how to manipulate data like awk or perl when visiting a file

how to manipulate data like awk or perl when visiting a file

[reader]

I want to get a leg up on how I might manipulate a file like I can
with shell tools or awk/perl.

Examples:  I want to display only the first field of lines 30 thru 75
           I want to reverse field 1 and 5 and print those plus 6 of
           each line.
           I want to add up the numbers that appears in field 7 of
           lines 11 through 28 printing the total to ~/sumtot.txt

I'd like to see a few basic examples and maybe I'll be able to get the
idea enough to do some of that when I need to.

I don't really want to start way at the beginning of elisp to get
started.   I hoped maybe there are some examples like that available
in existing info documents.

I haven't noticed much talk here about those kinds of chores so maybe
elisp isn't a good choice for that?

Re: how to manipulate data like awk or perl when visiting a file

[rustom]

On Dec 29, 9:14 am, rea...@newsguy.com wrote:
> I want to get a leg up on how I might manipulate a file like I can
> with shell tools or awk/perl.
>
> Examples:  I want to display only the first field of lines 30 thru 75
>            I want to reverse field 1 and 5 and print those plus 6 of
>            each line.
>            I want to add up the numbers that appears in field 7 of
>            lines 11 through 28 printing the total to ~/sumtot.txt
>
> I'd like to see a few basic examples and maybe I'll be able to get the
> idea enough to do some of that when I need to.
>
> I don't really want to start way at the beginning of elisp to get
> started.   I hoped maybe there are some examples like that available
> in existing info documents.
>
> I haven't noticed much talk here about those kinds of chores so maybe
> elisp isn't a good choice for that?

There was a discussion on something like this a few days ago. See
http://groups.google.com/group/gnu.emacs.help/browse_thread/thread/77dc549480d95f92#

Re: how to manipulate data like awk or perl when visiting a file

[Xah Lee]

On Dec 28, 8:14 pm, rea...@newsguy.com wrote:
> I want to get a leg up on how I might manipulate a file like I can
> with shell tools or awk/perl.
>
> Examples:  I want to display only the first field of lines 30 thru 75
>            I want to reverse field 1 and 5 and print those plus 6 of
>            each line.
>            I want to add up the numbers that appears in field 7 of
>            lines 11 through 28 printing the total to ~/sumtot.txt
>
> I'd like to see a few basic examples and maybe I'll be able to get the
> idea enough to do some of that when I need to.
>
> I don't really want to start way at the beginning of elisp to get
> started.   I hoped maybe there are some examples like that available
> in existing info documents.
>
> I haven't noticed much talk here about those kinds of chores so maybe
> elisp isn't a good choice for that?


do you want to do this as interactive use or do you mean how to do it
in elisp?

For elisp, here's a example:

• How To Process A File Line By Line In Elisp
  http://xahlee.org/emacs/elisp_process_lines.html

for your task, perhaps it's easier to simply read all content in a
file, use split-string so that each line becomes a list item, then use
split-string again on each item so that you got the fields.

Of course, you can also just write your code in awk or perl, and have
a emacs wrapper to call them as emacs commands. See

• Elisp Wrapper For Perl Scripts
  http://xahlee.org/emacs/elisp_perl_wrapper.html

For interactive use, depending what specifically what you want to do,
but it might be easier to simply use shell-command or shell-command-on-
region, and call your sed awk.

  Xah
∑ http://xahlee.org/

☄

Re: how to manipulate data like awk or perl when visiting a file

[poppyer]

rustom <rustompmody@gmail.com> writes:

: On Dec 29, 9:14 am, rea...@newsguy.com wrote:
:> I want to get a leg up on how I might manipulate a file like I can
:> with shell tools or awk/perl.
:>
:> Examples:  I want to display only the first field of lines 30 thru 75
:>            I want to reverse field 1 and 5 and print those plus 6 of
:>            each line.
:>            I want to add up the numbers that appears in field 7 of
:>            lines 11 through 28 printing the total to ~/sumtot.txt
:>
:> I'd like to see a few basic examples and maybe I'll be able to get the
:> idea enough to do some of that when I need to.
:>
:> I don't really want to start way at the beginning of elisp to get
:> started.   I hoped maybe there are some examples like that available
:> in existing info documents.
:>
:> I haven't noticed much talk here about those kinds of chores so maybe
:> elisp isn't a good choice for that?
:
: There was a discussion on something like this a few days ago. See
: http://groups.google.com/group/gnu.emacs.help/browse_thread/thread/77dc549480d95f92#

This seems not so relevant.

