Help with template like solution

Shell Programming and Scripting
1

hi experts,

i'm trying to do this:

file1 is a template. might have kinds of 'funny' characters.
sample:

<body>
<form>
<p><input type="text" name="abc"/></p>
�
<p><my_content></p>
</form>
</body>

file2 is a file that contains lots of text. this might be very big. might have 'funny' characters as well.
sample:

<tr><td align="center">1.1</td><td>1.2</td></tr>
... many many lines ...
<tr><td align="center">2000.1</td><td>2000.2</td></tr>

my question is, how do i read file1, replace <my_content> with the content of file2, and store/print out the result?

i've tried to use sed, i firstly escape the content of file2, these characters: \ &
then i use sed substitute. but when file2 gets very big, sed would display error that the parameter is too long.

any suggestions on how to do this? i do not need the regex feature. just search that <my_content> and replace it with the content from file2.

thanks for the help.

2

Hope you can modify your template like (so that "<my_content>" has its own line).

<body>
<form>
<p><input type="text" name="abc"/></p>
�
<p>
<my_content>
</p>
</form>
</body>

Then this works (takes everything before "<my_content>" , the file2 and then everything after "<my_content>" )

#!/bin/bash
{   sed -n '1,/\<my_content\>/p' file1 | sed '$d'
    cat file2
    sed -n '/\<my_content\>/,//p' file1 | sed '1d'
} > file3

If "<my_content>" is an html comment, you can remove everything after the pipes in the script, it will leave the comment in the output file as markers to see where you've put your data.

3

An AWK solution :

awk '
   NR==FNR { string = string $0 "\n" ; next}
   { gsub(/<my_content>/, string) ; print }
' file2 file1

Jean-Pierre.