Minor style tweaks

Author: acabal

README.md

@@ -67,9 +67,7 @@ These instructions were tested on Mac OS X 10.12. Your mileage may vary. Correct
 
 Help and pull requests are welcomed!
 
-- Move some legacy scripts like `build-kobo`, and `hyphenate` into appropriate libraries/scripts.
-
-- Update scripts to use new library global variables like `XHTML_NAMESPACES`, instead of redefining them in each script.
+- Move some legacy scripts like `build-kobo` and `hyphenate` into appropriate libraries/scripts.
 
 # Tool descriptions
 

british2american

@@ -4,9 +4,7 @@ import argparse
 import os
 import fnmatch
 import regex
-
-
-IGNORED_FILENAMES = ["colophon.xhtml", "titlepage.xhtml", "imprint.xhtml", "uncopyright.xhtml", "halftitle.xhtml", "toc.xhtml", "loi.xhtml"]
+import se
 
 
 def main():
@@ -26,7 +24,7 @@ def main():
 		if os.path.isdir(target):
 			for root, _, filenames in os.walk(target):
 				for filename in fnmatch.filter(filenames, "*.xhtml"):
-					if filename not in IGNORED_FILENAMES:
+					if filename not in se.IGNORED_FILENAMES:
 						target_filenames.add(os.path.join(root, filename))
 		else:
 			target_filenames.add(target)

build-images

@@ -10,6 +10,7 @@ import subprocess
 import regex
 import se
 
+
 def clean_inkscape_svg(filename, clean_path):
 	with open(filename, "r+") as file:
 		svg = file.read()
@@ -148,5 +149,6 @@ def main():
 			if args.verbose:
 				print("\t./images/titlepage.svg not found, skipping ...")
 
+
 if __name__ == "__main__":
 	main()

clean

@@ -8,6 +8,7 @@ import subprocess
 import regex
 import se
 
+
 def main():
 	parser = argparse.ArgumentParser(description="Prettify and canonicalize individual XHTML or SVG files, or all XHTML and SVG files in a source directory.  Note that this only prettifies the source code; it doesn't perform typography changes.")
 	parser.add_argument("-v", "--verbose", action="store_true", help="increase output verbosity")

dec2roman

@@ -5,6 +5,7 @@ import sys
 import roman
 import se
 
+
 def main():
 	parser = argparse.ArgumentParser(description="Convert a decimal number to a Roman numeral.")
 	parser.add_argument("-n", "--no-newline", dest="newline", action="store_false", help="don't end output with a newline")
@@ -30,5 +31,6 @@ def main():
 		else:
 			print(roman.toRoman(int(line)), end="")
 
+
 if __name__ == "__main__":
 	main()

extract-ebook

@@ -9,6 +9,7 @@ import magic
 from se.kindleunpack import kindleunpack # GPLv3: https://www.mobileread.com/forums/showthread.php?t=61986
 import se
 
+
 def main():
 	parser = argparse.ArgumentParser(description="Extract an EPUB, MOBI, or AZW3 ebook into ./FILENAME.extracted/ or a target directory.")
 	parser.add_argument("-v", "--verbose", action="store_true", help="increase output verbosity")
@@ -55,5 +56,6 @@ def main():
 		if args.verbose:
 			print(" OK")
 
+
 if __name__ == "__main__":
 	main()

find-mismatched-diacritics

@@ -6,6 +6,7 @@ import fnmatch
 import unicodedata
 import regex
 
+
 def main():
 	parser = argparse.ArgumentParser(description="Find words with mismatched diacritics in XHTML files.  For example, \"cafe\" in one file and \"café\" in another.")
 	parser.add_argument("-v", "--verbose", action="store_true", help="increase output verbosity")
@@ -68,5 +69,6 @@ def main():
 			if args.verbose:
 				print(" OK")
 
+
 if __name__ == "__main__":
 	main()

hyphenate

@@ -8,6 +8,7 @@ import regex
 from hyphen import Hyphenator, dict_info
 from bs4 import BeautifulSoup
 
