#!/usr/bin/env sh # lighttext.sh — Light Text (.lt) to HTML, Markdown or EPUB converter # Usage: lighttext.sh input.lt output.html # lighttext.sh input.lt output.md # lighttext.sh input.lt output.epub # --- Args --- if [ "$#" -ne 2 ]; then echo "Usage: lighttext.sh input.lt output.{html|md|epub}" >&2 exit 1 fi INPUT="$1" OUTPUT="$2" if [ ! -f "$INPUT" ]; then echo "Error: file not found: $INPUT" >&2 exit 1 fi # Detect output format from extension case "$OUTPUT" in *.html) FORMAT="html" ;; *.md) FORMAT="md" ;; *.epub) FORMAT="epub" ;; *) echo "Error: output must be .html, .md or .epub" >&2 exit 1 ;; esac TITLE=$(basename "$INPUT" .lt) # ============================================================ # SHARED HELPERS # ============================================================ html_escape() { sed 's/&/\&/g; s//\>/g; s/"/\"/g' } protect_escapes() { sed -e 's/\\\\/__ESC_BS__/g' -e 's/\\\*/__ESC_STAR__/g' -e 's/\\_/___ESC_UND__/g' -e 's/\\\[/__ESC_LB__/g' -e 's/\\\]/__ESC_RB__/g' -e 's/\\>/__ESC_GT__/g' -e 's/\\-/__ESC_DASH__/g' } restore_escapes() { sed -e 's/__ESC_BS__/\\/g' -e 's/__ESC_STAR__/*/g' -e 's/___ESC_UND__/_/g' -e 's/__ESC_LB__/[/g' -e 's/__ESC_RB__/]/g' -e 's/__ESC_GT__/>/g' -e 's/__ESC_DASH__/-/g' } inline_html() { echo "$1" | protect_escapes | awk ' function esc(s){gsub(/&/,"&",s); gsub(//,">",s); gsub(/"/,""",s); return s} { line=$0; out=""; while(match(line,/\[[^\]>]+\s*>\s*[^\]]+\]/)){ pre=substr(line,1,RSTART-1); link=substr(line,RSTART,RLENGTH); post=substr(line,RSTART+RLENGTH); sub(/^\[/,"",link); sub(/\]$/,"",link); split(link,p,/>/); txt=p[1]; url=p[2]; gsub(/^\s+|\s+$/,"",txt); gsub(/^\s+|\s+$/,"",url); out=out pre "" esc(txt) ""; line=post; } print out line; }' | html_escape | sed 's/<a href="//g; s/">/">/g; s/<\/a>/<\/a>/g' | sed 's/\*\([^*][^*]*\)\*/\1<\/strong>/g' | sed 's/_\([^_][^_]*\)_/\1<\/em>/g' | restore_escapes } inline_md() { echo "$1" | protect_escapes | sed 's/\[\([^]>]*[^]> ]\) *> *\([^]) ][^]]*\)\]/[\1](\2)/g' | sed 's/\*\([^*][^*]*\)\*/**\1**/g' | restore_escapes } is_heading() { stripped=$(echo "$1" | tr -d '[:space:][:digit:][:punct:]') [ -n "$stripped" ] && [ "$stripped" = "$(echo "$stripped" | tr '[:lower:]' '[:upper:]')" ] } # ============================================================ # PARSE — emit lines to stdout, caller redirects # ============================================================ parse_html() { in_ul=0 in_ol=0 para_lines="" para_open=0 flush_list_html() { [ "$in_ul" -eq 1 ] && { echo ""; in_ul=0; } [ "$in_ol" -eq 1 ] && { echo ""; in_ol=0; } } flush_para_html() { [ "$para_open" -eq 1 ] && { echo "

$(inline_html "$para_lines")

" para_lines="" para_open=0 } } echo "" echo '' echo "" echo '' echo '' echo "$TITLE" echo "" echo "" while IFS= read -r line || [ -n "$line" ]; do if [ -z "$line" ]; then flush_para_html; flush_list_html; continue fi if is_heading "$line"; then flush_para_html; flush_list_html echo "

$(inline_html "$line")

