Gossamer Forum
Home : Products : Links 2.0 : Installation -- Unix :

db.pl ....line 28

Quote Reply
db.pl ....line 28
I'm back! I know you've seen this 1000 times before (believe me, I've read loads of similar queries using the search engine but this is unique, possibly :-)) but I get the following error when trying to run admin.cgi

Error including libraries: Can't locate /home/username/freeque.port5.com/cgi-bin/links/admin/db.pl in @INC (@INC contains: /host/f/r/e/p/o/r/freeque.port5.com/cgi-bin/links/admin /usr/local/lib/perl5/5.00503/aix /usr/local/lib/perl5/5.00503 /usr/local/lib/perl5/site_perl/5.005/aix /usr/local/lib/perl5/site_perl/5.005 .) at /host/f/r/e/p/o/r/freeque.port5.com/cgi-bin/links/admin/admin.cgi line 28.

Make sure they exist, permissions are set properly, and paths are set correctly.

Now, I knwo you're going to say "set the absolute path coorectly " or some such, but unless I'm being absolutely dopey (which is more than possible) then I think I've done that. After reading the faq/support forum posts for my web server I know for a fact (cause the webmaster explicitly said so) that the absolute path is home/username . The error message seems to indicate that the path is /host/f/r/e/p/o/r/ and, despite being told "The correct path is /home/username as outlined in the FAQ, any other path ie /hosts/w/w/w/etc could and probably will change without warning. Therefore the only path you should use is /home/username" I tried using the aforementioned path, to no effect.

I have no telnet access so cannot find out my absolute path that way, although I have been directed to various scripts which will do it for you. However, I am wary of these due to the above quotation from the webmaster/administrator who advises only to use home/username as the absolute path.

I will be grateful if anyone can point out how much of an idiot I'm being and show me what I've done wrong. Just to help, my lings.cgf file is below (to make a lengthy post even longer):

# Paths and URL's to Important Stuff
# --------------------------------------------------------
# PATH and URL of Admin CGI directory. No Trailing Slash.
$db_script_path = "/home/username/freeque.port5.com/cgi-bin/links/admin";
$db_dir_url = "http://freeque.port5.com/cgi-bin/links/admin";

# URL of User CGI directory. No Trailing Slash.
$db_cgi_url = "http://freeque.port5.com/cgi-bin/links";

# PATH and URL of Pages to be built. No Trailing Slash.
$build_root_path = "/home/username/freeque.port5.com";
$build_root_url = "http://freeque.port5.com";

# PATH and URL of What's New page. No Trailing slash.
$build_new_path = "$build_root_path/New";
$build_new_url = "$build_root_url/New";

# PATH and URL of What's Cool page. No Trailing slash.
$build_cool_path = "$build_root_path/Cool";
$build_cool_url = "$build_root_url/Cool";

# PATH and URL of What's Rating page. No Trailing slash.
$build_ratings_path = "$build_root_path/Ratings";
$build_ratings_url = "$build_root_url/Ratings";

# Database Options
# --------------------------------------------------------
# You really need to use flock(). However if your system doesn't support it (Win95),
# then turn this off at your own risk!
$db_use_flock = 1;
# Use Pragma: no-cache to stop browsers from caching output? (0 = cache, 1 = nocache)
$db_nocache = 0;
# Use IIS headers? Set this to 1 if your web server needs the full headers (HTTP/1.0 etc.)
$db_iis = 0;
# Field number of category in category.db
$db_main_category = 1;
# Referers -- which hosts are allowed to add to your database.
@db_referers = ();

# Email Options
# --------------------------------------------------------
# NOTE: Be sure to edit the text of the message. It can be found
# in the templates directory: email-add.txt, email-del.txt and
# email-mod.txt respectively.

# Set only ONE of either $db_mail_path which should go to a
# mail program like sendmail, or $db_smtp_server which will use
# a SMTP server. Do not set both at once!