+
 def main():
 	parser = argparse.ArgumentParser(description="Insert soft hyphens at syllable breaks in XHTML files.")
 	parser.add_argument("-v", "--verbose", action="store_true", help="increase output verbosity")
@@ -139,5 +140,6 @@ def main():
 		if args.verbose:
 			print(" OK")
 
+
 if __name__ == "__main__":
 	main()

make-url-safe

@@ -4,6 +4,7 @@ import argparse
 import sys
 import se.formatting
 
+
 def main():
 	parser = argparse.ArgumentParser(description="Make a string URL-safe.")
 	parser.add_argument("-n", "--no-newline", dest="newline", action="store_false", help="don't end output with a newline")
@@ -25,5 +26,6 @@ def main():
 		else:
 			print(se.formatting.make_url_safe(line), end="")
 
+
 if __name__ == "__main__":
 	main()

prepare-release

@@ -100,5 +100,6 @@ def main():
 				file.write(processed_xhtml)
 				file.truncate()
 
+
 if __name__ == "__main__":
 	main()

print-manifest-and-spine

@@ -25,5 +25,6 @@ def main():
 	if args.manifest:
 		print(se_epub.generate_spine())
 
+
 if __name__ == "__main__":
 	main()

reading-ease

@@ -10,7 +10,6 @@ from bs4 import BeautifulSoup
 import se
 
 
-IGNORED_FILENAMES = ["colophon.xhtml", "titlepage.xhtml", "imprint.xhtml", "uncopyright.xhtml", "halftitle.xhtml", "toc.xhtml", "loi.xhtml"]
 INCLUDED_CHARACTERS = list(string.whitespace) + list(string.digits) + [":", ";", ".", "?", "!"]
 
 
@@ -149,7 +148,7 @@ def main():
 
 		for root, _, filenames in os.walk(directory):
 			for filename in fnmatch.filter(filenames, "*.xhtml"):
-				if filename not in IGNORED_FILENAMES:
+				if filename not in se.IGNORED_FILENAMES:
 					with open(os.path.join(root, filename), "r", encoding="utf-8") as file:
 						text += BeautifulSoup(file.read(), "lxml").body.get_text() + " "
 
@@ -192,5 +191,6 @@ def main():
 		else:
 			print(flesch_reading_ease)
 
+
 if __name__ == "__main__":
 	main()

reorder-endnotes

@@ -15,7 +15,6 @@ def is_positive_integer(value):
 
 	return int_value
 
-
 def main():
 	parser = argparse.ArgumentParser(description="Increment the specified endnote and all following endnotes by 1.")
 	group = parser.add_mutually_exclusive_group(required=True)

roman2dec

@@ -32,5 +32,6 @@ def main():
 			se.print_error("Not a Roman numeral: {}".format(line))
 			exit(1)
 
+
 if __name__ == "__main__":
 	main()

semanticate

@@ -6,6 +6,7 @@ import fnmatch
 import regex
 from bs4 import BeautifulSoup
 
+
 def main():
 	parser = argparse.ArgumentParser(description="Automatically add semantics to Standard Ebooks source directories.")
 	parser.add_argument("-v", "--verbose", action="store_true", help="increase output verbosity")
@@ -102,5 +103,6 @@ def main():
 		if args.verbose:
 			print(" OK")
 
+
 if __name__ == "__main__":
 	main()

simplify-tags

@@ -192,5 +192,6 @@ def main():
 		if args.verbose:
 			print(" OK")
 
+
 if __name__ == "__main__":
 	main()

split-file

@@ -5,6 +5,7 @@ import os
 import regex
 import se.epub
 
+
 def main():
 	parser = argparse.ArgumentParser(description="Split an XHTML file into many files at all instances of <!--se:split-->, and include a header template for each file.")
 	parser.add_argument("filename", metavar="FILE", help="an XHTML file")
@@ -42,5 +43,6 @@ def output(chapter_number, header_xhtml, chapter_xhtml):
 		file.write(header_xhtml.replace("NUMBER", str(chapter_number)) + "\n" + chapter_xhtml + "\n</section></body></html>")
 		file.truncate()
 