" prev_blank=0; continue fi if echo "$line" | grep -qE '^- .+'; then flush_para_html [ "$in_ol" -eq 1 ] && flush_list_html [ "$in_ul" -eq 0 ] && { echo "
    "; in_ul=1; } echo "
  • $(inline_html "$(echo "$line" | sed 's/^- //')")
    • " prev_blank=0; continue fi if echo "$line" | grep -qE '^[0-9]+\. .+'; then flush_para_html [ "$in_ul" -eq 1 ] && flush_list_html [ "$in_ol" -eq 0 ] && { echo "
      "; in_ol=1; } echo "
    1. $(inline_html "$(echo "$line" | sed 's/^[0-9]*\. //')")
      2. " prev_blank=0; continue fi flush_list_html if [ "$para_open" -eq 0 ]; then para_lines="$line"; para_open=1 else para_lines="$para_lines $line" fi prev_blank=0 prev_blank=0 done < "$INPUT" flush_para_html flush_list_html echo "" echo "" } # ---- parse_md() { in_ul=0 in_ol=0 para_lines="" para_open=0 ol_counter=0 flush_list_md() { in_ul=0; in_ol=0; ol_counter=0 } flush_para_md() { [ "$para_open" -eq 1 ] && { echo "$(inline_md "$para_lines")" echo "" para_lines="" para_open=0 } } echo "# $TITLE" prev_blank=1 while IFS= read -r line || [ -n "$line" ]; do if [ -z "$line" ]; then flush_para_md; flush_list_md; prev_blank=1 continue fi if [ "$prev_blank" -eq 1 ] && is_heading "$line"; then flush_para_md; flush_list_md echo "" echo "# $(inline_md "$line")" prev_blank=0; continue fi if echo "$line" | grep -qE '^- .+'; then flush_para_md [ "$in_ul" -eq 0 ] && { in_ul=1; } echo "- $(inline_md "$(echo "$line" | sed 's/^- //')")" prev_blank=0; continue fi if echo "$line" | grep -qE '^[0-9]+\. .+'; then flush_para_md [ "$in_ol" -eq 0 ] && { in_ol=1; ol_counter=0; } ol_counter=$((ol_counter + 1)) echo "$ol_counter. $(inline_md "$(echo "$line" | sed 's/^[0-9]*\. //')")" prev_blank=0; continue fi flush_list_md if [ "$para_open" -eq 0 ]; then para_lines="$line"; para_open=1 else para_lines="$para_lines $line" fi prev_blank=0 prev_blank=0 done < "$INPUT" flush_para_md } # ============================================================ # OUTPUT # ============================================================ case "$FORMAT" in html) parse_html > "$OUTPUT" ;; md) parse_md > "$OUTPUT" ;; epub) # EPUB = zip of: mimetype + META-INF/container.xml + OEBPS/content.opf + OEBPS/content.html if ! command -v zip > /dev/null 2>&1; then echo "Error: 'zip' is required for EPUB output" >&2 exit 1 fi TMPDIR=$(mktemp -d) trap 'rm -rf "$TMPDIR"' EXIT mkdir -p "$TMPDIR/META-INF" mkdir -p "$TMPDIR/OEBPS" # mimetype (must be uncompressed and first in zip) printf 'application/epub+zip' > "$TMPDIR/mimetype" # META-INF/container.xml cat > "$TMPDIR/META-INF/container.xml" << 'XML' XML # OEBPS/content.opf UUID=$(cat /proc/sys/kernel/random/uuid 2>/dev/null || od -x /dev/urandom | head -1 | awk '{print $2$3"-"$4"-"$5"-"$6"-"$7$8$9}') cat > "$TMPDIR/OEBPS/content.opf" << XML $TITLE en $UUID XML # OEBPS/content.html — reuse HTML parser but as XHTML for EPUB { echo '' echo '' echo '' echo '' echo '' echo "$TITLE" echo '' echo '' # Run parser but skip the outer html/head/body wrapper parse_html | sed '1,/^$/d' | sed '/<\/body>/,$d' echo '' echo '' } > "$TMPDIR/OEBPS/content.html" # Build EPUB zip — mimetype must be first and uncompressed ORIG_DIR=$(pwd) cd "$TMPDIR" zip -X0 "$ORIG_DIR/$OUTPUT" mimetype zip -rX9 "$ORIG_DIR/$OUTPUT" META-INF OEBPS cd "$ORIG_DIR" ;; esac echo "Done: $OUTPUT"