Thu, 18 Sep 2008 06:06:09 -0500, problems did catÂ*:

> I want to have a part of my bash-script testing the first char of $1, or
> better the string of 4-chars.
>
> A sh script on hand, apparently does something like required:
>> case `echo $1 |sed 's/^.*\.//'` in
> CASE ststement [write the $1 file (?1st line only) via sed which
> replaces with nil
>> c|h)
> OF branch "c" or "h" chars
>> indent -kr -pcs "$2"
> do actions for "c" or "h" case
>> ;;
>> C|cc|CC|cxx|CXX|cpp|CPP)
>> astyle "$2"
>> ;;
>> java|JAVA)
>> astyle --style=java --mode=java "$2"
>> ;;
> default CASE
>> esac
>
> Since the example is examining file extensions, is it restricted to
> examining the tail of $2 ?
>
> How would it examine the 'head' of $2 ?

there are several ways to ``examine the 'head' of $2 ($1 I suppose?) ''
anyway, as you started with sed let's go with it, using the
same loosy assumes about head/tail marked by a dot:
echo $1 |sed 's/\..*$//'

On Thu, 18 Sep 2008 06:06:09 -0500, problems@gmail wrote:
> I want to have a part of my bash-script testing the first char of $1,
> or better the string of 4-chars.

Some light reading found here
http://tldp.org/LDP/abs/html/index.html
for a hot start, check out the B-5. String Operations section.


> Since the example is examining file extensions, is it
> restricted to examining the tail of $2 ?

Tail is a bit of overkill, Try this command

fn=test.data; echo "${fn##*.}"

> How would it examine the 'head' of $2 ?

fn=/home/data/test.data; echo "$(basename $fn)"

problems@gmail wrote:
> I want to have a part of my bash-script testing the first char of $1,
> or better the string of 4-chars.
>
> A sh script on hand, apparently does something like required:
>> case `echo $1 |sed 's/^.*\.//'` in
> CASE ststement [write the $1 file (?1st line only)
> via sed which replaces with nil
>> c|h)
> OF branch "c" or "h" chars
>> indent -kr -pcs "$2"
> do actions for "c" or "h" case
>> ;;
>> C|cc|CC|cxx|CXX|cpp|CPP)
>> astyle "$2"
>> ;;
>> java|JAVA)
>> astyle --style=java --mode=java "$2"
>> ;;
> default CASE
>> esac
>
> Since the example is examining file extensions, is it
> restricted to examining the tail of $2 ?
>
> How would it examine the 'head' of $2 ?
>
> TIA
>
> == Chris Glur.
>
Something like this may help:

$ TESTSTRING=abcd
$ echo ${TESTSTRING}
abcd
$ echo ${TESTSTRING:0:1}
a
$ echo ${TESTSTRING:1:3}
bcd


Eric

On 2008-09-18, problems@gmail wrote:
> I want to have a part of my bash-script testing the first char of $1,
> or better the string of 4-chars.
>
> A sh script on hand, apparently does something like required:
>> case `echo $1 |sed 's/^.*\.//'` in

To get the first character of $1:

c1=${1%"${1#?}"}

To get the first 4 characters:

c4=${1%"${1#????}"}

In bash and ksh93, the above can be done with:

c1=${1:0:1}
c4=${1:0:4}

To test the first character:

case $1 in
a*) echo a ;;
b*) echo b ;;
## etc..
esac

> CASE ststement [write the $1 file (?1st line only)
> via sed which replaces with nil
>> c|h)
> OF branch "c" or "h" chars
>> indent -kr -pcs "$2"
> do actions for "c" or "h" case
>> ;;
>> C|cc|CC|cxx|CXX|cpp|CPP)
>> astyle "$2"
>> ;;
>> java|JAVA)
>> astyle --style=java --mode=java "$2"
>> ;;
> default CASE
>> esac
>
> Since the example is examining file extensions, is it
> restricted to examining the tail of $2 ?

To extract the extension (i.e., the final characters):

ext=${1##*.}

To test the extension:

case $1 in
*.c) echo "C source file" ;;
*.h) echo "C header file" ;;
# etc.
esac

> How would it examine the 'head' of $2 ?


--
Chris F.A. Johnson, author |
Shell Scripting Recipes: | My code in this post, if any,
A Problem-Solution Approach | is released under the
2005, Apress | GNU General Public Licence

On Thu, 18 Sep 2008 19:55:10 +0000, Chris F.A. Johnson wrote:

> On 2008-09-18, problems@gmail wrote:
>> I want to have a part of my bash-script testing the first char of $1,
>> or better the string of 4-chars.
>>
>> A sh script on hand, apparently does something like required:
>>> case `echo $1 |sed 's/^.*\.//'` in
>
> To get the first character of $1:
>
> c1=${1%"${1#?}"}
>
> To get the first 4 characters:
>
> c4=${1%"${1#????}"}
>
> In bash and ksh93, the above can be done with:
>
> c1=${1:0:1}
> c4=${1:0:4}
>
> To test the first character:
>
> case $1 in
> a*) echo a ;;
> b*) echo b ;;
> ## etc..
> esac

Chris F.A. Johnson: he represents what info pages could have been :-)