On Wed, 2002-11-27 at 21:12, Franki wrote:
> no,, 127.0.0.1 should work.. but it means those entries in your error log
> have nothing to do with you..
>
> give us more details..
>
> 1. where exactly is the file installed?? (eg /var/www/cgi-bin ) {There are two
>parts in two directories the xmlboard.cgi is in the /var/www/cgi-bin directory}
{And the rest of the program is in a directory called forum. I first put
it under var/www/ then I thought that maybe it should be in my home
directory so I made one there and changed the path to it in the
xmlboard.cgi script file}
> 2. Where did you get the script, whats it called, where did you get it and
> what does it do (supposed to)?{ It is a xml based forum. I got it from cgi resources
>the home site of the script is http://xmlboard.hypermart.net/ }
> 3. What exact url are you using in the browser to test the script.{
>http://localhost/cgi-bin/xmlboard.cgi}
>
> that info should help in working out what is going on.
{Also here is the readme file as an attachment.}
>
Thanks
Aaron
> rgds
>
> Frank
>
> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Aaron
> Sent: Thursday, 28 November 2002 4:49 AM
> To: [EMAIL PROTECTED]
> Subject: RE: [newbie] cgi help
>
>
> On Wed, 2002-11-27 at 20:29, Franki wrote:
> > is your IP address either of those?? my ipaddress?? I am using localhost
> do I need an ipaddress for this and if so???
> I am using apache to locally test a site I built. I would also love to
> let others see it. Am I barking up the wrong tree??
>
> Or is the ipaddress the one my machine uses to connect to the net???
>
> Thanks
> Aaron
> >
> > the vti and mem and msadc look like a frontpage application of some sort..
> >
> > just what is the script you are trying to run???
> >
> >{ xmlboard and clinks
> I am trying to get a usable free forum and a cgi script that will rotate
> a line of text weekly.}
> > rgds
> >
> > Frank
> >
> > -----Original Message-----
> > From: [EMAIL PROTECTED]
> > [mailto:[EMAIL PROTECTED]]On Behalf Of Aaron
> > Sent: Thursday, 28 November 2002 4:10 AM
> > To: [EMAIL PROTECTED]
> > Subject: RE: [newbie] cgi help
> >
> >
> > On Wed, 2002-11-27 at 19:18, Franki wrote:
> > > here are some steps to try.
> > > OK here is the output but I am not sure what it means
> > > 1. Check the error log.. open a console: tail /var/log/httpd/error_log
> > [Wed Nov 27 21:05:15 2002] [error] [client 62.219.125.72] File does not
> > exist: /var/www/html/_vti_bin
> > [Wed Nov 27 21:05:20 2002] [error] [client 62.219.125.72] File does not
> > exist: /var/www/html/_mem_bin
> > [Wed Nov 27 21:05:21 2002] [error] [client 62.219.125.72] File does not
> > exist: /var/www/html/msadc
> > [Wed Nov 27 21:05:23 2002] [error] [client 62.219.125.72] File does not
> > exist: /var/www/html/scripts
> > [Wed Nov 27 21:05:27 2002] [error] [client 62.219.125.72] File does not
> > exist: /var/www/html/scripts
> > [Wed Nov 27 21:05:29 2002] [error] [client 62.219.125.72] File does not
> > exist: /var/www/html/scripts
> > [Wed Nov 27 21:05:34 2002] [error] [client 62.219.125.72] File does not
> > exist: /var/www/html/scripts
> > [Wed Nov 27 21:05:36 2002] [error] [client 62.219.125.72] File does not
> > exist: /var/www/html/scripts
> > [Wed Nov 27 21:14:54 2002] [error] [client 62.219.128.77] File does not
> > exist: /var/www/html/scripts
> > [Wed Nov 27 21:14:57 2002] [error] [client 62.219.128.77] File does not
> > exist: /var/www/html/scripts
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > > 2. Make sure the file was uploaded as asci not binary.. if you used an
> ftp
> > > client.
> >
> > This is all localhost
> > > 3. Make sure /etc/httpd/conf/commonhttpd.conf or httd.conf has a section
> > for
> > > /cgi-bin/ and that its set to ALLOW.
