#!/usr/bin/env -S awk -f
BEGIN {
in_paragraph= 0
in_blockquote= 0
in_ul= 0
ul_depth= 0
in_ol= 0
in_italics= 0
in_bold= 0
in_link= 0
in_table= 0
emitted= 0
}
# Init lookahead
FNR == 1 {
getline nextline < FILENAME
getline nextline < FILENAME
}
# Drop setext header marks
/^[=-]+$/ {
getline
getline nextline < FILENAME
}
# Escape all ampersands (HTML rule)
/&/ {
gsub(/&/, "\\&", $0)
}
# Restore ampersands part of HTML character escape
/&[A-Za-z]{2,4};/ {
gsub(/&/, "\\&", $0)
}
# Escape less than (<) (HTML rule)
/ {
gsub("<", "\\<", $0)
}
# line break
/^.*[ ]{2}$/ {
sub( /[ ]{2}$/, "
", $0 )
}
# block quote
/^>/ {
if( (! in_blockquote) && (! emitted) ) {
print "
" in_paragraph= 0 } if( in_italics ) { print "" in_italics= 0 } if( in_bold ) { print "" in_bold= 0 } if( in_strike ) { print "" in_code= 0 } if( in_code ) { print "" in_code= 0 } if( in_paragraph ) print "" }" in_blockquote= 1 in_paragraph= 1 } sub(/^>[ ]*/, "", $0) } # table /^[ ]*[|]/ { #print "" if( ! in_table ) { print "
" in_table= 1 in_paragraph= 1 } if( match($0, /---/) ) emitted= 1 else if( match(nextline, /---/) ) { sub(/^[ ]*[|][ ]*/, "
" in_table= 0 in_paragraph= 0 } if( in_ul ) { while( in_ul > 0 ) { print "" in_ul-- } ul_depth= 0 in_paragraph= 0 } if( in_ol ) { print "" in_ol= 0 in_paragraph= 0 } if( in_italics ) { print "" in_italics= 0 } if( in_bold ) { print "" in_bold= 0 } if( in_paragraph ) { print "" in_paragraph= 0 } } # paragraph and/or setext header $0 !~ /^$/ { if( match(nextline, /^[=]+$/) ) { printf("", $0) gsub(/[ ]*[|][ ]*/, " ", $0) sub(/[ ]*[|][ ]*$/, " ", $0) } else { sub(/^[ ]*[|][ ]*/, " ", $0) gsub(/[ ]*[|][ ]*/, " ", $0) sub(/[ ]*[|][ ]*$/, " ", $0) } } # ATX headers /^#/ { n= match($0, /^#+/) header_lvl= RLENGTH repl= sprintf(" ", header_lvl) sub(/^#+ ?/, repl, $0) repl= sprintf(" ", header_lvl) if( match($0, /#+$/) ) sub(/ ?#+$/, repl, $0) else sub(/$/, repl, $0) print emitted= 1 } # unordered list /^[ ]*[*+-] / { uld= match($0, /[*+-]/) if( uld > ul_depth ) { sub(/^[ ]*[*+-][ ]*/, "
- ", $0) in_ul++ in_paragraph= 1 } else if( uld < ul_depth ) { sub(/^[ ]*[*+-][ ]*/, "
", $0) in_ul-- } else { sub(/^[ ]*[*+-][ ]*/, " ", $0) } ul_depth= uld uld= match(nextline, /[*+-]/) if( uld <= ul_depth ) { sub(/$/, " ", $0) } } # ordered list /^(([0-9]+)|([A-Za-z])|([iIvVxXlL]{1,4}))[.]/ { t= "1" if( $0 ~ /[ivxl]+[.]/ ) { t= "i" sub(/^[ivxl]{1,4}[.][ ]*/, "", $0) } else if( $0 ~ /[IVXL]+[.]/ ) { t= "I" sub(/^[IVXL]{1,4}[.][ ]*/, " ", $0) } else if( $0 ~ /^[A-Z]./ ) { t= "A" sub(/^[A-Z][.][ ]*/, " ", $0) } else if( $0 ~ /^[a-z]./ ) { t= "a" sub(/^[a-z][.][ ]*/, " ", $0) } else { sub(/^[0-9]+[.][ ]*/, " ", $0) } if( ! in_ol ) { printf(" ", $0) } # Links /\[.*\]\(.*\)/ { in_link= 1 s = "" if( match($0, /(\[.*\])(\(.*)( ".*"\))/, tla) ) { l= substr(tla[2], 2) match(tla[3], /".*"/) a= substr(tla[3], RSTART, RLENGTH) match(tla[1], /\[.*\]/) t= substr(tla[1], RSTART+1, RLENGTH-2) s= sprintf("%s", l, a, t) } else if( match($0, /(\[.*\])(\(.*\))/, tl) ) { match(tl[2], /\(.*\)/) l= substr(tl[2], RSTART+1, RLENGTH-2) match(tl[1], /\[.*\]/) t= substr(tl[1], RSTART+1, RLENGTH-2) s= sprintf("%s", l, t) } sub(/\[.*\)/, s) } # bold /[*_][*_]/ { if( !in_link && !in_codeblock && !in_code ) { while( $0 ~ /[*_][*_]/ ) { if( ! in_bold ) { sub(/[*_][*_]/, "", $0) in_bold= 1 } else { sub(/[*_][*_]/, "", $0) in_bold= 0 } } } } # italics /[*_]/ { if( !in_link && !in_codeblock && !in_code ) { while( $0 ~ /[*_]/ ) { if( ! in_italics ) { sub(/[*_]/, "", $0) in_italics= 1 } else { sub(/[*_]/, "", $0) in_italics= 0 } } } } # strikethrough /~~/ { if( !in_link && !in_codeblock && !in_code ) { while( $0 ~ /~~/ ) { if( ! in_strike ) { sub(/~~/, "\n", t) in_ol= 1 in_paragraph= 1 } sub(/$/, "
", $0) in_strike= 1 } else { sub(/~~/, "", $0) in_strike= 0 } } } } # code block /^```/ { if( ! in_codeblock ) { sub(/```/, "", $0) in_codeblock= 1 } else { sub(/```/, "", $0) in_codeblock= 0 } } # code in text /`/ { if( !in_link ) { while( $0 ~ /`/ ) { if( ! in_code ) { sub(/`/, "", $0) in_code= 1 } else { sub(/`/, "", $0) in_code= 0 } } } } # empty line /^$/ { if( in_blockquote && (! match(nextline, /^>/)) ) { print "" in_blockquote= 0 in_paragraph= 0 } if( in_table ) { print "%s
\n", $0) emitted= 1 } if( match(nextline, /^[-]+$/) ) { printf("%s
\n", $0) emitted= 1 } if( (! in_paragraph) && (! emitted) ) { print "" in_paragraph= 1 } } # Finally: print the line { if( ! emitted ) print emitted= 0 getline nextline < FILENAME } END { if( in_table ) { print "" in_paragraph= 0 } if( in_ul ) { print "" in_paragraph= 0 } if( in_ol ) { print "" in_paragraph= 0 } if( in_blockquote ) { print "