# PATH of sendmail. # SMTP Server
$db_mail_path = '/usr/lib/sendmail'; $db_smtp_server = '';

# Email address of the administrator. BE SURE TO SET!
$db_admin_email = '';

# Notify visitors automatically when their links are validated?
$db_email_add = 0;

# Notify visitors automatically when their links are modified?
$db_email_modify = 0;

# Log all outgoing messages? Put the full path to a logfile and make sure it
# exists and is writeable (chmod 666 -rw-rw-rw-). This logfile can get pretty
# big!
$db_mailer_log = '';

# Build Options
# --------------------------------------------------------
# Setting for popular cutoff, can either be a percent (i.e. top 3%), or
# a fixed value (i.e. top 10). Enter 0.03 or 10 respectively.
$db_popular_cutoff = 0.03; # Top 3%
# Number of days a link is considered New.
$db_new_cutoff = 14; # 14 days old.
# Return search results bolded (1 = yes, 0 = no).
$search_bold = 1;
# The name of an index file.
$build_index = "index.html";
# The extension to use for non index files:
$build_extension = ".html";

# Should your add.cgi automatically select what category the user
# is adding his link into?
$db_single_category = 1;

# Directory Permissions. Do NOT put quotes around it. Set to 0777
# for world writeable. If you put this at 0755, you might need a
# script to remove any directories created as they will be owned
# by the web server or user nobody. Better solution might be to
# run the nph-build routine from the command shell.
$build_dir_per = 0777;

# The script can now break up category pages that have more then x
# number of links in it. It will also create next and previous links
# to help navigate. Set this to 1 if you want to break up pages or 0 if
# you want to have only one page per category.
# Setting this to one, will also break up the What's New pages into
# per day sections (not affected by number of links per page).
$build_span_pages = 1;

# Number of links per page.
$build_links_per_page = 25;

# Detailed View: The script can build a single html page per link. This is
# great if you have a review in your database. To enable, you must set
# the directory where all the pages will be stored, the url to that directory
# and set the enable option.
$build_detailed = 0;
$build_detail_path = "$build_root_path/Detailed";
$build_detail_url = "$build_root_url/Detailed";

# Use Templates: instead of editing site_html.pl, you can edit regular html pages
# encoded with special <%links_tags%>. It's not as flexible as site_html.pl, but it
# may be easier.
$build_use_templates = 1;

# Site Title, the default title to use. You should really spend some time customizing
# the directory though!
$build_site_title = 'Gossamer Threads Links';

# Date Routines
# --------------------------------------------------------
# Your date format can be whatever you like, as long as the following
# two functions are defined &date_to_unix and &unix_to_date:
# The default is dd-mmm-yyyy.

sub date_to_unix {
# --------------------------------------------------------
# This routine must take your date format and return the time a la UNIX time().
# Some things to be careful about..
# timelocal does not like to be in array context, don't do my($time) = timelocal (..)
# int your values just in case to remove spaces, etc.
# catch the fatal error timelocal will generate if you have a bad date..
# don't forget that the month is indexed from 0!
my $date = shift; my $i;
my %months = map { $_ => $i++ } qw!Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec!;
my ($day, $mon, $year) = split(/-/, $date);

exists $months{$mon} or return undef;
$day = int($day); $year = $year - 1900;

require Time::Local;
my $time = 0;
eval {
$time = &Time::Local::timelocal(0,0,0, $day, $months{$mon}, $year);
if ($@) { die "invalid date format: $date - parsed as (day: $day, month: $months{$mon}, year: $year). Reason: $@"; }
return $time;

sub unix_to_date {
# --------------------------------------------------------
# This routine must take a unix time and return your date format
# A much simpler routine, just make sure your format isn't so complex that
# you can't get it back into unix time.
my $time = shift;
my ($sec, $min, $hour, $day, $mon, $year, $dweek, $dyear, $tz) = localtime $time;
my @months = qw!Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec!;
$year = $year + 1900;
return "$day-$months[$mon]-$year";

sub long_date {
# --------------------------------------------------------
# This routine is for printing a nicer date format on the what's new page. It should
# take in a date in your current format and return a new one.
my $time = shift;
$time = &date_to_unix ($time);
my ($sec, $min, $hour, $day, $mon, $year, $dweek, $dyear, $tz) = localtime $time;
my @months = qw!January February March April May June July August September October November December!;
my @days = qw!Sunday Monday Tuesday Wednesday Thursday Friday Saturday!;
$year = $year + 1900;
return "$days[$dweek], $months[$mon] $day $year";