> > >
> > > 5 make sure the permissions and ownership are correct.. as a basis to
> see
> > if
> > > you can get it working, try this:
> > >
> > > chmod 755 myscript.pl
> > > chown apache.apache myscript.pl
> > >
> > > then try it again..
> > >
> > > they are the most common problems I have seen with cgi scripts and
> server
> > > 500 errors.
> > >
> > >
> > > rgds
> > >
> > > Frank
> > >
> > >
> > > -----Original Message-----
> > > From: [EMAIL PROTECTED]
> > > [mailto:[EMAIL PROTECTED]]On Behalf Of Aaron
> > > Sent: Thursday, 28 November 2002 12:37 AM
> > > To: [EMAIL PROTECTED]
> > > Subject: RE: [newbie] cgi help
> > >
> > >
> > > Since were on the subject of cgi.
> > > I setup apache and put a test site in the var/www/html/ directory. I
> > > added some cgi scripts and programs I download from the net. However non
> > > of them will work for me. I checked that I have the correct path to perl
> > > But I still get Internal Server Error from the apache server.
> > > I don't know what I am missing.
> > > Could someone help??
> > > Thanks
> > > Aaron
> > >
> > > On Wed, 2002-11-27 at 11:28, Franki wrote:
> > > > try swapping the url to the image and css from relative to absolute,
> and
> > > > move both out of the cgi-bin.
> > > >
> > > > they should be somewhere like /var/www/html/images
> > > >
> > > > in the script they should link to the URL..
> > > > http://mydomain.com/images/gandhi.jpg
> > > >
> > > > see how you go with that. some setups will not server images and stuff
> > > from
> > > > the cgi-bin.
> > > >
> > > > rgds
> > > >
> > > > Franki
> > > >
> > > > -----Original Message-----
> > > > From: [EMAIL PROTECTED]
> > > > [mailto:[EMAIL PROTECTED]]On Behalf Of L.V.Gandhi
> > > > Sent: Tuesday, 26 November 2002 11:27 PM
> > > > To: [EMAIL PROTECTED]
> > > > Subject: [newbie] cgi help
> > > >
> > > >
> > > > When I call the cgi file given below, I don't get the effects of
> > > stylesheet
> > > > mentioned in the header and also the picture. both the files are in
> the
> > > same
> > > > folder as that of the cgi file.
> > > > ===============================================
> > > > #!/usr/bin/perl
> > > >
> > > > print <<End;
> > > > Content-Type: text/html\n\n
> > > > <html>
> > > > <head>
> > > > <title>test</title>
> > > > <link rel="stylesheet" type="text/css" href="teststyle.css" />
> > > > </head>
> > > > <body>
> > > > <h1>test</h1>
> > > > <img src="gandhi.jpg" alt="My picture" height="153" width="131"
> > > > align="left">
> > > > </body>
> > > > </html>
> > > > End
> > > > ==============================
> > > > I called the script with test .html given below.
> > > > <html><head><title>Test</title></head>
> > > > <body>
> > > > <a href="cgi-bin/test1.pl">csstest<a><br>
> > > > </body></html>
> > > > What should be done to get the desired effect.
> > > > --
> > > > L.V.Gandhi
> > > > 203, Soundaryalahari Apartments, Lawsons Bay colony, Visakhapatnam,
> > 530017
> > > > MECON, 5th Floor, RTC Complex, Visakhapatnam AP 530020 INDIA
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > ----
> > > >
> > >
> > > > Want to buy your Pack or Services from MandrakeSoft?
> > > > Go to http://www.mandrakestore.com
> > >
> > >
> > >
> > >
> > >
> > > ----
> > >
> >
> > > Want to buy your Pack or Services from MandrakeSoft?
> > > Go to http://www.mandrakestore.com
> >
> >
> >
> >
> >
> > ----
> >
>
> > Want to buy your Pack or Services from MandrakeSoft?
> > Go to http://www.mandrakestore.com
>
>
>
>
>
> ----
>
> Want to buy your Pack or Services from MandrakeSoft?
