#!/usr/bin/awk -f
# -*- indent-tabs-mode: t; -*-
# HAT TRICK
# (C) 2022 C. Duckworth

### Commentary:

### Code:
BEGIN {
	split("html,gemini,gopher", HT_FORMATS_AVAILABLE, ",")
	process_arguments()
	normalize_ht_formats()
	if (! HT_TAGCHARS[1]) {
		split("b:**,i://,code:``", HT_TAGCHARS, ",")
	}
	# Output buffer.  The output will be chunked into blocks.
	BUFFER = ""
	# Default block type
	DEFAULT_BLOCK = "p"
	# The current block type.  We start with a standard paragraph.
	BLOCK = DEFAULT_BLOCK
}

### RAW TEXT
/^>>>/ {
	BLOCK = "raw"
	if ($1) {
		split($1, raw_type, ",")
	}
}

/^<<</ {
	BLOCK = DEFAULT_BLOCK
	for (type in raw_type) {
		delete raw_type[type]
	}
}

### BLOCKS
block == "raw" {
	bufpush($0, "\n")
	next
}

/^#+/ {	# Headers
	match($0, /^#+/)
	if (! (BLOCK == "h" RLENGTH)) {
		buflush()
		tag = "h" RLENGTH
	}
	sub(/^#+[ \t]*/, "", $0)
}

/^>/ {	# Block quote
}

/^-/ {	# Unordered list
}

/^[0-9]\./ {	# Ordered list
}

/^---$/ {	# Section break
}

/^$/ {	# End a block
	buflush()
	BLOCK = DEFAULT_BLOCK
}

### LINES
/^=>/ {	# Link
}

/^</ {	# HTML tag
}

/^;/ {	# Comment
}

# EVERYTHING ELSE
/./ {
}

### FINISH
END {
	buflush()
}


### FUNCTIONS
function buflush()
{
	# Print the buffer and close the current block.
	if (BUFFER) {
		ht_print(BUFFER)
	}
	BUFFER = ""
	if ("html" in HT_FORMATS && BLOCK != "raw") {
		ht_print("</" BLOCK ">")
	}
	print
}

function bufpush(str, sep)
{
	# Push STR onto the buffer after SEP (defaults to space).
	sep = sep ? sep : " "
	BUFFER = BUFFER (BUFFER ? sep : "") str
}

function ht_print(str)
{
	if (HT_FORMATS_COUNT == 1) {
		print str
	} else {
		split(str, arr, "\n")
		for (format in HT_FORMATS) {
			line = 1
			while (arr[line]) {
				printf "%s\t%s\n", format, arr[line++]
			}
		}
	}
}

function html_escape(str)
{
	# Escape HTML entities and beginning-line spaces.
	gsub(/&/, "\\&amp;", t)
	gsub(/</, "\\&lt;", t)
	gsub(/>/, "\\&gt;", t)
	sub(/^ /, "\\&nbsp;", t)
	return t
}

function normalize_ht_formats()
{
	for (format in HT_FORMATS_AVAILABLE) {
		normat[format] = 0
	}
	if (! HT_FORMATS[1]) {
		for (i in HT_FORMATS_AVAILABLE) {
			HT_FORMATS[i] = HT_FORMATS_AVAILABLE[i]
		}
	}
	for (format in HT_FORMATS) {
		if (format == "all") {
			for (i in HT_FORMATS_AVAILABLE) {
				HT_FORMATS[i] = HT_FORMATS_AVAILABLE[i]
			}
			return
		} else if (format ~ /^-/) {
			delete normat[substr(format, 2)]
		} else {
			normat[format] = 1
		}
	}
	for (format in normat) {
		if (normat[format]) {
			HT_FORMATS[format] = format
		}
	}
	for (format in HT_FORMATS) {
		HT_FORMATS_COUNT++
	}
}

function process_arguments()
{
	a = 1
	HT_FORMATS[1] = 0
	HT_TAGCHARS[1] = 0
	while (ARGV[a]) {
		if (a == "-c" || a ~ /^--chars=/) {
			# HTML tag <-> markup character correspondance
			if (a == "-c") {
				a++
			} else if (a ~ /^--chars=/) {
				sub(/^[^=]*=/, "", a)
				# HT_TAGCHARS is an array
			}
			split(a, HT_TAGCHARS, ",")
		} else if (a == "-f" || a ~ /^--format=/) {
			# Output format
			if (a == "-f") {
				a++
			} else if (a ~ /^--format=/) {
				sub(/^[^=]*=/, "", a)
				# HT_FORMATS is an array
			}
			split(a, HT_FORMATS, ",")
		}
		a++
	}
}