#!/home/rebelsky/perl/bin/perl # /home/rebelsky/public_html/Blazers/Annotations/Summer1999/Insert.pm # Sarah Luebke # converted from insert.cgi to Insert.pm to work # with Ravel 07/15/99 # # This script gets a URL, html, name, email and group info from # Plugin.pm. Then it reads information from # /home/rebelsky/public_html/Blazers/Annotations/Stored_Annotations/index.txt # and determines if annotations exist for that URL. # If there are annotations for that document, find the directory containing # those annotations and search the document for the correct placement # of the annotations. Then insert the annotation flags (small triangles) # and the Javascript code to open a window. # # INPUT: URL, html name, email and group info from Plugin.pm # # OUTPUT: returns the page with the proper # javascript and annotation flags added (and prints it out). #Package ID package Insert; use Exporter (); @ISA = (Exporter); @EXPORT = qw(main); use LWP::Simple; use HTML::Parse; use URI::Escape; use AFile; use Search; use History; # call the subroutines #&main; # extractInfo is called in main and # insertAnnotation is called in extractInfo sub main { ########################################################################## # This subroutine takes url, html and user info and calls # extractInto to do the rest. # INPUT: url, html, name of user, email of user, a list of # the user's groups # OUTPUT: The new html page ########################################################################## #VARIABLES my $url = shift; my $html = shift; my $name = shift; #the users name my $email = shift; #the email address of the user my @groups = shift; #groups that user belongs to my $first; #boolean: first group in list my $group; #each group individually my $lasttime; #last time that this page was accessed as returned by History->history #update the history and get the last time that the page was accessed by #this user. $lasttime = history($name,$url); #read in the groups....added by Rachel Heck on 7/12/99 $first = 1; foreach $group (@groups) { if ($first == 1) { $groups = $group; $first = 0; } #if ($first == 1) else { $groups = "$groups,$group"; } #else } #foreach $group (param("groups") ################################################################ # Call extractInfo with $url and $html &extractInfo($url,$html,$name,$email,$groups,$lasttime); } # main sub extractInfo { ########################################################################## # This subroutine checks if the page has been annotated by looking in # # index.txt. If it has, then extract the annotation info from the # # files in the appropriate directory. # ########################################################################## #VARIABLES my $url = shift; my $html = shift; my $name = shift; #the users name my $email_user = shift; #the users email addres my $groups = shift; #the groups that the user is a member my $lasttime = shift; #the last time this url was accessed my $dir; my @stored; my $author; #author of annotation my $email; #email of the author of the annotation my $date; #date and time of the annotation my $title; #Title of the annotation my $annotated_text; #text that was annotated - the flags #go on either side of this my $annotation; #text of annotation my $context; #context of the annotation my $times; #number of times annotated_text exists my $protection; #boolean of whether the user has #permission to view the annotation my $permission; my $time; #the time that the annotation was made my @user_groups; my @annotation_groups; my $file; local (*fields); # Open index.txt and check if the URL is there. If it is, get the name # of the directory associated with it. # readIndex is from Search.pm. It takes a url and returns the directory # associated with that url. $dir = readIndex($url); # Now look at each file in the directory (only # the .txt files) if ($dir ne 0) { foreach $file (<$dir\/*.txt>) { #call readFile from the AFile module. This will parse the annotation file. readFile(*fields,$file); #store the results from parsing the file $author = $fields{'author'}; $email = $fields{'email'}; $date = $fields{'date'}; $time = $fields{'time'}; $title = $fields{'title'}; $annotated_text = $fields{'annotated_text'}; $annotation = $fields{'annotation'}; $context = $fields{'context'}; $times = $fields{'times'}; # number of times the annotated_text occurs $protection = $fields{'protection'}; #whether it's public, private or groups #strip extra white-space characters from the beginning and ending of #variables @stored = ($author,$email,$date,$time,$title,$annotated_text,$annotation,$context,$times,$protection); foreach $thing (@stored) { # Strip extra white-space characters from the beginning and end. # Written with help from Sam $thing =~ s/^\s*//; $thing =~ s/\s*$//; } #foreach # extract the author's name, email, date and title, annotated_text, # annotation, context and times $author = $stored[0]; $email = $stored[1]; $date = $stored[2]; $time = $stored[3]; $title = $stored[4]; $annotated_text = $stored[5]; $annotation = $stored[6]; $context = $stored[7]; $times = $stored[8]; $protection = $stored[9]; # Call insertAnnotation with $html, $author, $email, $date, $time, # $title, $annotated_text, $annotation, $context, $times, $file, # $url, and $lasttime # Check if the user has permissions. Taken from TreeCreation.pm, # written by Rachel Heck. #remove extra spaces $fields{'protection'} =~ s/^(\s*)//i; $fields{'protection'} =~ s/(\s*)$//i; # Check if the user is the author, or the sysadmin if (($fields{'author'} eq $name) || ("sysadmin" eq $name)) { $permission = 1; } elsif ($fields{'protection'} eq "private") { $permission = 0; } elsif ($fields{'protection'} eq "public") { $permission = 1; } else { $permission = 0; @user_groups = split(/\,/,$groups); @annotation_groups = split(/\,/,$fields{'protection'}); foreach $annotation_group (@annotation_groups) { foreach $user_group (@user_groups) { if ($annotation_group eq $user_group) { $permission = 1; } } } } if ($permission == 1) { $html = &insertAnnotation($html,$author,$email,$date,$time,$title,$annotated_text,$annotation,$context,$times,$file,$url,$name,$email_user,$groups,$lasttime); } # if (permission == 1) } #foreach $file (<$dir\/*.txt>) } # Add a toolbar at the top of the page with the code for calling the add page my $disclaimer = "
This page was created as part of a research project with Samuel Rebelsky, Rachel Heck and Sarah Luebke.
"; # call load_add.cgi and give it the URL, name, users's email and groups my $add_button = "\"Add<\/a>"; # call search.cgi and give it the URL, name, users's email and groups my $search_button = "\"Search<\/a>"; #my $help_button = ""; my $help_button = "\"Help<\/a>"; my $red_arrow = ""; my $green_arrow = ""; my $yellow_arrow = ""; # We don't need the arrow key anymore since it is on the help page #my $arrow_key = "$red_arrow = New annotation since last visit
$green_arrow = New reply since last visit
$yellow_arrow = Annotation and all replies added prior to last visit"; if ($html =~ //i) { $html =~ s/()/$1$add_button$search_button$help_button
\n/i; } #if #add javascript script for opening a new window into the head of the document my $java = "\n\n"; # add javascript for opening a new window for the search my $java_search = "\n\n"; # add javascript for opening a new window for the help page my $java_help = "\n\n"; if ($html =~ //i) { #insert the javascripts $html =~ s/()/$1$java \n $java_search \n $java_help/i; } # if elsif ($html =~ //i) { $html =~ s/()/$1$java \n $java_search \n $java_help<\/head>\n/i; } # elsif else { $html = "$java \n $java_search \n $java_help<\/head>\n" . $html; } # else # We're done, so return the new document return "$html"; } # extractInfo() sub insertAnnotation { ######################################################################## # This subroutine searches for the annotated text and inserts # # the annotation markers around it, along with the javascript # # to open a new window displaying the annotation. # # It calls textToSearch in the Search.pm module ######################################################################## # VARIABLES my $html = shift; my $author = shift; my $email = shift; my $date = shift; my $time = shift; #the time that the annotation was made my $title = shift; my $annotated_text = shift; my $annotation = shift; my $context = shift; my $times = shift; my $file = shift; my $url = shift; my $name = shift; #the users name my $email_user = shift; #the email address of the user my $groups = shift; #the groups that the user is a part of my $lasttime = shift; #the last time that this page was accessed my $found = 0; my $new_html; my $before; #used to hold the stuff matched before the context my $after; #stuff matched after the context my $stuff; # the context that was matched my $temp_before; # stuff matched before the annotated_text my $temp_string; # the annotated_text that was matched my $temp_after; # stuff after the annotated_text my $new; #boolean: is the annotation new since last viewing my $forward_arrow; my $back_arrow; #determine whether this is an annotation that the user has not seen $new = lessThan($lasttime,substr($date,0,2) . "/" . substr($date,2,2) . "/" . substr($date,4,2) . "\|" . $time); #use the appropriate arrow if (($new == 1) || ($new eq "equal")) { $forward_arrow = "\"""; $back_arrow = "\"""; }#if (($new == 1) || ($new eq "equal")) else { $new = replyTest($lasttime,$file,$name,$groups); if ($new == 1) { $forward_arrow = "\"""; $back_arrow = "\"""; }#if ($new == 1) else { $forward_arrow = "\"""; $back_arrow = "\"""; }#else }#else # Forward_arrow and back_arrow are icons to mark the annotations # and add the code to open a new window when clicked $forward_arrow = "$forward_arrow<\/a>"; $back_arrow = "$back_arrow<\/a>"; # Call the textToSearch function from the Search module # to slashify, change all white-spaces to account # for HTML tags or any number of white-spaces and # add possible tags around parens and such. $annotated_text = textToSearch($annotated_text); $context = textToSearch($context); $context =~ s/

/

(<.\*>)\*/gi; #ACK! I put this in because I had # "

Text" that wasn't matching and a general rule was too slow # Search for $context in the html. If it is found, # add the links and javascript for displaying the annotation. # FIX: Right now it is adding the annotate flags around the # first instance of $annotated_text if the context is found! if ($html =~ m/($context)/i) { # save the parts of the document $before = "$`"; #everything before the context $after = "$'"; # everything after the context $stuff = "$1"; # the context that was matched # If the annotated text is matched, with a link around it, # keep the tag as part of the annotated text, unless # it is a javascript tag that I added. #if ($stuff =~ m/((?:\s*)+(?:[^<]*))($annotated_text(?:\s|.)*?(?:<\/a>)+)/i) if ($stuff =~ m/(\s*)+($annotated_text(?:\s|[^>])*(?:<\/a>)+)/is) { $found = 1; $temp_before = "$`"; # everything before the $temp_string = "$1" . "$2"; # the annotated text, including $temp_after = "$3" . "$'"; # everything after } #if ($stuff =~ m/(\s*$annotated_text\s*<\/a>)/i) # If there isn't a link, match the annotated_text. # Also make sure it isn't inside a tag (taken from Rachel's code) elsif ($stuff =~ m/(>|(?:>[^<]*[^a-zA-Z_0-9<])|\W)($annotated_text)/is) { $found = 1; $temp_before = "$`" . "$1"; # stuff before the annotated_text $temp_string = "$2"; # the annotated_text $temp_after = "$'"; # stuff after } #elsif elsif ($stuff =~ m/($annotated_text)/is) { # In this case the annotated_text was only matched in a tag print "Matched just annotated_text\n"; $found = 1; $temp_before = "$`"; $temp_string = "$1"; $temp_after = "$'"; }#elsif if ($found == 1) { # add the annotation flags around the annotated_text $temp_string = "$forward_arrow" . $temp_string . "$back_arrow"; # put the context string back together $stuff = "$temp_before" . "$temp_string" . "$temp_after"; # put the html document back together $html = "$before" . "$stuff" . "$after"; # we're done! $new_html = $html; } else { print "Annotated_text not found: $annotated_text\n"; } } # if ($html =~ m/$context/i) { else { # If the context wasn't matched, # look for the annotated text anyway. # WE SHOULD BE LOOKING FOR THE