Cristian Rodriguez wrote:
> 2005/5/18, Tom Eastep <[EMAIL PROTECTED]>:
> 
> 
>>>- We need to look at the big issues and make sure we've got a good
>>>handle on them.  For me, these include (in rough order of priority from
>>>my perspective):
>>>* Multiple ISPs & load balancing
>>>* Features to enable building a personal firewall with shorewall
>>>* IPv6
>>>
> 
> My ideas  are the following:
> 
> 1. Declare Shorewall "alive but feature frozen",only bug fixes, until
> we can organize and get the needed coders,especially some good bash
> coders and an iptables geek (¿anyone?)

After 2.4.0 is out...

> 
> 2.I can take care of a mirror and enrole me to the website and 
> documentation. team.(please¡)
> 

I think that we already have enough mirrors -- I think that the issue of
 where to place the rsync server and how to keep the mirrors and
sourceforge synced needs addressing. Basically, we need someone to be
Webmaster.

Since you are volunteering ...

I've attached the simple scripts that I use to publish to Sourceforge
and to my own server (shorewall.net AKA lists.shorewall.net AKA
cvs.shorewall.net AKA rsync.shorewall.net).

The way that this works currently is that I run 'publish' which:

a) Converts files from XML to HTML if the original file is XML.
b) Uses SCP to copy the file(s) to my server and to sourceforge (you
will need a user account on Sourceforge).

All of the current documentation and the website files are in CVS
(Shorewall-Website/ and Shorewall-docs2/).

Possibly Mike Noyes can help you with coming up with a new arrangement;
he has offered to help with things at the Sourceforge end. I'm uncertain
how to set up a publishing scheme based at Sourceforge that can keep the
mirrors updated.

One of the flukes with the HTML documentation is that part of the files
have an ".htm" extension and part have ".html". The attached 'publish'
script depends on a "../Website" directory to be populated with the .htm
files. The script that I use to build Shorewall releases
(http://shorewall.net/pub/shorewall/contrib/makeshorewall.sh) includes a
list of the .htm files -- that's a better approach but I just never got
around to adding it to the publish script (if it ain't broke, don't fix it).

Both the publish and makeshorewall scripts require recent Docbook XML
tools to be installed.

I use XXE (Xmlmind XML Editor -- http://www.xmlbind.com/xmleditor) to
edit the Docbook XML files. The standard version is free and it's a nice
WYSIWYG editor.

Note that there are two different files used for the Left Frame on the
website -- one for Sourceforge and one for the other sites (Sourceforge
has Logo requirements for the sites that they host).

I publish those files using an scp command rather than the publish script.

Example for Sourceforge:

scp Shorewall_sfindex_frame.htm
shorewall.sf.net:/home/groups/s/sh/shorewall/htdocs/Shorewall_index_frame.htm

> 3. Files,mailist,and cvs stuff "should" be transfered to SF and a
> "help needed" item opened.

Nod -- we will move the mailing lists when Sourceforge has upgraded
their Mailman installation (I'm currently running a later release of
mailman so our files aren't compatible). CVS will be moved after 2.4.0.

> 
> 4. creat a more conservative release schema. IMHO shorewall have most
> of the needed features,but ipv6 is a good new one.
> 

That is in line with Paul's recommendation.

-Tom
-- 
Tom Eastep    \ Nothing is foolproof to a sufficiently talented fool
Shoreline,     \ http://shorewall.net
Washington USA  \ [EMAIL PROTECTED]
PGP Public Key   \ https://lists.shorewall.net/teastep.pgp.key
#!/bin/sh

#STYLESHEET=/home/teastep/Shorewall/docbook-xsl-1.62.4//xhtml/docbook.xsl
STYLESHEET=/usr/share/xml/docbook/stylesheet/nwalsh/current/xhtml/docbook.xsl
WEBSITE=/home/teastep/Shorewall/Website

if [ $# -eq 0 ]; then
    echo "Usage: $(basename $0) <XML file> ..."
    exit 2
fi

while [ $# -gt 0 ]; do

    case $1 in
        *.xml)
            b=${1%.*}

            if [ -f ../Website/$b.htm ]; then
                b=$b.htm
                f="$WEBSITE/$b"
            else
                b=$b.html
                f="$WEBSITE/$b"
            fi

            echo "Converting $1 from XML to HTML ($b) ..."

            if xmllint --valid --noout $1 && xsltproc --output $f --stringparam 
html.stylesheet html.css -param toc.section.depth 3 $STYLESHEET $1 ; then
                filestocopy="$filestocopy $f"
            fi
            ;;

        *)
            filestocopy="$filestocopy $1"
            ;;
    esac

    shift

done

if [ -n "$filestocopy" ]; then
    for f in $filestocopy; do
        filestoreport="$filestoreport $(basename $f)"
    done

    echo "Copying$filestoreport to SourceForge..."
    scp $filestocopy [EMAIL PROTECTED]:/home/groups/s/sh/shorewall/htdocs > 
/dev/null
    echo "Copying$filestoreport to Mail..."
    scp -p $filestocopy [EMAIL PROTECTED]:/var/www/html > /dev/null
fi


#!/bin/sh

        scp $@ [EMAIL PROTECTED]:/home/groups/s/sh/shorewall/htdocs/images
        scp -p  $@ [EMAIL PROTECTED]:/var/www/html/images
#       cp  -pf $@ /home/teastep/Shorewall-docs


#!/bin/sh

        scp $@ [EMAIL PROTECTED]:/home/groups/s/sh/shorewall/htdocs
        scp -p  $@ [EMAIL PROTECTED]:/var/www/html
        scp -p  $@ [EMAIL PROTECTED]:/var/www/lists
#       cp  -pf $@ /home/teastep/Shorewall-docs


Reply via email to