#!/bin/sh exit 1 # http://www.berklix.net/~jhs/src/bsd/jhs/bin/public/tidy/tidy.sh # Copyright Julian H. Stacey jhs @ berklix.net # Purpose: To help debug someone's web who doesn't have tidy command. # Installed to /usr/local/www/cgi-bin/tidy # Called by http://www.berklix.net/cgi-bin/tidy # PS I have run this script, captured the output, # & checked the output with tidy, # This is a recovered merge of 2 version, JJLATER To Be reviewed & tested. # To ensure this script generates error free HTML. I even ran ispell too. # PATH=/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin PATH=/sbin:/bin:/usr/sbin:/usr/bin export HOME PATH TZ # Do not change TMP without also editing embedded /tmp/ elsewhere TMP=/usr/local/www/data/tmp # /usr/local/www/apache22/data/tmp # /site/usr/local/www/Data/virtual/tmp # Remember the $TMP should be 755 root staff # Also apache.conf needs to allow .txt # BUG: For some reason, can not click in the table of 6 # on localhost - files do not display , # yet files get created. Maybe its cos a sym link local -> ../usr1/local # when this script run on berklix servers it is OK rm -f \ $TMP/tidy_tidied_errors.txt \ $TMP/tidy_tidied_html.txt \ $TMP/tidy_tidied_lines.txt \ $TMP/tidy_original_errors.txt \ $TMP/tidy_original_html.txt \ $TMP/tidy_original_lines.txt \ $TMP/tidy_txt.tmp # In case of some previous abort. URL=http://www.berklix.com # All my temporary files end in .txt as: # Better chance apache .conf will know how to display them, # than a random name like .error # Stops firefox rendering them as HTML, leaves it raw, easier for user to see. cd $TMP fetch -o tidy_original_html.txt $URL cp tidy_original_html.txt tidy_tidied_html.txt /usr/local/bin/tidy -i -m tidy_tidied_html.txt 2>1 > /dev/null HEADER="http://www.berklix.net/cgi-bin/tidy `date`" HOSTNAME=`hostname` echo "$HEADER" > tidy_original_lines.txt echo "$URL original_lines" >> tidy_original_lines.txt echo "" >> tidy_original_lines.txt /bin/cat -n tidy_original_html.txt >> tidy_original_lines.txt echo "$HEADER" > tidy_tidied_lines.txt echo "$URL tidied_lines" >> tidy_tidied_lines.txt echo "" >> tidy_tidied_lines.txt /bin/cat -n tidy_tidied_html.txt >> tidy_tidied_lines.txt echo "$HEADER" > tidy_original_errors.txt echo "$URL original_errors" >> tidy_original_errors.txt echo "" >> tidy_original_errors.txt /usr/local/bin/tidy -errors tidy_original_html.txt \ 2>> tidy_original_errors.txt echo "$HEADER" > tidy_tidied_errors.txt echo "$URL tidied_errors" >> tidy_tidied_errors.txt echo "" >> tidy_tidied_errors.txt /usr/local/bin/tidy -errors tidy_tidied_html.txt \ 2>> tidy_tidied_errors.txt # Now put headers on to make it a bit harder for someone # to abuse this as a proxy echo "$HEADER" > tidy_txt.tmp echo "$URL original_html" >> tidy_txt.tmp echo "" >> tidy_txt.tmp cat tidy_original_html.txt >> tidy_txt.tmp rm tidy_original_html.txt mv tidy_txt.tmp tidy_original_html.txt echo "$HEADER" > tidy_txt.tmp echo "$URL tidied_html" >> tidy_txt.tmp echo "" >> tidy_txt.tmp cat tidy_tidied_html.txt >> tidy_txt.tmp rm tidy_tidied_html.txt mv tidy_txt.tmp tidy_tidied_html.txt echo 'Content-type: text/html' echo "" echo '' echo '' #{ echo '' #{ echo '' echo '' # 'LINK' above over rides the firefox blue icon of an envelope. echo '' #{ echo 'HTML Tidy' # Text Shows in the header bar in Firefox. echo '' #} echo '' #} echo '' #{ echo '

' #{ echo '' #{ echo -n 'http://www.berklix.net/cgi-bin/tidy' echo -n '' #} echo '

' #} echo 'This is an HTML Tidy diagnostic script,' echo 'analysing initially a single URL.' # echo '(Currently:' # echo -n '' # echo -n $URL # echo ' )' #} echo '