> Go to http://www.mandrakestore.com
#############################################################################
#XMLBoard : An XML-Driven Multi-Forum Message Board v 1.5 #
#Copyright (C) 2002 Joel Rosen #
#WWW: http://xmlboard.hypermart.net #
#Contact: [EMAIL PROTECTED] #
#############################################################################
XMLBoard Features:
* Data is stored in XML format; this format is a rather dynamic one
as it is quickly becoming the universally preferred method of
storing data, and keeping information separate from appearance in
use on the web. (See sites such as xml.com and xml.apache.org for
more info on the possibilities of XML.) The data is read via
the perl module XML::Parser, which interacts with James Clark's
Expat.
* Both significant disk space is saved and much functionality is added
by storing the posts in XML as opposed to HTML -- design changes
for all posts can be easily made by modifying one file, and separate
HTML files with all the design tags do not need to be written out for
each post.
* XMLBoard supports multiple forums and multiple designs with just
one installation. Setup is simple, and the HTML presentation for
each forum can be easily edited with user-specified HTML files which
are read by the script.
* Optional cookies can remember the user's name, and tell the user
the last time they visited the forum, and the number of posts
added since then. New posts are indicated with "new". Each forum
uses a different set of cookies, to differentiate between the
forums.
* Users can search the forum for posts containing words or phrases, in
any of the author, subject, or message fields.
* One administrator is allowed per forum, and he/she may remove posts
and entire threads from the index, or remove all posts added before
a specified date. A name and password is required.
* Each forum has several other configurable options that tweak the
operation of the script, located in a different configuration
file for each forum, allowing for variation in the functionality
of the forums, in addition to variation in design.
Installation:
Requirements:
* The Apache webserver.
* Perl version 5.004 or higher.
* James Clark's Expat -- Note: Apache versions 1.3.9 and above
already come with Expat installed; I recommend upgrading Apache
if you can, because setting up Expat for Apache is a mystery
to me.
* The XML::Parser perl module. You should be able to find this
at my website, or at CPAN. This script was developed with v 2.29,
but earlier versions should work.
The xmlboard.tar.gz package should include:
* xmlboard.cgi (the script)
* LICENSE (the GPL)
* README (this file)
* msg.dtd (doctype definition for all forums)
* sample/top.html (html for the sample forum)
* sample/bottom.html (html for the sample forum)
* sample/msg.xml (a mostly empty xml file for the sample forum)
* sample/msg.temp (an empty file for the sample forum)
* sample/pass.txt (the admin password file for the sample forum)
* sample/forum.conf (the configuration file for the sample forum)
First, you should extract this package to your downloads directory:
> tar zxvf xmlboard.tar.gz
> cd XMLBoard
Second, set the permissions on xmlboard.cgi so that it can be run as
an executable:
> chmod 755 xmlboard.cgi
Move the script into your cgi-bin directory:
> mv xmlboard.cgi /path/to/cgi-bin
Creating Your First Forum:
I recommend you keep a copy of the sample directory, and the files
inside it, for easy setup for new forums. You'll start by copying the sample
directory to the location and name you wish to use for your first forum:
> cp -r sample /path/to/your/forum
> cd /path/to/your/forum
This directory should be one in which you have write access (don't
try to put it in /etc). I recommend you set the permissions on the directory
so that all users can read and write to it. This will allow for the removal
of the temp file the script creates, although it is not necessary (read more
in the script configuration section). I also suggest that this directory be
not referred to by Apache, and especially that Apache's Indexes option is
off for this directory. This will keep your forum more secure and prevent
people from looking at pass.txt.
> chmod 777 .
Now, make sure that read and write permissions are set on the files
in your forum directory:
> chmod 666 *
You are now ready to configure the script for this forum.
Configuring the essentials of the script:
Go back to the cgi-bin directory...
> cd /path/to/cgi-bin
... and open xmlboard.cgi with an editor. Any editor that does not
automatically go around inserting line breaks for lines that wrap around
will do.
> emacs xmlboard.cgi
The first and most important thing you need to do is change the path
on the first line of code. It should say something like:
#!/usr/bin/perl
Set to the location of the perl binary on your computer. While the
above is a likely location, it might be at /usr/local/bin/perl, if you don't
quite know.