+
 if __name__ == "__main__":
 	main()

titlecase

@@ -26,5 +26,6 @@ def main():
 		else:
 			print(se.formatting.titlecase(line), end="")
 
+
 if __name__ == "__main__":
 	main()

typogrify

@@ -6,11 +6,8 @@ import fnmatch
 import html
 import regex
 import smartypants
+import se
 
-# Some convenience aliases
-WORD_JOINER = "\u2060"		# word joiner, U+2060
-HAIR_SPACE = "\u200a"		# hair space, U+200A
-NBSP = "\u00a0"			# no-break space, U+00A0
 
 def main():
 	parser = argparse.ArgumentParser(description="Apply some scriptable typography rules from the Standard Ebooks typography manual to XHTML files.")
@@ -63,47 +60,47 @@ def main():
 				processed_xhtml = regex.sub(r"—”([a-z])", r"—“\1", processed_xhtml, flags=regex.IGNORECASE)
 				processed_xhtml = regex.sub(r"—’([a-z])", r"—‘\1", processed_xhtml, flags=regex.IGNORECASE)
 				processed_xhtml = regex.sub(r"-“</p>", r"—”</p>", processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"‘”</p>", r"’{}”</p>".format(HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"‘”</p>", r"’{}”</p>".format(se.HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
 
 				# Remove spaces between en and em dashes
 				# Note that we match at least one character before the dashes, so that we don't catch start-of-line em dashes like in poetry.
 				processed_xhtml = regex.sub(r"([^\.\s])\s*([–—])\s*", r"\1\2", processed_xhtml)
 
 				# First, remove stray word joiners
-				processed_xhtml = processed_xhtml.replace(WORD_JOINER, "")
+				processed_xhtml = processed_xhtml.replace(se.WORD_JOINER, "")
 
 				# Some older texts use the ,— construct; remove that archaichism
 				processed_xhtml = processed_xhtml.replace(",—", "—")
 
 				# Em dashes and two-em-dashes can be broken before, so add a word joiner between letters/punctuation and the following em dash
-				processed_xhtml = regex.sub(r"([^\s{}{}{}])([—⸻])".format(WORD_JOINER, NBSP, HAIR_SPACE), r"\1{}\2".format(WORD_JOINER), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"([^\s{}{}{}])([—⸻])".format(se.WORD_JOINER, se.NO_BREAK_SPACE, se.HAIR_SPACE), r"\1{}\2".format(se.WORD_JOINER), processed_xhtml, flags=regex.IGNORECASE)
 
 				# Add en dashes; don't replace match that is within an html tag, since ids and attrs often containg the pattern DIGIT-DIGIT
 				processed_xhtml = regex.sub(r"(?<!<[^>]*)([0-9]+)\-([0-9]+)", r"\1–\2", processed_xhtml)
 
 				# Add a word joiner on both sides of en dashes
-				processed_xhtml = regex.sub(r"{}?–{}?".format(WORD_JOINER, WORD_JOINER), r"{}–{}".format(WORD_JOINER, WORD_JOINER), processed_xhtml)
+				processed_xhtml = regex.sub(r"{}?–{}?".format(se.WORD_JOINER, se.WORD_JOINER), r"{}–{}".format(se.WORD_JOINER, se.WORD_JOINER), processed_xhtml)
 
