ONLINE/COMMAND-LINE DICTIONARY

05-03-2004, 03:07 PM

Need a dictionary, but don't haveone in your system.

Create the following script called "dict" (or name of your choice).It uses the lynx web browser (freelyavailable on the internet) togo out to dictionary.com and find the word your looking for.

Code:

#!/bin/csh -f

set word=$1

# lynx is a text-based browser 
# available free on the web/net.
lynx -cfg=/dev/null -dump "http&#58;//www.dictionary.com/cgi-bin/dict.pl?term=$&#123;word&#125;" | more

**lodder** · 05-03-2004, 06:24 PM

thx man will try it out

**Asbenson** · 05-03-2004, 09:52 PM

Wow, I like it :shock:

**Ludootje** · 08-09-2004, 03:56 PM

Cool idea!
Of course, it made me want to write a bash version right away, here it is for those who don't have (t)csh installed:

Code:

#!/bin/bash
lynx -dump "http&#58;//www.dictionary.com/cgi-bin/dict.pl?term=$1" | less

It uses the 'less' pager instead of 'more', so you can scroll back up (ok, a lot of people probably have alias more="less" in their .bashrc, so to most people it probably won't matter a lot)
For those who use a pager a lot, I'd recommend to check out 'most' - it definitely beats less or one of the i'm-a-textbrowser-but-also-a-pager apps.
As you probably know, your pager is used to view man pages - here's an example of less versus most:
less pager
most pager
As you can see, 'most' highlights keywords in man pages.
You can set up your pager by modifying the $PAGER environment variable, so in your ~/.bashrc you can write:
export PAGER=/usr/bin/most

BTW I wonder why you specify "-cfg=/dev/null", this would break the script for those who have set up lynx to use a proxy, and *can't* browse the net without a proxy. Any specific reason for this?
Oh, and what does the '-f' do in csh? I don't have a tcsh or csh man page apparently (weird...) and the --help isn't very helpful :-/

**Ludootje** · 08-09-2004, 04:57 PM

I'm kinda bored right now, so I decided to make a "decent" version - yeah I know, almost no improvement or anything, but again, I did it out of pure boredom.

Code:

#!/bin/bash

# BASH script created by Ludootje <ludootje at linux dot be>
# written on&#58; 09/08/2004 &#40;Belgian date format&#41;
# released under the GNU General Public License &#40;a.k.a. GPL&#41;
# original csh version written by Mark "mcangeli" Angeli

help&#40;&#41; &#123;
        echo "Usage&#58;"
        echo "  --help&#58; display this help."
        echo "  $0 word1 &#91;word2 word3 ... wordn&#93;"
        echo
        echo "You can use as many words as you like."
        echo "When you've read the explanation of the word, hit 'q' to move on to the next word."
        echo "To exit the program when it's running, hit ^C &#40;i.e. control+c&#41;, followed by 'q'."

        exit 0
&#125;

dict&#40;&#41; &#123;
       for i in "$@"; do
               lynx -dump "http&#58;//www.dictionary.com/cgi-bin/dict.pl?term=$i" | less
       done
       exit 0
&#125;

case $1 in
   --help&#41; help
      ;;
   *&#41; for i in "$@"; do
         lynx -dump "http&#58;//www.dictionary.com/cgi-bin/dict.pl?term=$i" | less
    done
        exit 0
esac
#EOF

Anyone an idea what's wrong with the dict() function? I wanted to use that after the *), but than it doesn't work... If I call the help() function after *), it works fine, so I must've misconstructed something in the dict()...
Now I think about it, maybe I should pass arguments to it? But I have no idea how I could determine how many arguments (i.e. words in this case) are passed to the script, so I can't but them in variables. And dict($@) { ... } doesn't work.

**Prospero** · 11-18-2004, 09:09 PM

Here's one for a thesaurus.

Code:

#!/bin/bash
lynx -dump "http&#58;//www.m-w.com/cgi-bin/thesaurus?book=Thesaurus&va=$1" | less

**maccorin** · 11-19-2004, 02:06 AM

Originally Posted by Ludootje";p="2296

I'm kinda bored right now, so I decided to make a "decent" version - yeah I know, almost no improvement or anything, but again, I did it out of pure boredom.

Code:

#!/bin/bash

# BASH script created by Ludootje <ludootje at linux dot be>
# written on&#58; 09/08/2004 &#40;Belgian date format&#41;
# released under the GNU General Public License &#40;a.k.a. GPL&#41;
# original csh version written by Mark "mcangeli" Angeli

help&#40;&#41; &#123;
        echo "Usage&#58;"
        echo "  --help&#58; display this help."
        echo "  $0 word1 &#91;word2 word3 ... wordn&#93;"
        echo
        echo "You can use as many words as you like."
        echo "When you've read the explanation of the word, hit 'q' to move on to the next word."
        echo "To exit the program when it's running, hit ^C &#40;i.e. control+c&#41;, followed by 'q'."

        exit 0
&#125;

dict&#40;&#41; &#123;
       for i in "$@"; do
               lynx -dump "http&#58;//www.dictionary.com/cgi-bin/dict.pl?term=$i" | less
       done
       exit 0
&#125;

case $1 in
   --help&#41; help
      ;;
   *&#41; for i in "$@"; do
         lynx -dump "http&#58;//www.dictionary.com/cgi-bin/dict.pl?term=$i" | less
    done
        exit 0
esac
#EOF

Anyone an idea what's wrong with the dict() function? I wanted to use that after the *), but than it doesn't work... If I call the help() function after *), it works fine, so I must've misconstructed something in the dict()...
Now I think about it, maybe I should pass arguments to it? But I have no idea how I could determine how many arguments (i.e. words in this case) are passed to the script, so I can't but them in variables. And dict($@) { ... } doesn't work.

