Print missing html entities from XML file in DOCTYPE ready to paste format

README.md

Description

Command Line that print entities (like & or ") in way that can be paste into <!DOCTYPE if xml file don't define them and parser return error, that entity is not defined. Command check if Entity is not already defined.

gistfile1.sh

ALIGN=18;grep -oE '&[^#][^;]+;' foo.xml | sort | uniq | while read entity; do name=$(echo -n $entity | sed -e 's/[;&]//g');grep ENTITY foo.xml | grep " $name " > /dev/null || echo '<!ENTITY '$(echo -n $entity | html2text | perl -e "printf '$name %'.($ALIGN-(length '$name')).'s;\'>','\'&#'.(ord <>);"); done

output.txt

<!ENTITY aacute       '&#195;'>
<!ENTITY acute        '&#194;'>
<!ENTITY aelig        '&#195;'>
<!ENTITY agrave       '&#195;'>
<!ENTITY alpha        '&#206;'>
<!ENTITY amp           '&#38;'>
<!ENTITY apos          '&#38;'>
<!ENTITY auml         '&#195;'>
<!ENTITY eacute       '&#195;'>
<!ENTITY Eacute       '&#195;'>
<!ENTITY ecirc        '&#195;'>
<!ENTITY egrave       '&#195;'>
<!ENTITY empty         '&#38;'>
<!ENTITY gt            '&#62;'>
<!ENTITY hArr          '&#38;'>
<!ENTITY iota         '&#206;'>
<!ENTITY iuml         '&#195;'>
<!ENTITY kappa        '&#206;'>
<!ENTITY lambda       '&#206;'>
<!ENTITY Lambda       '&#206;'>
<!ENTITY lang          '&#38;'>
<!ENTITY larr         '&#226;'>
<!ENTITY ldquo         '&#38;'>
<!ENTITY lsquo         '&#38;'>
<!ENTITY lt            '&#60;'>
<!ENTITY mdash         '&#38;'>