#!/home/rebelsky/perl/bin/perl ##########NEW BLOCKIFY #########VERSION 0.05, pre-vivek-blockify #%%%%%%%%%%%%%%%%%%%%%%%%%%%% ############# BUG LIST ###################### #$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ #$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ ### Jared Seaman Last Updated: 7/12/99 ### A program for doing some serious blockification of ### files to be edited... ### Begun 7/9/99 ############################################## #=pod # =head1 blokify.pl # This should \do some quick-n-dirty blockification of to-be-edited # files; i.e. it will \read stuff in, BLOK at a \time [which is to say, up to # an HTML tag], & should import \some hardcoded rule-predicessors # to apply to BLOKs.... # =cut #This gives us some facilities for slashifying strings & things, so that # Perl's reg.exps will be ok with them. #use searchslash; #This gives us facilities for reading files into strings (read_in_file) # or arrays (file_into_array). use filereaders; #This contains the siteweave routine that applies the local Rules. use siteweave; local ($default_rules) = 'sample/rules.swr'; #This variable is here to indicate the default file we are to be editing local ($default_file) = 'sample/nuts.html'; #This variable indicates the default location for the updated file local ($default_update) = 'sample/nnuts.html'; #If outy is not locally defined up here, they will be globally #defined by open_files, below. #the (current) definitions of what = a block local (@block_defs) = (2, "^(<.*?>)","^([^<]*)"); local (*outy); #the output File-Handle local (@rules); #the array that contains all rules we use local ($oldfile); #the string that contains the contents of the old file &main; #The MAIN is what we want to use. ############################################## ###Main; contains the NEW version of the BLOK-readin' in thing sub main () {#main &open_files(); my $block; #the thing we use to process a block of text #Now, @rules & $oldfile will have their contents filled. #print $oldfile; #Let's blockify the $oldfile, & write out the modified blocks: my $i = 0; my $j = @rules; my $dmzstart, $dmzstop; #We check to see if there is a DMZ tag. If there is more than 1, we #take the first, so there. while ($i < $j) {#check rules for DMZ if ($rules[$i] =~ /(.*?)\s+DMZ=(-|)(\d*)\s+(.*)/) {#we've found a DMZ rule... $rules[$i] = ""; #blank that variable, cuz we're implementing it here $dmzstart = $1; $dmzstop = $4; #there are now 3 rules, including the DMZ tag... $block_defs[0] = 3; #The "string of stuff" can not contain a DMZ start.. #The new tag is the "DMZ--stuff not to change--/DMZ"\ # & just to be nice, if they don't close the DMZ tag, it keeps # going until it gets to the end of the file. $block_defs[3] = '^\s*'.$dmzstart.'((?:.|\n)*)'.$dmzstop; last; }#close-- found-DMZ-rule $i++; }#close while-loop-for-checking-rules-for-DMZ while ($oldfile) {#while stuff in $oldfile, keep processing the file #HERE'S WHERE IT ACTUALLY DOES BLOKIFYING! #There CAN be N ways to define a blok: for ($i=$block_defs[0]; $i >0; $i--) {#checking to cut out a blok. if ($oldfile =~ s/$block_defs[$i]//) {#get that blok $block = $1; #If we are in an HTML tag.. if ($i == 1) {#we pass in an IN-HTML value as the 2nd value. $block = &siteweave($block,1); } else {#if not In HTML #if we're in a DMZ tag.. if ($i == 3) { #print it out unmodified! &outprint ($block); next; } #if we're NOT in a DMZ tag: else { #we modify the not-HTML, non-DMZ section $block = &siteweave($block); } }#close If not In HTML #in all cases but DMZ (where we already printed it out # unmodified & went through the loop again), we print out #print $block."\n"; &outprint ($block); }#end of get-that-blok }#close checking to cut out a block }#close $oldfile while loop }#close main #%%%%%%%%%%%%%%%%%%%%%%%%%% ###########NON-MAIN SUBPROGRAMS######### #%%%%%%%%%%%%%%%%%%%%%%%%%% ##########################OUTPRINT #This is a subprogram that prints the string it is #given to outy. sub outprint () { my $block = $_[0]; print outy $block."\n"; } ##########################OPEN_FILES #Here I'll use some code from earlier programs, that do #the file-opening. sub open_files () {#open files #local (*rulz); #the rules File-Handle my ($rulefile); #file name for Rules file #if the user gave us a 3rd input in command-line, #that was the name of their rules file if ($ARGV[0]) {$rulefile = $ARGV[0];} #if the user did not, we ask for it here else { print "What rules file would you like to use?"; chomp($rulefile = ); if (!$rulefile) {$rulefile = $default_rules;} } my $fil;#file name for stuff to be read IN #if the user gave us a 2nd input in command-line, #that was the name of the file they wanted to modify.. if ($ARGV[1]) {$fil = $ARGV[1];} #if the user did not, we ask for it here. else { print "What file would you like to convert?"; chomp($fil = ); if (!$fil) {$fil = $default_file;} #my testing file default }#close get-where-to-write-input-file-to print $fil." will be blockified & sitewoven!\n"; my $new; #where we're writing out the Newly-Edited thing #If they input a 3rd value on the command line, #that's where it is to be written to if ($ARGV[2]) {$new = $ARGV[2];} #if the user did not, we ask for it here else { print "What file would you like the converted version written to?"; chomp ($new = ); if (!$new) #in case they wanted to write & save to the same place, or the default, is: {$new = $default_update;} #my testing output default }#close get-where-to-write-output-file-to #reggy is the regular expression we'll use to sort the rules in file_into_sorted... #$2 is the one we want, as reflected below. # What reggy matches is "Chunk, followed by =, then priority, then * or \s my $reggy = '(.*?)=(\d+|-\d+)?(\*|\s)'; #Fill the @rules array with the rules, sorted by Priority, then rules-file order @rules = &file_into_sorted_array($rulefile,$reggy); $oldfile = &read_in_file($fil); #fill the $oldfile with the contents of oldfile #If outy is not locally defined at the beginning, it will be globally #defined here. open (outy, ">$new") or die "ZAM! Bad Write-to attempt"; #& the to-file return; #all is well in open_file-vil }#close open_files