One way I can think of is to replace TAB to | first
and then use orgtbl-mode, the Org Table editor to do the work.

Is there any way without replace TAB first and manipulate them directly?

Cheers,
poppyer

Re: how to manipulate data like awk or perl when visiting a file

[Harry Putnam]

Xah Lee <xahlee@gmail.com> writes:

> For interactive use, depending what specifically what you want to do,
> but it might be easier to simply use shell-command or shell-command-on-
> region, and call your sed awk.

Thanks for your thoughtful answer. Interesting pages you have.

I got pretty interested reading your comparison of perl vs elisp. 

What prompted my question here was that I had no access to shell tools
or perl on a specific machine I was on right then, so it got me to
thinking why I never see much mention of using elisp for those kinds
of jobs.

Its an interesting subject but in fact I'm rarely in that situation
so usually rely on shell tools awk and perl.  Those I have some small
grasp of.

Your examples may lead me to work on elisp more than I have.

Thanks again.

Re: how to manipulate data like awk or perl when visiting a file

[Kevin Rodgers]

reader@newsguy.com wrote:
> I want to get a leg up on how I might manipulate a file like I can
> with shell tools or awk/perl.
> 
> Examples:  I want to display only the first field of lines 30 thru 75
>            I want to reverse field 1 and 5 and print those plus 6 of
>            each line.
>            I want to add up the numbers that appears in field 7 of
>            lines 11 through 28 printing the total to ~/sumtot.txt

Are you familiar with M-| (shell-command-on-region)?

-- 
Kevin Rodgers
Denver, Colorado, USA

Re: how to manipulate data like awk or perl when visiting a file

[Ted Zlatanov]

On Tue, 30 Dec 2008 00:59:28 -0600 Harry Putnam <reader@newsguy.com> wrote: 

HP> I got pretty interested reading [Xah's] comparison of perl vs elisp. 

HP> What prompted my question here was that I had no access to shell tools
HP> or perl on a specific machine I was on right then, so it got me to
HP> thinking why I never see much mention of using elisp for those kinds
HP> of jobs.

HP> Its an interesting subject but in fact I'm rarely in that situation
HP> so usually rely on shell tools awk and perl.  Those I have some small
HP> grasp of.

The biggest problem with ELisp file processing vs. Perl/awk/etc. is that
efficient line-by-line processing is not possible.  Thus you're limited
to either byte offsets or what will fit in a buffer.  This is a big
inconvenience for large files.

Also, concise one-liners are trivial with Perl and awk, while Emacs
one-liners are painfully verbose (with the one saving grace that $
doesn't show up much in ELisp code).  By the time you're on the third
screen line, you've forgotten what you needed in the first place.

Ted

Re: how to manipulate data like awk or perl when visiting a file

[Richard Riley]

Kevin Rodgers <kevin.d.rodgers@gmail.com> writes:

> reader@newsguy.com wrote:
>> I want to get a leg up on how I might manipulate a file like I can
>> with shell tools or awk/perl.
>>
>> Examples:  I want to display only the first field of lines 30 thru 75
>>            I want to reverse field 1 and 5 and print those plus 6 of
>>            each line.
>>            I want to add up the numbers that appears in field 7 of
>>            lines 11 through 28 printing the total to ~/sumtot.txt
>
> Are you familiar with M-| (shell-command-on-region)?

See Xah Lee's excellent pages too:

http://xahlee.org/emacs/emacs_unix.html