#!/bin/sh
#
# extrablatt - EXTRAct karolenderBLATT.
#
# Searches XXXXXXXXXXXX (or any file(s) given as arguments) for
# karolenderblatt entries and prints the corresponding date stamp,
# the karolenderblatt entry and 9 additional lines of trailing
# context.

# TODO:
#
# * parse date stamps (for year, month, day).
#
# * instead of printing the date stamp line, prefix the following output
#   with the (shortened) date stamp.
#
# * adjust date stamp by one resp. two days when the karolenderblatt
#   entry starts with 'gestern vor' or 'vorgestern vor', then replace
#   those terms by 'heute vor' (further processing mentioned below).
#
# * unHTMLIze the entries (e.g. &quot; -> ", strip <BR> tags etc.).
#
# * replace the relative terms ('heute vor N Jahren') by absolute
#   dates ('am dd.mm.yyyy'). Note: N may be a number or a noun
#   (like 'sechs', 'neun', including typos like 'neuen').

[ $# -gt 0 ] || set -- XXXXXXXXXXXX

# Could almost be done with grep -A9, except for the date stamp.
# So we use some awk, carefully obfuscated with love.
awk '
	function ref(d, e) {
		e = d
		d = "<span id=\"kb" ++i "\">" d "</span>"
		sub(/^[^0-9]*/, "", e)
		sub(/[^0-9]*$/, "", e)
		return ds[d]++ ? d : "<a name=\"" e "\"></a>" d
	}
	/^----/ && n = !(d = $0)
	/^<B>(kili|quabla)<\/B> \([^)]*\): (K|Reservek)arolenderblatt/ \
		&& !(n = 9 + !!($0 = ref(d) "\n" $0))
	n && n--
' "$@" |
sed -e 's@[[:space:]]*<BR>@<br/>@g' \
    -e 's@<B>@<b>@g' \
    -e 's@</B>@</b>@g' \
    -e 's@\([^[:alnum:]]\)\(GID=[0-9][0-9]*\)@\1<a href="https://www.karopapier.de/showmap.php?\2">\2</a>@' \
    -e 's@#[[:space:]]*\([0-9][0-9]*\)\(<br/>\)$@<a href="https://www.karopapier.de/chat#\1">#\1</a>\2@'