First stab at automated conversion

Makefile

@@ -0,0 +1,9 @@
+PAGES = page*.tex
+
+horrors2.dvi: horrors2.ltx $(PAGES)
+	latex $<
+	latex $<
+
+horrors2.pdf: horrors2.ltx $(PAGES)
+	pdflatex $<
+	pdflatex $<

horrors2.ltx

@@ -0,0 +1,60 @@
+\documentclass{book}
+
+%\usepackage[T1]{fontenc}
+%\usepackage{anttor}
+%\usepackage{tgothic}
+
+\title{Horrors 2: Bavarious Reasons}
+
+\newcommand{\by}[1]{{\large By {\sf #1}}\\\bigskip}
+
+
+\begin{document}
+\tableofcontents
+
+\part{Bavarious Reasons}
+\include{page1}
+
+\part{You're no Hakan}
+%\include{page2}
+
+\part{The Chronicles Of Biddick}
+%\include{page3}
+
+\part{My God, It's Full Of Stars}
+%\include{page4}
+
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+% Appendices
+
+
+\appendix
+\part{Inflamed Appendices}
+
+
+\chapter{Colophon}
+
+This book is a collection of stories from the Something Awful Forums,
+inspired by the 1991 book {\em Horrors} by Ben Biddick.  Most were written
+as a response to a challenge by forums moderator AYBraham:
+
+\begin{quotation}
+  Your goal: Write a short story in the theme of {\em Horrors}, the collection
+  of short stories written by Mr. Ben Beddick when he was 13 years
+  old. Each story is encouraged to feature the tragic protagonist Luke
+  Bavarius, and must be more than 500 words each. Each story must show
+  that "kids need to be respected and listened to."
+\end{quotation}
+
+This book was typeset by the \LaTeX{} document preparation system, using
+the {\sf book} class.
+
+It was set in the Antykwa Toru\'nska (``Antiqua of Torun'') typeface,
+the most horriffying of all the standard \TeX{}live typefaces.  The face
+was designed by Zygfryd Gardzielewski (1914--2001), and first cast in
+metal in 1960 in the Grafmasz typefoundry in Warsaw.  The variant used
+in this document was digitized by Janusz Marian Nowacki, a tremendously
+cool dude from Poland, based off of photocopies of the original design.
+
+\end{document}

sa2ltx.py

@@ -0,0 +1,76 @@
+#! /usr/bin/python3
+
+import optparse
+import xml.dom.minidom
+import xml.etree.ElementTree
+import re
+
+quotes_re = re.compile(r'"([^"]+)"')
+dots_re = re.compile(r'\.\.\.+')
+crap_re = re.compile(r'<p class="editedby">.*</p>', re.DOTALL)
+tag_re = re.compile(r'<[^>]+>')
+
+def by_class(e, classname):
+    todo = [e]
+    while todo:
+        i = todo.pop(0)
+        if i.get('class') == classname:
+            yield i
+        todo = i.getchildren() + todo
+
+def first_by_class(e, classname):
+    for i in by_class(e, classname):
+        return i
+
+def table_to_ltx(t):
+    dt = first_by_class(t, 'author')
+
+    username = dt.text
+    if not username:
+        # Moderators
+        username = dt.getchildren()[-1].tail
+    print('%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%')
+    print('\\by{%s}' % username)
+
+    body = first_by_class(t, 'postbody')
+
+    s = xml.etree.ElementTree.tostring(body)
+    s = s.replace('<br />', '\n')
+    s = s.replace('<i>', '{\\em ')
+    s = s.replace('</i>', '}')
+    s = s.replace('<b>', '{\\bf ')
+    s = s.replace('</b>', '}')
+    s = crap_re.sub('', s)
+    s = tag_re.sub('', s)
+    s = dots_re.sub('{\ldots}', s)
+    s = quotes_re.sub(r"``\1''", s)
+    print(s)
+
+def doc_to_ltx(doc):
+    for e in doc.getiterator('table'):
+        if e.get('class') == 'post':
+            table_to_ltx(e)
+
+def main():
+    p = optparse.OptionParser()
+    (opts, args) = p.parse_args()
+
+    for a in args:
+        f = open(a, encoding='iso-8859-1')
+        parser = xml.etree.ElementTree.XMLTreeBuilder()
+        parser.entity.update(nbsp=" ",
+                             rsaquo=">",
+                             lsquo="`",
+                             rsquo="'",
+                             ldquo="``",
+                             rdquo="''",
+                             hellip="{\\ldots}",
+                             ndash="---",
+                             mdash="---",
+                             iexcl="{\\!`}",
+                             copy="{\\copyright}",
+                             )
+        doc = xml.etree.ElementTree.parse(f, parser)
+        doc_to_ltx(doc)
+
+main()