 				# Add a word joiner if eliding a word with a two-em-dash
 				# Word joiner isn't necessary if punctuation follows
 				# Note the \p{{P}}.  We must double-curl {} because that's the escape sequence when using .format().  The actual regex should be \p{P} to match punctuation
-				processed_xhtml = regex.sub(r"([^\s{}{}{}])⸺".format(WORD_JOINER, NBSP, HAIR_SPACE), r"\1{}⸺".format(WORD_JOINER), processed_xhtml)
-				processed_xhtml = regex.sub(r"⸺([^\s\p{{P}}{}])".format(WORD_JOINER), r"⸺{}\1".format(WORD_JOINER), processed_xhtml)
+				processed_xhtml = regex.sub(r"([^\s{}{}{}])⸺".format(se.WORD_JOINER, se.NO_BREAK_SPACE, se.HAIR_SPACE), r"\1{}⸺".format(se.WORD_JOINER), processed_xhtml)
+				processed_xhtml = regex.sub(r"⸺([^\s\p{{P}}{}])".format(se.WORD_JOINER), r"⸺{}\1".format(se.WORD_JOINER), processed_xhtml)
 
 				# Remove word joiners from following opening tags--they're usually never correct
-				processed_xhtml = regex.sub(r"<([a-z]+)([^>]*?)>{}".format(WORD_JOINER), r"<\1\2>", processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"<([a-z]+)([^>]*?)>{}".format(se.WORD_JOINER), r"<\1\2>", processed_xhtml, flags=regex.IGNORECASE)
 
 				# Finally fix some other mistakes
 				processed_xhtml = processed_xhtml.replace("—-", "—")
 
 				# Replace Mr., Mrs., and other abbreviations, and include a non-breaking space
-				processed_xhtml = regex.sub(r"\b(Mr|Mr?s|Drs?|Profs?|Lieut|Fr|Lt|Capt|Pvt|Esq|Mt|St|MM|Mmes?|Mlles?)\.?\s+", r"\1.{}".format(NBSP), processed_xhtml)
-				processed_xhtml = regex.sub(r"<abbr>(Mr|Mr?s|Drs?|Profs?|Lieut|Fr|Lt|Capt|Pvt|Esq|Mt|St|MM|Mmes?|Mlles?)\.</abbr>?\s+", r"<abbr>\1.</abbr>{}".format(NBSP), processed_xhtml)
+				processed_xhtml = regex.sub(r"\b(Mr|Mr?s|Drs?|Profs?|Lieut|Fr|Lt|Capt|Pvt|Esq|Mt|St|MM|Mmes?|Mlles?)\.?\s+", r"\1.{}".format(se.NO_BREAK_SPACE), processed_xhtml)
+				processed_xhtml = regex.sub(r"<abbr>(Mr|Mr?s|Drs?|Profs?|Lieut|Fr|Lt|Capt|Pvt|Esq|Mt|St|MM|Mmes?|Mlles?)\.</abbr>?\s+", r"<abbr>\1.</abbr>{}".format(se.NO_BREAK_SPACE), processed_xhtml)
 
-				processed_xhtml = regex.sub(r"\bNo\.\s+([0-9]+)", r"No.{}\1".format(NBSP), processed_xhtml)
-				processed_xhtml = regex.sub(r"<abbr>No\.</abbr>\s+", r"<abbr>No.</abbr>{}".format(NBSP), processed_xhtml)
+				processed_xhtml = regex.sub(r"\bNo\.\s+([0-9]+)", r"No.{}\1".format(se.NO_BREAK_SPACE), processed_xhtml)
+				processed_xhtml = regex.sub(r"<abbr>No\.</abbr>\s+", r"<abbr>No.</abbr>{}".format(se.NO_BREAK_SPACE), processed_xhtml)
 
-				processed_xhtml = regex.sub(r"([0-9]+)\s<abbr", r"\1{}<abbr".format(NBSP), processed_xhtml)
+				processed_xhtml = regex.sub(r"([0-9]+)\s<abbr", r"\1{}<abbr".format(se.NO_BREAK_SPACE), processed_xhtml)
 
 				# A note on spacing:
 				# 					ibooks	kindle (mobi7)
@@ -130,9 +127,9 @@ def main():
 				# nth (as in nth degree)
 				processed_xhtml = regex.sub(r"\bn\-?th\b", r"<i>n</i>th", processed_xhtml)
 