The second thing you need to do is set the %boards hash. Don't worry
if you don't know Perl; this is easy. In the section marked "VARIABLE
DECLARATION" near the top of the script, you should find a line that looks like
this:
%boards = ("sampleName", "sampleDir");
What you need to do is replace "sampleName" with the name of your new
forum. Use simple characters, with no spaces, but you can call this forum
whatever you wish. Replace "sampleDir" with the path of the directory you
have just created. This can be either relative to xmlboard.cgi or absolute
(beginning with /). I recommend an absolute path just to be safe. It
is very important that the directory name does NOT end with a slash (/). So
this line should now look something like this:
%boards = ("beer", "/home/jojo/beerboard");
You also need to set $dtdLocation. You should find a line in the
script that looks like this:
$dtdLocation = "msg.dtd";
You can move the file "msg.dtd" anywhere you like; it will be used
by every forum you set up. This must be an absolute location (beginning with
a slash (/)). So if you moved it into a directory called /home/jojo,
the line should look like this:
$dtdLocation = "/home/jojo/msg.dtd";
* Note: This variable is probably not essential to the workings of
the forum; the string is just placed in the "<!DOCTYPE" line of all "msg.xml"
files. The XML::Parser does not validate XML code so in theory it shouldn't
matter, but I suggest you fill in the correct location to be safe, and
have complete XML if you ever do decide to use "msg.xml" for something other
than this script.
Another variable you may need to change is the $leaveTemp variable,
which you will find in the forum's forum.conf file.
> cd /path/to/your/forum
> emacs forum.conf
If you were unable to set the permissions on your forum's directory
for read and write access to all (777), you need to set this variable to true
(1). This will keep the script from trying to remove the temp file and create
a new one when a user posts to the forum. The line should look like this:
$leaveTemp = 1;
Ok, you should now be set to use your new forum! Make sure your
webserver is running, and open up a new browser window. Enter the url:
http://myWebServerName/cgi-bin/xmlboard.cgi
You should see an attractive green page telling you that you have
specified an invalid board (because you have specified no board.) It should
show a list of the forums installed, and you should see your new forum in this
list. Click on it to go to your forum. That's it -- it's up and running.
For each new forum you add, simply repeat the above steps, making sure
to add the forum's name and directory to the %boards hash. So if you were
to add another forum, the line in the script would look like this:
%boards = ("beer", "/home/jojo/beerboard", "women", "/home/jojo/ladies");
Additionally, you will want to change the administrator's name and
password. The board comes with a preset name and password, "XMLBoard" and
"XMLBoard". To change the name and password, go into the admin action on the
board. If there is a link in the index to "Admin", click on that, otherwise
change "action=index" to "action=admin" in the URL. Remember, names and
passwords are case-sensitive. Enter the preset name and password, followed
by the new name and password you want. Passwords must be at least 4 letters
long, and names may not contain colons (:).
If you forget your name and password, simply remove the file
"pass.txt", and create a new one:
> rm pass.txt
> touch pass.txt
> chmod 666 pass.txt
This will set the name to "" and the password to "", so be quick about
running the admin action and changing them so as to not compromise security.
The Rest of the Configuration:
Please refer to the comments in "forum.conf" in your forum's directory
for further instructions on setup. The file should come with appropriate
default values.
Remember to maintain the proper Perl syntax -- variables begin with
"$", statements end with ";", and strings should be kept in quotes, preferably
single (').
Setting HTML Code (Forum Design):
In each forum's directory are two files, "top.html", and "bottom.html".
These are the standard HTML files that are used by the forums for their design.
Although the forum will work as is, you will want to either edit these files,
and include other HTML files, to go with the decor of your webpage. You should
know some basic HTML and stylesheets if you want to do this, however. (Some
nice tutorials can be found at http://hotwired.lycos.com/webmonkey/ .)
"top.html" contains all of the HTML after the <html>, <head>, and <title>
tags, and before the HTML for the script output (a message index or a post's
contents, etc.). "bottom.html" contains all of the HTML for the end of the
page, that follows the script output. It should include </body> and </html>
tags. So, if I wanted a page with a blue background, and a horizontal rule
before and after the script output, this would be the contents of top.html
and bottom.html:
top.html:
</head>
<body bgcolor="blue">
<hr>
and bottom.html:
<hr>
</body>
</html>
Using Includes:
Many larger websites are based on using server-side-includes. Each
page might have a general table layout, but the data for each section of the
page is read via includes. While XMLBoard does not support the parsing of
standard server-side include notation, and the execution of other CGI scripts,
it does have its own format for including HTML:
<?xmlboard include="/path/to/file.html"?>
When you edit a line in top.html or bottom.html to look like the
above, the script will parse this line for the location of the file specified
and print out the contents of the file. This directive must be located on a
line by itself for the parsing to work, and the location must be absolute.
