Avoid non-ASCII characters in m4 files.

5 files changed, 27 insertions, 68 deletions

diff --git a/TODO b/TODO
index 9456fc6..88bae11 100644
--- a/TODO
+++ b/TODO
@@ -1,4 +1,4 @@
-#+TITLE: Autoconf Archive TODO    -*- mode:org;coding:utf-8;fill-column:79; -*-
+#+TITLE: Autoconf Archive TODO                 -*- mode:org;fill-column:79; -*-
 
 * TODO Document our new mailing lists.
 
@@ -21,32 +21,6 @@
     been changed in Git. Other branches, like 'maint', aren't tracked; there is
     only a moderate amount of traffic.
 
-
-* TODO Decide how to handle non-ASCII content in m4 files
-
-  Several macros contain non-ASCII characters, for example:
-
-  : $ grep@ Ludo m4/*.m4
-  : m4/_ac_c_ifdef.m4:#   Copyright (c) 2008 Ludovic Courtès <ludo@chbouib.org>
-  : m4/ac_cxx_compiler_vendor.m4:#   Copyright (c) 2008 Ludovic Courtès <ludo@chbouib.org>
-  : m4/ac_cxx_cppflags_std_lang.m4:#   Copyright (c) 2008 Ludovic Courtès <ludo@chbouib.org>
-  : m4/ac_cxx_cxxflags_std_lang.m4:#   Copyright (c) 2008 Ludovic Courtès <ludo@chbouib.org>
-  : m4/ac_cxx_ldflags_std_lang.m4:#   Copyright (c) 2008 Ludovic Courtès <ludo@chbouib.org>
-  : m4/ac_libtoolize_cflags.m4:#   Copyright (c) 2008 Ludovic Courtès <ludo@chbouib.org>
-
-  The encoding currently in use is "latin1", but the m4 files don't expose
-  that information. The situation can be improved as follows:
-
-  1) Add machine-readable charset declarations (i.e. for Emacs) to all m4
-     files. (Or just to those that actually use non-ASCII?)
-
-  2) Encode all files in UTF-8, like everyone else seems to do. (But add
-     charset declarations anyway?)
-
-  3) Restrict m4 files to ASCII.
-
-  Pure ASCII is my favorite.
-
 * TODO Figure out how to generate the AUTHORS file
 
   The file should contain the people who committed to Git and the people who
diff --git a/cfg.mk b/cfg.mk
index cf0914d..f3b308d 100644
--- a/cfg.mk
+++ b/cfg.mk
@@ -26,11 +26,11 @@ ALL_RECURSIVE_TARGETS += website
 website: $(HTML_FILES)
 
 $(STAGEDIR)/%.html : $(M4DIR)/%.m4 $(STAGEDIR)/.dirCreated $(srcdir)/macro.py $(srcdir)/macro2html.py
-	@$(srcdir)/macro2html.py --input-encoding=latin-1 --output-encoding=latin-1 --output-dir=$(STAGEDIR) --output-suffix=.html $<
+	@$(srcdir)/macro2html.py "$<" "$@"
 
 $(HTMLDIR)/%.html : $(STAGEDIR)/%.html
-	@echo publish $*
-	@tidy -quiet -ascii --indent yes --indent-spaces 1 --tidy-mark no -wrap 80 --hide-comments yes $< >$@
+	@echo generating $*.html
+	@tidy -quiet -ascii --indent yes --indent-spaces 1 --tidy-mark no -wrap 80 --hide-comments yes "$<" >"$@"
 
 $(STAGEDIR)/.dirCreated:
 	@$(MKDIR_P) $(STAGEDIR)
diff --git a/m4 b/m4
-Subproject 15c3cc5e5fdd75385ec266a935dc25519754596
+Subproject ad017ae36905e9ec0e6ad17e8b73c8b70e1a0c0
diff --git a/macro.py b/macro.py
index d3fe9fa..45b06cd 100755
--- a/macro.py
+++ b/macro.py
@@ -6,14 +6,13 @@ import re
 import os.path as path
 import sys
 import textwrap
-import encodings
 