INSTANCE OF THE TEXT! if ($html =~ m/($annotated_text)/i) { $found = "-1"; # This is an exception because the context wasn't matched $before = "$`"; #everything before the annotated text $after = "$'"; # everything after the text $stuff = "$1"; # the text that was matched # If the annotated text is matched, with a link around it, # keep the tag as part of the annotated text if ($stuff =~ m/(\s*)+($annotated_text\s*(?:<\/a>)+)/i) { # we have a match, with a link in front $temp_before = "$`"; $temp_string = "$1"; $temp_after = "$'"; } #if # If there isn't a link, match the annotated_text. # Also make sure it isn't inside a tag (taken from Rachel's code) #elsif ($stuff =~ m/(>|(?:>[^<]*[^a-zA-Z_0-9<]))((?:$annotated_text)(?:s\*)?)/i) #this new line was modeled after line 165 of Ambiguity.pm elsif ($stuff =~ m/((?:\W$annotated_text)(?:s\*)?)(?:<|[^a-zA-Z_0-9<>][^>]*?<)/si) { $temp_before = "$`" . "$1"; $temp_string = "$2"; $temp_after = "$'"; }#elsif else { # the annotated text was only matched w/in a tag print "error! "; #print "$annotated_text was only matched in a tag
\n"; $temp_string = $stuff; } #else # add the annotation flags around the annotated_text $temp_string = "$forward_arrow" . $temp_string . "$back_arrow"; # rejoin the "stuff annotated_text stuff" $stuff = "$temp_before" . "$temp_string" . "$temp_after"; # rejoin the "stuff context stuff" $html = "$before" . "$stuff" . "$after"; # we're done $new_html = $html; } #if ($html =~ m/($annotated_text)/i) } #else (the context wasn't matched) if ($found == 0) { print "Not Found
\n"; #ERROR: What should we do if an annotation cannot be inserted? #print "annotated_text: $annotated_text
\n\n"; #print "Context: $context\n\n"; return $html; } # if ($found == 0) elsif ($found == -1) { print "Context wasn't matched for: $context
\n"; } #return the edited source code return $new_html; } # insertAnnotation() ###################################################################### # Pod Documentation # ##################### #=head1 Name #Insert #=over 4 #=item main #This subroutine gets the gets the URL, name, email and group info #from the form geturl.html, calls unescape #from the URI package, gets the HTML and calls extractInfo. #=item extractInfo #This subroutine takes the url and html page. It checks if the page already #has annotations by looking in index.txt. If it does, it goes takes the #info from each annotation file in the appropriate directory. #=item insertAnnotation #This subroutine takes the html, author, email, date, title, annotated_text, #annotation, context and file from extractInfo. It searches for the #annotated text and inserts the annotation markers around it, #along with the javascript to open a new window displaying the annotation. #=back #=head 1 History #=over 4 #=item [09 June 99] #Began coding. #=item [11 June 99] #Edited by Rachel Heck. #=item [14 June 99] #Now works correctly. #=item [29 June 99] # Added call to textToSearch to slashify special characters. #=item [6 July 99] # Cleaned up a little and added comments (e.g., commented closing braces). #=item [7 July 99] # Look for context when matching annotation (it was matching # the annotated_text before). #=item [12 July 1999] # Edited by Rachel Heck....program now gets and passes the groups that the user # belongs to #=item [13 July 1999] # Checks that user has permissions to read the file before # inserting annotation flags around the annotated text. #=item [14 July 1999] # Fixed errors: If the annotated text is a link, insert the flags # outside of the link. #=item [15 July 1999] # Now receives the URL, html, name, email, and groups as # parameters (from Plugin.pm) instead of a form. #=item [16 July 1999] # Works with Ravel. #=item [20 July 1999] #Editted by Rachel Heck to check whether annotations are new or # added previous to the last visit to the page. #=item [23 July 1999] #When an annotation or search window opens, it will pop to the front. =item [4 August 1999] All cgi calls are now Proxy URLs. The name, email, and groups information are no longer being passed to cgi scripts. #=back #=cut