Newsgroups: best.www
Subject: Inserting header/footer without #include (server-parsed-html)
Summary: 
Expires: 
References: <642t07$8im$1@nntp1.ba.best.com> <65v196$g4n$1@shell3.ba.best.com> <65vln3$aru$1@nntp1.ba.best.com> <660g2s$mkd$1@nntp1.ba.best.com>
Sender: 
Followup-To: 
Distribution: best
Organization: Chez INWAP (people, computers, cats)
Keywords: 
Cc: 

In article <660g2s$mkd$1@nntp1.ba.best.com>,  <plambert$1@plambert.org> wrote:
>> At the risk of seeming simplistic, *why*?  
>If it's parsed, then the server has to search through it.  This step is
>probably 20-30x slower than straight html.
>
>Save the program at the end of this message as "preparse", ...
>Of course, I haven't really tested this program, and it's pretty incomplete.
>After reading this, wait 24 hours and check again for a followup from
>Joe Smith, who will post a version of this program that is two lines
>long, including the "#!/..." header.  ;-)

Sorry to disappoint you, Paul, it took 123 lines.  :-)
But that's because I took your idea and improved on it.
1) Eliminate the need for a separate template file,
2) Do variable substitution on the replacement text (like a macro).

It is now available in http://www.inwap.com/mybin/ .
	-Joe

(home/inwap) inwap@shell3% html-include
usage: /home/inwap/mybin/html-include include_dir $HOME/public_html
  where 'include_dir' is the directory containing the files to be included
  and '$HOME/public_html' is the directory with HTML files to be updated.
 
  For each *.html file, this program looks for lines like this:
  <!-- html-include file=footer prev=1.html next=2.html up=index.html -->
  several lines that will be replaced with contents of include_dir/footer
  <!-- html-include-end -->
 
  Each file that has any changes will be renamed to have a *.bak extension.
  The names of these *.bak files are sent to STDOUT so that they can
  be deleted if there are no errors.

(home/inwap) inwap@shell3% cat includes/footer
  <hr><table width="100%"><tr><td align=left>
  <a href="$prev"><img src="left.gif"></a>
  <a href="$up"><img src="up.gif"></a>
  <a href="$next"><img src="next.gif"></a>
  </td><td align=right>
  Maintained by js-cgi at inwap.com
  </td></tr></table>

(home/inwap) inwap@shell3% cat public_html/02.html
This line of text is before the footer
<!-- html-include file="footer" prev="01.html" up="00.html" next="02.html" -->
These lines of text get replaced
by executing html-include.
<!-- html-include-end -->
This line of text is after the footer

(home/inwap) inwap@shell3% html-include includes public_html/02.html
public_html/02.html - 1 change
public_html/02.html.bak

(home/inwap) inwap@shell3% cat public_html/02.html
This line of text is before the footer
<!-- html-include file="footer" prev="01.html" up="00.html" next="02.html" -->
        <!-- Contents of file includes/footer -->
  <hr><table width="100%"><tr><td align=left>
  <a href="01.html"><img src="left.gif"></a>
  <a href="00.html"><img src="up.gif"></a>
  <a href="02.html"><img src="next.gif"></a>
  </td><td align=right>
  Maintained by js-cgi@inwap.com
  </td></tr></table>
<!-- html-include-end -->
This line of text is after the footer

(home/inwap) inwap@shell3% html-include includes public_html/02.html
public_html/02.html - 0 changes

====================================================================

The section of the HTML file that needs to be updated is bracketed by
  <!-- html-include file="filename" optional_name="optional_value" -->
and
  <!-- html-include-end -->
and all the text between the two will be replaced by the contents of the
specified file, after doing variable substitution.  Since the start and
end tags are preserved, there HTML file is its own template; the output
from the html-include script can be used as input to the script later on.

The first argument to the html-include script is the name of the directory
that has the files to be included.  A good choice for this is "$HOME/includes"
which will keep the header and footer files in your home directory instead
of on the web area.  The remaining command-line arguments are names of
files and/or directories to process; the script does directories recursively
looking for *.html and *.htm files to process.  Only files that actually
need to be changed are rewritten and their modification dates are incremented
by 60 seconds.

If you have several levels of subdirectories, then you probably want to
create several header and footer files.
  includes/footer1 - uses <IMG SRC="img/next.gif">
  includes/footer2 - uses <IMG SRC="../img/next.gif">
  includes/footer3 - uses <IMG SRC="../../img/next.gif">
(Maybe I ought to build that into the program; have it use a single footer
file but insert "../" where appropriate.  Hmm.)

To do all of the files and subdirectories of your web area, use this:
	html-include includes public_html >temp
	rm `cat temp`

This is only the first cut at this html-include program.  Any suggestions?
	-Joe
The html-include.txt file has an example of its use, but you're right, it's
not written as a tutorial.  I'll have to work on getting a better example
written up.

1) Create a directory $HOME/includes .

2) Create two files, $HOME/includes/header and $HOME/includes/footer that
   have the HTML code that you want to have at the top of the page and
   at the bottom of the page, respectively.

3) On each file in $HOME/public_html, add two lines near the top and to 
   near the bottom.  For example, create test.html with this:

	<HTML><HEAD><TITLE>the title goes here</TITLE><BODY>
	<!-- html-include file=header -->
	<!-- html-include-end -->
	<H1>This is a test</H1>
	<!-- html-inclue file=footer -->
	<!-- html-include-end -->
	</BODY></HTML>

4) To do just the one file, use this command:
	html-include $HOME/includes public_html/test.html

5) Later on, when $HOME/includes/header and/or footer are changed, use
	html-include $HOME/includes public_html
   and it will locate all the files that need to be changed.

	-Joe