try

Code:

for i in $*

instead of

Code:

for i in $@

**knt** · 11-22-2004, 09:11 AM

I've just written a Q&D dictionary lookup script takes reference.com as base. Features:
- Have both thesaurus and dictionary support
- Omiting trailing (advertisement and unrequired) lines support

Code:

#!/bin/sh

# dict.sh      &#58; Simple command line dictionary.
# Requirements &#58; Internet connection to dictionary.com, lynx.
# Written by   &#58; Volkan YAZICI &#40;a.k.a. `knt'&#41; <v0lkany'at'yah*oo~com>
# License      &#58; GNU GPL &#40;General Public License&#41;


# &#123;&#123;&#123; Variables
LYNX="/usr/bin/lynx"
FILE_TMP=`mktemp`
URL_DICT="http&#58;//dictionary.reference.com/search?q="
URL_THES="http&#58;//thesaurus.reference.com/search?q="
# &#125;&#125;&#125;

# &#123;&#123;&#123; help&#40;&#41;
help&#40;&#41; &#123;
    echo "Usage&#58; $0 <-dict|-thes> <word> &#91;<word2> <word3> ...&#93;"
    exit 0
&#125;
# &#125;&#125;&#125;

# &#123;&#123;&#123; Check execution parameters
until &#91; -z "$1" &#93;
do
    case $1 in
        "-dict"&#41;
            URL=$URL_DICT
            ;; 

        "-thes"&#41;
            URL=$URL_THES
            ;;

        *&#41;
            WORD=" $1"
            ;;
    esac
    shift
done
# &#125;&#125;&#125;

# Remove trailing space at the start of the words sentence
WORD="`echo $WORD | sed 's/^ //g'`"

# Did the user satisfy the requirements?
&#91;&#91; $&#123;#URL&#125; -eq 0  || $&#123;#WORD&#125; -eq 0 &#93;&#93; && help

# Dump web page
$LYNX -dump "$&#123;URL&#125;$&#123;WORD&#125;" 1> $FILE_TMP 2>&1

# &#123;&#123;&#123; Chop trailing pieces
LN_COUNT=`wc -l $FILE_TMP | cut -d' ' -f1`
tail -n $&#91;$LN_COUNT - 12&#93; $FILE_TMP 1> $FILE_TMP.0
> $FILE_TMP
while read line
do
    &#91; "`echo $line | grep ADVERTISEMENT`" &#93; && break
    echo $line >> $FILE_TMP
done < $FILE_TMP.0
# &#125;&#125;&#125;

# Dump output
rm -f $FILE_TMP.0
cat $FILE_TMP && rm -f $FILE_TMP || echo "FILE_TMP&#58; $FILE_TMP"

# vim&#58; set ai tabstop=4 fdm=marker&#58;

**knt** · 11-29-2004, 01:57 PM

Originally Posted by knt

I've just written a Q&D dictionary lookup script takes reference.com as base. Features:
- Have both thesaurus and dictionary support
- Omiting trailing (advertisement and unrequired) lines support

Code:

#!/bin/sh

# dict.sh      &#58; Simple command line dictionary.
# Requirements &#58; Internet connection to dictionary.com, lynx.
# Written by   &#58; Volkan YAZICI &#40;a.k.a. `knt'&#41; <v0lkany'at'yah*oo~com>
# License      &#58; GNU GPL &#40;General Public License&#41;


# &#123;&#123;&#123; Variables
LYNX="/usr/bin/lynx"
FILE_TMP=`mktemp`
URL_DICT="http&#58;//dictionary.reference.com/search?q="
URL_THES="http&#58;//thesaurus.reference.com/search?q="
# &#125;&#125;&#125;

# &#123;&#123;&#123; help&#40;&#41;
help&#40;&#41; &#123;
    echo "Usage&#58; $0 <-dict|-thes> <word> &#91;<word2> <word3> ...&#93;"
    exit 0
&#125;
# &#125;&#125;&#125;

# &#123;&#123;&#123; Check execution parameters
until &#91; -z "$1" &#93;
do
    case $1 in
        "-dict"&#41;
            URL=$URL_DICT
            ;; 

        "-thes"&#41;
            URL=$URL_THES
            ;;

        *&#41;
            WORD=" $1"
            ;;
    esac
    shift
done
# &#125;&#125;&#125;

# Remove trailing space at the start of the words sentence
WORD="`echo $WORD | sed 's/^ //g'`"

# Did the user satisfy the requirements?
&#91;&#91; $&#123;#URL&#125; -eq 0  || $&#123;#WORD&#125; -eq 0 &#93;&#93; && help

# Dump web page
$LYNX -dump "$&#123;URL&#125;$&#123;WORD&#125;" 1> $FILE_TMP 2>&1

# &#123;&#123;&#123; Chop trailing pieces
LN_COUNT=`wc -l $FILE_TMP | cut -d' ' -f1`
tail -n $&#91;$LN_COUNT - 12&#93; $FILE_TMP 1> $FILE_TMP.0
> $FILE_TMP
while read line
do
    &#91; "`echo $line | grep ADVERTISEMENT`" &#93; && break
    echo $line >> $FILE_TMP
done < $FILE_TMP.0
# &#125;&#125;&#125;

# Dump output
rm -f $FILE_TMP.0
cat $FILE_TMP && rm -f $FILE_TMP || echo "FILE_TMP&#58; $FILE_TMP"

# vim&#58; set ai tabstop=4 fdm=marker&#58;

in the above code, replace 'WORD=" $1" ' line with 'WORD="$WORD $1"'