# -------------------------------------------------------------
# Extra Paths -- unless you feel the need to rename files, you
# can leave this as is.
# -------------------------------------------------------------
$db_script_url = $db_dir_url . "/admin.cgi"; # Admin script.
$db_header_path = $db_script_path . "/header"; # Place to store header and footers.
$db_footer_path = $db_script_path . "/footer";
$build_search_url = $db_cgi_url . "/search.cgi"; # URL of search script.
$build_add_url = $db_cgi_url . "/add.cgi"; # URL of add script.
$build_modify_url = $db_cgi_url . "/modify.cgi"; # URL of modify script.
$build_jump_url = $db_cgi_url . "/jump.cgi"; # URL of jump script.
$build_email_url = $db_cgi_url . "/subscribe.cgi"; # URL of email update script.
$build_rate_url = $db_cgi_url . "/rate.cgi"; # URL of rate script.
$db_mail_url = $db_dir_url . "/nph-email.cgi"; # URL of admin mass mail script.
$build_css_url = $build_root_url . "/links.css"; # URL to your CSS file.

$db_lib_path = $db_script_path; # PATH of library files.
$db_links_name = "$db_script_path/data/links.db"; # PATH to links database.
$db_category_name = "$db_script_path/data/categories.db"; # PATH to category database.
$db_valid_name = "$db_script_path/data/validate.db"; # PATH to validation database.
$db_modified_name = "$db_script_path/data/modified.db"; # PATH to modification database.
$db_url_name = "$db_script_path/data/url.db"; # PATH to URL lookup database.
$db_email_name = "$db_script_path/data/email.db"; # PATH to email database.
$db_links_id_file_name = "$db_script_path/data/linksid.txt"; # PATH to links counter file.
$db_category_id_file_name = "$db_script_path/data/categoryid.txt"; # PATH to category counter file.
$db_hits_path = "$db_script_path/data/hits"; # PATH to hits directory.
$db_template_path = "$db_script_path/templates"; # PATH to templates.
$db_rates_path = "$db_script_path/data/rates"; # PATH to ratings.

Also, here is my admin.cgi file. I know you would have said to set the require "/links.cgi" to the absolute path, but I tried that, it didn't work and so I followed a bit of advice (which worked for someone else) which said to add a fullstop/period/dot before require "./links.cgi" (again to no effect). Anyways, here's my admin.cgi file for you to analyse (has anyone told you you all do a great job?)

# Required Librariers
# --------------------------------------------------------
eval {
($0 =~ m,(.*)/[^/]+,) && unshift (@INC, "$1"); # Get the script location: UNIX /
($0 =~ m,(.*)\\[^\\]+,) && unshift (@INC, "$1"); # Get the script location: Windows \

require 5.001; # Make sure we have at least perl 5.001.
require "./links.cfg"; # Change this to full path to links.cfg if you have problems.
require "$db_lib_path/db.pl"; # Database Routines.
require "$db_lib_path/db_utils.pl"; # Database Support utilities.
require "$db_lib_path/admin_html.pl"; # Admin HTML routines.

if ($@) {
print "Content-type: text/plain\n\n";
print "Error including libraries: $@\n";
print "Make sure they exist, permissions are set properly, and paths are set correctly.";

# ========================================================
eval { &main; }; # Trap any fatal errors so the program hopefully
if ($@) { &cgierr("fatal error: $@"); } # never produces that nasty 500 server error page.
exit; # There are only two exit calls in the script, here and in in &cgierr.
# ========================================================

sub main {
# --------------------------------------------------------
$| = 1; # Flush Output Right Away

# Main Menu. Check to see what the user requested, then if he has permission for that
# request, do it. Otherwise send the user off to an unauthorized request page.
%in = &parse_form; # Get form input so we know which database to load.

