#!/usr/bin/python
# text2html - trivial html encoding of normal text

import sys
import re

# precompile regular expressions
re_quoted = re.compile(r"(?m)^(>.*?)$")
re_url = re.compile(r"<URL:(.*)>")
re_http = re.compile(r"(http:\S+)")
re_strong = re.compile(r"\*(\S+)\*")
re_em = re.compile(r"\b_(\S+)_\b")

# split paragraphs
for para in open(sys.argv[1]).read().split("\n\n"):
    # TODO encode entities: dont encode "<>" but do "&"
    if para.startswith(" "):
        print "<pre>\n%s\n</pre>" % para
    else:
        para = re_quoted.sub(r"\1<br />", para)          # quoted text
        para = re_url.sub(r'<a href="http://pleac.sourceforge.net/include/python/ch20/\1">\1</a>', para)  # embedded URL
        para = re_http.sub(r'<a href="http://pleac.sourceforge.net/include/python/ch20/\1">\1</a>', para) # guessed URL
        para = re_strong.sub(r"<strong>\1</strong>",para)   # this is *bold* here
        para = re_em.sub(r"<em>\1</em>",para)            # this is _italic_ here
        print "<p>\n%s\n</p>" % para                     # add paragraph tags