-def loadFile(path, encoding):
-  with closing( encodings.search_function(encoding).streamreader(open(path)) ) as fd:
+def loadFile(path):
+  with closing( open(path) ) as fd:
     return fd.read()
 
-def writeFile(path, encoding, buffer):
-  with closing( encodings.search_function(encoding).streamwriter(open(path, "w")) ) as fd:
+def writeFile(path, buffer):
+  with closing( open(path, "w") ) as fd:
     fd.write(buffer)
 
 def splitSections(buffer):
@@ -61,10 +60,10 @@ def collapseText(lines, width = 72):
   return body
 
 class Macro:
-  def __init__(self, filePath, encoding):
+  def __init__(self, filePath):
     self.name = path.splitext(path.basename(filePath))[0]
     # header and body are separated by an empty line.
-    (header,body) = loadFile(filePath, encoding).split("\n\n", 1)
+    (header,body) = loadFile(filePath).split("\n\n", 1)
     self.body = body.split('\n')
     # drop initial header (if present)
     header = re.sub(r"^\n*# =+\n#[^\n]*\n# =+\n(#\n)+", '', header, 1)
diff --git a/macro2html.py b/macro2html.py
index 7a81316..d49d9c9 100755
--- a/macro2html.py
+++ b/macro2html.py
@@ -2,17 +2,8 @@
 
 assert __name__ == "__main__"
 
-import os.path as path
+import sys
 from macro import Macro, writeFile
-from optparse import OptionParser
-
-opts = OptionParser()
-opts.add_option('-v', "--verbose", dest = "verbose", default = False, action = "store_true")
-opts.add_option('', "--input-encoding", dest = "inEncode", default = "latin-1")
-opts.add_option('', "--output-encoding", dest = "outEncode", default = "latin1")
-opts.add_option('', "--output-dir", dest = "outDir", default = "stage")
-opts.add_option('', "--output-suffix", dest = "suffix", default = ".html")
-(options, args) = opts.parse_args()
 
 tmpl = """\
 <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN"
@@ -105,17 +96,14 @@ def formatAuthor(a):
   else:
     return "Copyright &copy; %(year)s %(name)s" % a
 
-for m4File in args:
-  (stem,suff) = path.splitext(path.basename(m4File))
-  assert suff == ".m4"
-  outFile = path.join(options.outDir, stem + options.suffix)
-  assert outFile != m4File
-  if options.verbose:
-    print m4File, "->", outFile
-  m = Macro(m4File, options.inEncode)
-  m.synopsis = "<br>\n".join([ "<code>%s</code>" % quoteHtml(l) for l in m.synopsis ])
-  m.description = '\n\n'.join(map(formatParagraph, m.description))
-  m.description = m.description.replace("</pre>\n\n<pre>", "\n\n")
-  m.authors = "<br>\n".join(map(formatAuthor, m.authors))
-  m.license = '\n'.join(map(formatParagraph, m.license))
-  writeFile(outFile, options.outEncode, tmpl % m.__dict__)
+if len(sys.argv) != 3:
+  raise Exception("invalid command line syntax: %s" % ' '.join(map(repr, sys.argv)))
+(m4File,outFile) = sys.argv[1:]
+assert outFile != m4File
+m = Macro(m4File)
+m.synopsis = "<br>\n".join([ "<code>%s</code>" % quoteHtml(l) for l in m.synopsis ])
+m.description = '\n\n'.join(map(formatParagraph, m.description))
+m.description = m.description.replace("</pre>\n\n<pre>", "\n\n")
+m.authors = "<br>\n".join(map(formatAuthor, m.authors))
+m.license = '\n'.join(map(formatParagraph, m.license))
+writeFile(outFile, tmpl % m.__dict__)