Why Use It

' echo '
' #{ echo 'For Microsoft users who do not have local Unix tools. To reformat,' echo 'check, & insert line numbers to help debug their HTML.' echo 'Presume search engines will rate you higher' echo 'if you have clean HTML.' echo 'Some commercial HTML generator' echo 'programs fail to generate clean HTML, some use browser dependent' echo 'proprietary extensions, & humans fail too with' echo 'hand written HTML' echo '
' #} # I hope This is no longer needed now I am importing $HOSTNAME # echo '

If your browser is on any of these

' # echo '' #} # echo '

Click to one of these instead

' # echo '' #} # echo '' #{ # echo '(Because Berklix is served by a cluster of web servers, ' # echo '& the server that has served you this text & has analysed your URL,' # echo 'may not be the same server that next responds when you click' # echo 'one of the 6 files below (so you may get an obsolete old copy, or even an error of no existent file)).' # echo '' #} # echo '
' echo -n '

' #{ echo 'Analysis of HTML compliance for:' echo -n '' echo -n $URL echo '' #} echo '

' #} echo '
' #{ echo '' #{ echo '' #{ echo '' #} echo '' #} echo '' #} # The update was imported from date/xearth, # but I dont think I need it here, # the user could also push refetch on browser button. # echo '' #{ # echo '' #} # echo '' #} # echo '' #} # echo '' #{ echo '' #} echo '' #} echo '' #} echo '' #{ echo '' #} echo '' #} echo '' #} echo '' #{ echo '' #} echo '' #} echo '' #} echo '' #{ echo '' #} echo '' #} echo '' #} echo '' #{ echo '' #} echo '' #} echo '' #} echo '' #{ echo '' #} echo '' #} echo '' #} echo '
' #{ echo 'File' echo '' #{ echo 'Content' echo '
' #{ # echo '' #{ # echo 'UPDATE' # echo '' #} # echo '' #{ # echo 'Update the contents of the 6 files below.' # echo '
' #{ echo -n 'HTML of Original' echo '' #{ echo 'Original source HTML' # echo 'From fetch' echo '
' #{ echo -n 'Errors Of Original' echo '' #{ echo -n 'Errors' # echo 'From ' # echo 'tidy -errors original_html' echo '
' #{ echo -n 'Lines Numbers Of Original' echo '' #{ echo 'With line numbers' # echo 'From cat -n original_html' echo '
' #{ echo -n 'HTML tidied' echo '' #{ echo 'HTML Tidied (cleaned & formated)' # echo 'From' # echo 'tidy -m -i tidied_html,' # echo '
With indents & line feeds to easier read & debug.' echo '
' #{ echo -n 'Errors of Tidied' echo '' #{ echo 'Errors' # echo 'From tidy -errors tidied_html' echo '
' #{ echo -n 'Line numbers of Tidied' echo '' #{ echo 'With line numbers' # echo 'From cat -n tidied_html' echo '
' #} echo '
' #} echo '

If you think it slow

' echo '
' #{ echo 'It is a script: It fetches your URL' echo 'from the internet (including whatever delays that URL might' echo 'impose with scripts or link redirections etc), then' echo 'analyses the data & writes analysis in 6 files' echo 'on a Berklix server,' echo '(for you to later click on),' echo 'then builds this page, & sends it to your browser.' echo '
' #} echo '

Credits & Copyrights

' echo '
' #{ echo '' #{ echo -n '' #{ echo -n 'CGI Script' echo ', ' #} echo 'by' echo -n '' #{ echo -n 'Julian H. Stacey' echo ', ' #} echo 'calls ' echo -n '' #{ echo -n 'Tidy' echo ',' #} echo 'Indexed by ' echo -n '' #{ echo -n 'FreeBSD ports/www/tidy' echo ', ' #} echo 'Port wrapper in ' echo -n '' #{ echo -n 'SVN (subversion repository)' echo '' #} echo 'on Operating System: ' echo -n '' #{ echo -n 'FreeBSD' echo ', ' #} echo 'Admin by ' echo -n '' #{ echo -n 'Berklix.com' echo ', ' #} echo 'Servers: ' echo -n '' #{ echo -n 'Berklix.Net' echo ', ' #} echo '& ' echo -n '' #{ echo -n 'Sites' echo '.' #} echo 'Copyright ' echo -n '' #{ echo -n 'Julian H. Stacey' echo ', ' #} echo 'Munich 2015, ' echo 'You may use script under BSD licence, Please retain this credits line.' echo '' #} echo '
' #} echo '

Later Development

' echo '
' #{ echo '' #{ echo 'it could take a parameter to sample other URLs,' echo 'but there are issues to be considered first' echo '(security design comments welcome to' echo -n '' echo 'www.berklix.com/~jhs/contact/ )' echo '' #} echo '' #} echo '
' #} echo 'Other tools at www.berklix.net' #} echo '' #} echo '' #}