation part was fixed, but not the matching part.

Nikolai Weibull says "Adding 'r:|[:]=*' to ones matcher-list makes it
work, but should this really be necessary?".

Indeed, adding that to the matcher-list works for me, but I guess it's
not ideal to modify that global list.

I found a workaround in the zshcompsys manual at tag-order:
"Another use of patterns is to try multiple match specifications one
after another.  The matcher-list style offers something similar, but
it is tested very early in the completion system and hence can't be
set for single commands nor for more specific contexts."

  zstyle ':completion:*:*:rake:*' tag-order '*' '*:-colon'
  zstyle ':completion:*-colon' matcher 'r:|[:]=*'

Could someone confirm this is appropriate?

And if so, could it be added to zsh (or be documented)?

Cheers,
Benoit Daloze

Fwd: Re: Loading functions

Paul Hoffman — 2012-04-19T23:33:03

[Sorry for the duplicate, Mark; I accidentally replied to you privately 
and not to the list.]

On Thu, Apr 19, 2012 at 11:03:11PM +0000, Mark wrote:

I just started doing this sort of thing recently.  Here's a snippet from 
my ~/.zshrc:

cd() {
    if [[ -x .autodir/exit ]]; then
        print -- '(-autodir)' >&2
        source .autodir/exit
    fi
    builtin cd $< at > || return $?
    if [[ -x .autodir/enter ]]; then
        print -- '(+autodir)' >&2
        source .autodir/enter
    fi
}

HTH,

Paul.

gmane.comp.shells.zsh.user

Autocomplete for umount with space

globbing in assignment

completion oddity

Extracting the 4th word of the first line in a file - is there a more elegant solution?

How do I find shortest match?

global aliases substituting *within* a path

global aliases substituting *within* a path

changing case to Title Case

FNG specify a count

Tip to clean up your unused scripts

logical NOT

how to use character range [] in substitution pattern :gs

zargs: unexpected, non-xargs behaviour

a 'require' function for zsh scripts and interactive functions

Command not found handler for non-searched commands?

compinit causes completion to fail?

do you use separate .zshenv and .zshrc files?

can strftime show 'p.m.' instead of 'PM'?

${(s::)VAR} vs "${(s::)VAR}"

rake completion with colons in their names

Fwd: Re: Loading functions

global aliases substituting within a path

global aliases substituting within a path