On Nov 30, 9:57 am, Ed Morton <mor... [at] lsupcaemnt.com> wrote:
> On 11/29/2007 10:44 PM, sant... [at] gmail.com wrote:
>
>
>
> > On Nov 29, 8:49 am, Ed Morton <mor... [at] lsupcaemnt.com> wrote:
>
> >>On 11/28/2007 12:23 AM, sant... [at] gmail.com wrote:
>
> >>>Sorry I have posted this before but I have a slight change in the
> >>>question.
>
> >>>I have an html file.The entire script is in one line only. The
> >>>following is the script.
>
> >>><table><tbody><tr><td class="r">Chapter
> >>>1: ................................................. </td></tr></
> >>>tbody></table><p Hare Krishna ......................</p> <p Hare
> >>>Rama ......................</p>
>
> >>>where .............. is a variable text
>
> >>>In the above script I want to delete the text
>
> >>><table><tbody><tr><td class="r">Chapter
> >>>1: ................................................. </td></tr></
> >>>tbody></table>
>
> >>>where ........ represents variable content.
>
> >>>I have 100 files with names 1.htm to 100.htm
>
> >>>How can i do this using unix commands rather than selecting the text
> >>>and deleting.
>
> >>Depending on whether not "<table>" or "</table>" can occur multiple times on a
> >>line, this may be all you need:
>
> >>for file in *.htm
> >>do
> >> sed 's:<table>.*</table>::' "$file" > tmp &&
> >> mv tmp "$file"
> >>done
>
> >>Regards,
>
> >> Ed.
>
> > "<table>" or "</table>" can occur multiple times on a line then what
> > can be done
>
> Use all of the unique text you mentioned and replace the chain of periods with ".*":
>
> sed 's:<table><tbody><tr><td class="r">Chapter 1:.*</td></tr></tbody></table>::'
>
> If the text on either side of the ".*" can appear elsewhere on the same line,
> then it's a harder problem that needs a different approach.
>
> Ed.
Thank you.