# Load the database definition file and set the link url.
$in{'db'} ?
require "$db_lib_path/$in{'db'}.def" :
require "$db_lib_path/links.def";

$db_script_link_url = "$db_script_url?db=$in{'db'}";

# The functions beginning with &html_ can be found in admin_html.pl, while the other
# functions can be found in db.pl

if ($in{'add_form'}) { &html_add_form; } # Display the Add Record Form.
elsif ($in{'add_record'}) { &add_record; } # Add the Actual Record.
elsif ($in{'add_record_mult'}) { &add_record_mult; } # Add Multiple Records at once (beta).
elsif ($in{'view_search'}) { &html_view_search; } # Display form to search database.
elsif ($in{'view_records'}) { &view_records; } # Search database and print results.
elsif ($in{'delete_search'}) { &html_delete_search; } # Display the form to search for records to delete.
elsif ($in{'delete_form'}) { &html_delete_form; } # Display the form to pick records to delete.
elsif ($in{'delete_records'}) { &delete_records; } # Delete records.
elsif ($in{'modify_search'}) { &html_modify_search; } # Display the form to search for records to modify.
elsif ($in{'modify_mult_form'}) { &html_modify_mult_form; } # Display multiple records to modify at one time.
elsif ($in{'modify_form'}) { &html_modify_form; } # Display the form to pick record to modify.
elsif ($in{'modify_form_record'}) { &html_modify_form_record; } # Display the form to modify a record.
elsif ($in{'modify_record'}) { &modify_record; } # Modify the record.
elsif ($in{'modify_mult_record'}) { &modify_mult_record; } # Modify multiple records at once.
elsif ($in{'validate_form'}) { &html_validate_form; } # Display the Validation Form.
elsif ($in{'validate_records'}) { &validate_records; } # Validate records.
elsif ($in{'check_links'}) { &check_links; } # Display/Check to make sure links have matching catgories.
elsif ($in{'fix_links'}) { &fix_links; } # Fixes up any errors found in check links.
elsif ($in{'check_duplicates'}) { &check_duplicates; } # Display/Check for duplicate links.
elsif ($in{'html_mail_target'}) { &html_mail_target; } # Form to do targeted mass mailing.
elsif ($in{'html_mail_form'}) { &html_mail_form; } # Form to do mass mailing.
elsif ($in{'html_mail_update'}) { &html_mail_update; } # Form to do email updates.
elsif ($in{'html_edit_template'}) { &html_edit_template; } # Edit Templates
elsif ($in{'save_template'}) { &save_template; } # Save templates.
elsif ($in{'html_template_help'}) { &html_template_help; } # Template Help.
elsif ($in{'display'} eq 'navigation') { &html_navigation; } # Display HTML Header
elsif ($in{'display'} eq 'body') { &html_body; } # Display HTML Body
elsif ($in{'display'} eq 'top') { &html_top; } # Display HTML Top
else { &html_home; } # Display Frame Index page.

# &cgierr("Done"); # Uncomment this line for Debugging... Will tack on form variables and environment variables
# to the end of every page. Quite Useful.

Quote Reply
Re: db.pl ....line 28 In reply to
Sounds like you have a little problem with permissions. Have another look at the readme and make sure you have all of your permissions set correctly. By the way you sound, you might not be too sure of your full system path. Download the script at http://dotmatrix.net/scripts/setup/ and give it a shot to see if your paths are in fact correct.

Good luck!

Quote Reply
Re: db.pl ....line 28 In reply to
The absolute path you need to set is in the admin.cgi for the links.cfg file:


require "/home/username/freeque.port5.com/cgi-bin/links/admin/links.cfg";

If your script cannot find the links.cfg initially, it will not be able to find the location of the $db_lib_path directory, and thus you will receive the error message that the script cannot find the db.pl file.



Quote Reply
Re: db.pl ....line 28 In reply to
Thanks for all your help. I have learned so much installing this script.

It turns out that my absolute path was entirely different to the one I perceived it to be. Several other people had been having problems finding out the absolute path and with a bit of trial and error we finally figured out what it was.

Thanks for all your help, but I'm sure I won't be gone for long. I've got loads of questions, but I'll just have to see if I can figure them out on my own first.



Quote Reply
Re: db.pl ....line 28 In reply to
>>Download the script at http://dotmatrix.net/scripts/setup/ and give it a shot to see if your
paths are in fact correct.

Yippee!!! This just saved my night!! Thank you for posting and saving me hours of trying to figure this out. My path was *not* anything like I thought it was.