-				# Remove double spaces that use nbsp for spacing
-				processed_xhtml = regex.sub(r"{}[{} ]+".format(NBSP, NBSP), r" ", processed_xhtml)
-				processed_xhtml = regex.sub(r" [{} ]+".format(NBSP), r" ", processed_xhtml)
+				# Remove double spaces that use se.NO_BREAK_SPACE for spacing
+				processed_xhtml = regex.sub(r"{}[{} ]+".format(se.NO_BREAK_SPACE, se.NO_BREAK_SPACE), r" ", processed_xhtml)
+				processed_xhtml = regex.sub(r" [{} ]+".format(se.NO_BREAK_SPACE), r" ", processed_xhtml)
 
 				# House style: remove spacing from common Latinisms
 				processed_xhtml = regex.sub(r"([Ii])\.\s+e\.", r"\1.e.", processed_xhtml)
@@ -143,40 +140,40 @@ def main():
 				processed_xhtml = regex.sub(r"B\.\s+C\.", r"BC", processed_xhtml)
 
 				# Put spacing next to close quotes
-				processed_xhtml = regex.sub(r"“[\s{}]*‘".format(NBSP), r"“{}‘".format(HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"’[\s{}]*”".format(NBSP), r"’{}”".format(HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"“[\s{}]*’".format(NBSP), r"“{}’".format(HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"‘[\s{}]*“".format(NBSP), r"‘{}“".format(HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"“[\s{}]*‘".format(se.NO_BREAK_SPACE), r"“{}‘".format(se.HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"’[\s{}]*”".format(se.NO_BREAK_SPACE), r"’{}”".format(se.HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"“[\s{}]*’".format(se.NO_BREAK_SPACE), r"“{}’".format(se.HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"‘[\s{}]*“".format(se.NO_BREAK_SPACE), r"‘{}“".format(se.HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
 
 				# We require a non-letter char at the end, otherwise we might match a contraction: “Hello,” ’e said.
-				processed_xhtml = regex.sub(r"”[\s{}]*’([^a-zA-Z])".format(NBSP), r"”{}’\1".format(HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"”[\s{}]*’([^a-zA-Z])".format(se.NO_BREAK_SPACE), r"”{}’\1".format(se.HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
 
 				# Fix ellipses spacing
 				processed_xhtml = regex.sub(r"\s*\.\s*\.\s*\.\s*", r"…", processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"[\s{}]?…[\s{}]?\.".format(NBSP, NBSP), r".{}…".format(HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"[\s{}]?…[\s{}]?".format(NBSP, NBSP), r"{}… ".format(HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"<p([^>]*?)>{}…".format(HAIR_SPACE), r"<p\1>…", processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"[\s{}]?…[\s{}]?\.".format(se.NO_BREAK_SPACE, se.NO_BREAK_SPACE), r".{}…".format(se.HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"[\s{}]?…[\s{}]?".format(se.NO_BREAK_SPACE, se.NO_BREAK_SPACE), r"{}… ".format(se.HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"<p([^>]*?)>{}…".format(se.HAIR_SPACE), r"<p\1>…", processed_xhtml, flags=regex.IGNORECASE)
 
 				# Remove spaces between opening tags and ellipses
-				processed_xhtml = regex.sub(r"(<[a-z0-9]+[^<]+?>)[\s{}]?…".format(NBSP), r"\1…", processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"(<[a-z0-9]+[^<]+?>)[\s{}]?…".format(se.NO_BREAK_SPACE), r"\1…", processed_xhtml, flags=regex.IGNORECASE)
 