So, here is an example of what your top.html and bottom.html files might look
like if you were using server side includes for the layout of your page:
top.html:
</head>
<body bgcolor="blue">
<table width="100%">
<tr>
<td colspan="2">
<?xmlboard include="/home/bean/http/topInclude.html"?>
</td>
</tr>
<tr>
<td>
<?xmlboard include="/home/bean/http/sidebar.html"?>
</td>
<td>
and bottom.html:
</td>
</tr>
<tr>
<td colspan="2">
<?xmlboard include="/home/bean/http/footer.html"?>
</td>
</tr>
</table>
</body>
</html>
That's it.
Administration of a Forum:
To delete posts or threads, or change your administration name and
password, use the XMLBoard's "admin" action. If there is a link in the index,
use that, otherwise put "action=admin" in the URL query string. It should
look like this:
http://domain/cgi-bin/xmlboard.cgi?board=forumname&action=admin
A page will appear, similar to a message index, but with radio buttons
next to each post. You may select one. Enter your name and password into the
field below, and the post you selected, and all posts which followup to that
post (in that post's thread) will be permanently deleted.
Another form is also displayed for the removal of old posts. Just
enter the maximum age of a post in the field (in seconds), enter your name
and password, and all posts that are older than the number of seconds specified
will be permanently deleted. However, followups to old posts will not be
deleted.
A third form is also shown below, for changing your name and password.
(See Configuring the essentials.)
Installing Without Root Access:
(Thanks to Paul Johnson for info on how to do this.)
If you don't have root access, and can't get your system administrator
to install the XML::Parser perl module, you can set it up in your home
directory by following these steps.
First, you need to download the module and untar/ungzip it. After
entering the directory it extracted to, enter these commands:
> mkdir ~/user_perl
> perl Makefile.PL PREFIX=~/user_perl
> make
> make install
Now enter the directory it installed to:
> cd ~/user_perl/lib/site_perl
> pwd
You should now see a full path of the directory to which the perl
module installed. If you haven't already followed the instructions for
installing XMLBoard, you should do so now. Now enter the directory to
which you installed the script and open it with an editor. On the second
line of the script, right beneath where it says #!/usr/bin/perl, type
in the following:
use lib '/path/to/your/directory/user_perl/lib/site_perl';
and insert the path you saw after typing "pwd" in the console in
between the single quotes. Once you finish setting up your forums, XMLBoard
should be good to go. If it still doesn't work after following these steps,
you should probably go with XMLBoard Solo, which doesn't require Expat or
XML::Parser.
For Inquiring Minds:
If you have no interest in modifying the code to the XMLBoard, good
for you, the board perfect as it is ;) But if you do wish to edit the code
or simply want to know how the board works, here's how. This is a general
description of the flow of the program; you should look at the comments in
the code for specifics.
Basically, the top section defines all the global variables that
configure the script. Following is a large section of subroutines, which are
called from the section at the bottom of the script. Every time the script
runs, it needs to know what forum it is dealing with. This value is
found in $FORM{'board'}. All user-inputted values are placed in the hash,
%FORM. $FORM{'board'} is checked against the list in the hash, %boards, which
is defined in setup. If it does not exist in the list, a page out printed
listing the existing boards and the script exits. The next value that is
checked is $FORM{'action'}. If $FORM{'action'} is not a valid action, it is
set to "index". If it equals "changepw", the script checks the name and
password, and changes the password or not accordingly. If it equals
"searchpage", the script prints a form for searching. These are the only
actions that do not involve the parsing of XML, and they exit the script
at this point.