 				# Remove spaces between closing tags and ellipses
-				processed_xhtml = regex.sub(r"…[\s{}]?(</[a-z0-9]+>)".format(NBSP), r"…\1", processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"…[\s{}]+([\)”’])".format(NBSP), r"…\1", processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"([\(“‘])[\s{}]+…".format(NBSP), r"\1…", processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"…[\s{}]?([\!\?\.\;\,])".format(NBSP), r"…{}\1".format(HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"([\!\?\.\;”’])[\s{}]?…".format(NBSP), r"\1{}…".format(HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"\,[\s{}]?…".format(NBSP), r",{}…".format(HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"…[\s{}]?(</[a-z0-9]+>)".format(se.NO_BREAK_SPACE), r"…\1", processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"…[\s{}]+([\)”’])".format(se.NO_BREAK_SPACE), r"…\1", processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"([\(“‘])[\s{}]+…".format(se.NO_BREAK_SPACE), r"\1…", processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"…[\s{}]?([\!\?\.\;\,])".format(se.NO_BREAK_SPACE), r"…{}\1".format(se.HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"([\!\?\.\;”’])[\s{}]?…".format(se.NO_BREAK_SPACE), r"\1{}…".format(se.HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"\,[\s{}]?…".format(se.NO_BREAK_SPACE), r",{}…".format(se.HAIR_SPACE), processed_xhtml, flags=regex.IGNORECASE)
 
 				# Remove spaces between ellipses and endnotes directly after
-				processed_xhtml = regex.sub(r"…[\s{}]?(<a[^>]+?id=\"note-[0-9]+\"[^>]*?>)".format(NBSP), r"…\1", processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"…[\s{}]?(<a[^>]+?id=\"note-[0-9]+\"[^>]*?>)".format(se.NO_BREAK_SPACE), r"…\1", processed_xhtml, flags=regex.IGNORECASE)
 
 				# Add non-breaking spaces between amounts with an abbreviated unit.  E.g. 8 oz., 10 lbs.
-				processed_xhtml = regex.sub(r"([0-9])\s+([a-z]{1,3}\.)", r"\1{}\2".format(NBSP), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"([0-9])\s+([a-z]{1,3}\.)", r"\1{}\2".format(se.NO_BREAK_SPACE), processed_xhtml, flags=regex.IGNORECASE)
 
 				# Add non-breaking spaces between Arabic numbers and AM/PM
-				processed_xhtml = regex.sub(r"([0-9])\s+([ap])\.m\.", r"\1{}\2.m.".format(NBSP), processed_xhtml, flags=regex.IGNORECASE)
-				processed_xhtml = regex.sub(r"([0-9])\s+<abbr([^>]*?)>([ap])\.m\.", r"\1{}<abbr\2>\3.m.".format(NBSP), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"([0-9])\s+([ap])\.m\.", r"\1{}\2.m.".format(se.NO_BREAK_SPACE), processed_xhtml, flags=regex.IGNORECASE)
+				processed_xhtml = regex.sub(r"([0-9])\s+<abbr([^>]*?)>([ap])\.m\.", r"\1{}<abbr\2>\3.m.".format(se.NO_BREAK_SPACE), processed_xhtml, flags=regex.IGNORECASE)
 
 				processed_xhtml = processed_xhtml.replace("Ph.D", "PhD")
 				processed_xhtml = regex.sub(r"P\.\s*S\.", r"P.S.", processed_xhtml)
@@ -212,5 +209,6 @@ def main():
 		if args.verbose:
 			print(" OK")
 
+
 if __name__ == "__main__":
 	main()

unicode-names

@@ -4,6 +4,7 @@ import argparse
 import sys
 import unicodedata
 
+
 def main():
 	parser = argparse.ArgumentParser(description="Display Unicode code points, descriptions, and links to more details for each character in a string.  Useful for differentiating between different flavors of spaces, dashes, and invisible characters like word joiners.")
 	parser.add_argument("strings", metavar="STRING", nargs="*", help="a Unicode string")
@@ -22,5 +23,6 @@ def main():
 		for character in line:
 			print(character + "\tU+{:04X}".format(ord(character)) + "\t" + unicodedata.name(character) + "\t" + "http://unicode.org/cldr/utility/character.jsp?a={:04X}".format(ord(character)))
 
+
 if __name__ == "__main__":
 	main()