If $FORM{'action'} is "admin", $FORM{'action'} is set to "index",
but will set a variable to note that the user requested "admin". The
script then sets up the parser, using functions from the XML::Parser module,
for the action requested. When the parser is set up, it associates different
elements it encounters with a handler (a subroutine). So when the parser is
going through an XML document, and it sees a <post> tag, it calls a start-tag
handler subroutine. When it sees a </post> tag, it calls an end-tag handler
subroutine. And when it sees character data, it calls a character-data
handler.
The script then sets up the %wantedTags hash for the action requested.
This hash contains all the elements the script is looking for when it is
parsing XML. By referencing the element it finds when parsing against this
list of elements wanted, the script can decide what to do with the data
it collects -- scrap it or store it and print something.
If $useCookies is true, the script gets all the cookies for the forum
requested and puts the names and values into the %COOKIES hash.
If $FORM{'action'} is "index", the file for parsing is set to msg.xml,
in the forum's directory, and the parsing begins. To simplify: if the parser
encounters character data and it is within a tag that is found in %wantedTags,
the data is added to the %POST hash. If all of the wanted tags are found,
the script will then print out the post in the index. The index is either
organized as a table, with one row per post, or as an unordered list (this is
defined in setup). If it is using a table, the post is indented depending on
the current depth of the parse (how many elements it is in), by using .
If it is using an unordered list, it uses nested <ul>'s to indent.
If $FORM{'action'} is "read", it parses through but only saves data
to the %POST hash if the parser has run into a <post> tag that has an id equal
to that of $FORM{'id'}. Otherwise, it adds the current post's id to an array
(@idList). The id is placed in the number corresponding with the depth of the
parse, and when the post requested is found, the id with the depth immediately
above that of the requested post is saved as $PREVPOST{'id'}. This is how
the script determines what the post is in reply to. The script then finishes
parsing and calls the "read" parse again, but this time looking for a post
with an id that equals $PREVPOST{'id'}. Now, it has info in %POST and
%PREVPOST, and this info is printed out. The script then calls the "index"
action, but sets a value for a parent element ($baseId). An index of
responses to the post is printed out.
If $FORM{'action'} is "write", the script makes sure that the user
has provided all the data needed for a post. If not, an error message is
printed; otherwise msg.xml is copied to msg.temp and it begins parsing
msg.temp. The contents of msg.temp are copied verbatim to msg.xml, but when
the parser finds the place where the post should be put, this is also
printed to msg.xml, and the parse finishes. A message is then printed,
thanking the user for their kindly contribution to the forum.
If $FORM{'action'} is "remove", it works the same way as the "write"
action does, but when the parser finds the thread or post to remove, it simply
neglects to copy this information to msg.xml. When it finds the post selected,
it records the depth at which this post was found, and does not print any
more xml until this depth is reached again. This ends up removing the post
and all responses to the post.
If $FORM{'action'} is "removeold", the script runs similarly to the
"index" action. It stores all the info it collects from msg.temp in the
%POST hash, and when it reaches an end element it checks to see if all the
tags in %wantedTags have been collected. If they have, it checks the date
of the post to see if it is older than the administrator has specified, and
if it is an old post it is not printed. The script remembers that a post
has been skipped at the current depth, and then continues parsing for more
posts. If it finds all the info for a post that is not old, it prints the
info out to msg.xml, all except for the </post> tag. When the parser
encounters a </post> tag, it checks to see if a post has been skipped at
the current depth, and does not print the tag if one has, and then forgets
that a post has been skipped. But if it has not been skipped at the current
depth, it prints the </post> tag.
If $FORM{'action'} is "search", the script gathers the info about all
posts like with the index, and when it has all the needed info, it globs
together the fields the user requested for searching and runs a regular
expression on it to see if the terms the user specified match the string. If
it matches, a link is printed in the results list.
That is the basic flow of the program; for info on specific aspects
and subroutines, look at the comments in the code.
Fin:
I hope you enjoy the XMLBoard script. If you have suggestions,
questions, comments, bug-reports, etc., please check
http://xmlboard.hypermart.net or http://beanpole.cjb.net to see if there
are any appropriate forums up. I will check them frequently when I have time
(which I frequently have), so you're most likely to get a question answered
there without cluttering my inbox. But if you must, email me at
[EMAIL PROTECTED]
- Joel Rosen
Want to buy your Pack or Services from MandrakeSoft?
Go to http://www.mandrakestore.com
