def modernize_spelling(plain_output: bool) -> int:
"""
Entry point for `se modernize-spelling`
"""
parser = argparse.ArgumentParser(description="Modernize spelling of some archaic words, and replace words that may be archaically compounded with a dash to a more modern spelling. For example, replace `ash-tray` with `ashtray`.")
parser.add_argument("-n", "--no-hyphens", dest="modernize_hyphenation", action="store_false", help="don’t modernize hyphenation")
parser.add_argument("-v", "--verbose", action="store_true", help="increase output verbosity")
parser.add_argument("targets", metavar="TARGET", nargs="+", help="an XHTML file, or a directory containing XHTML files")
args = parser.parse_args()
return_code = 0
console = Console(highlight=False, theme=se.RICH_THEME, force_terminal=se.is_called_from_parallel()) # Syntax highlighting will do weird things when printing paths; force_terminal prints colors when called from GNU Parallel
for filename in se.get_target_filenames(args.targets, ".xhtml"):
if args.verbose:
console.print(se.prep_output(f"Processing [path][link=file://{filename}]{filename}[/][/] ...", plain_output), end="")
try:
with open(filename, "r+", encoding="utf-8") as file:
xhtml = file.read()
try:
new_xhtml = se.spelling.modernize_spelling(xhtml)
problem_spellings = se.spelling.detect_problem_spellings(xhtml)
for problem_spelling in problem_spellings:
console.print(se.prep_output(f"{('[path][link=file://' + str(filename) + ']' + filename.name + '[/][/]') + ': ' if not args.verbose else ''}{problem_spelling}", plain_output))
except se.InvalidLanguageException as ex:
se.print_error(f"{ex}{' File: [path][link=file://' + str(filename) + ']' + str(filename) + '[/][/]' if not args else ''}", plain_output=plain_output)
return ex.code
if args.modernize_hyphenation:
new_xhtml = se.spelling.modernize_hyphenation(new_xhtml)
if new_xhtml != xhtml:
file.seek(0)
file.write(new_xhtml)
file.truncate()
except FileNotFoundError:
se.print_error(f"Couldn’t open file: [path][link=file://{filename}]{filename}[/][/].", plain_output=plain_output)
return_code = se.InvalidInputException.code
if args.verbose:
console.print(" OK")
return return_code
def extract_ebook(plain_output: bool) -> int:
"""
Entry point for `se extract-ebook`
"""
parser = argparse.ArgumentParser(description="Extract an .epub, .mobi, or .azw3 ebook into ./FILENAME.extracted/ or a target directory.")
parser.add_argument("-o", "--output-dir", type=str, help="a target directory to extract into")
parser.add_argument("-v", "--verbose", action="store_true", help="increase output verbosity")
parser.add_argument("targets", metavar="TARGET", nargs="+", help="an epub, mobi, or azw3 file")
args = parser.parse_args()
console = Console(highlight=False, theme=se.RICH_THEME, force_terminal=se.is_called_from_parallel()) # Syntax highlighting will do weird things when printing paths; force_terminal prints colors when called from GNU Parallel
if args.output_dir and len(args.targets) > 1:
se.print_error("The [bash]--output-dir[/] option can’t be used when more than one ebook target is specified.", plain_output=plain_output)
return se.InvalidArgumentsException.code
for target in args.targets:
target = Path(target).resolve()
if args.verbose:
console.print(se.prep_output(f"Processing [path][link=file://{target}]{target}[/][/] ...", plain_output), end="")
if not path.isfile(target):
se.print_error(f"Not a file: [path][link=file://{target}]{target}[/][/].", plain_output=plain_output)
return se.InvalidInputException.code
if args.output_dir is None:
extracted_path = Path(target.name + ".extracted")
else:
extracted_path = Path(args.output_dir)
if extracted_path.exists():
se.print_error(f"Directory already exists: [path][link=file://{extracted_path}]{extracted_path}[/][/].", plain_output=plain_output)
return se.FileExistsException.code
with open(target, "rb") as binary_file:
file_bytes = binary_file.read()
if _is_mobi(file_bytes):
# kindleunpack uses print() so just capture that output here
old_stdout = sys.stdout
sys.stdout = TextIOWrapper(BytesIO(), sys.stdout.encoding)
kindleunpack.unpackBook(str(target), str(extracted_path))
# Restore stdout
sys.stdout.close()
sys.stdout = old_stdout
elif _is_epub(file_bytes):
with zipfile.ZipFile(target, "r") as file:
file.extractall(extracted_path)
else:
se.print_error("File doesn’t look like an epub, mobi, or azw3 file.")
return se.InvalidFileException.code
if args.verbose:
console.print(" OK")
return 0
def hyphenate(plain_output: bool) -> int:
"""
Entry point for `se hyphenate`
"""
parser = argparse.ArgumentParser(description="Insert soft hyphens at syllable breaks in XHTML files.")
parser.add_argument("-i", "--ignore-h-tags", action="store_true", help="don’t add soft hyphens to text in <h1-6> tags")
parser.add_argument("-l", "--language", action="store", help="specify the language for the XHTML files; if unspecified, defaults to the `xml:lang` or `lang` attribute of the root <html> element")
parser.add_argument("-v", "--verbose", action="store_true", help="increase output verbosity")
parser.add_argument("targets", metavar="TARGET", nargs="+", help="an XHTML file, or a directory containing XHTML files")
args = parser.parse_args()
console = Console(highlight=False, theme=se.RICH_THEME, force_terminal=se.is_called_from_parallel()) # Syntax highlighting will do weird things when printing paths; force_terminal prints colors when called from GNU Parallel
for filename in se.get_target_filenames(args.targets, ".xhtml"):
if args.verbose:
console.print(se.prep_output(f"Processing [path][link=file://{filename}]{filename}[/][/] ...", plain_output), end="")
with open(filename, "r+", encoding="utf-8") as file:
xhtml = file.read()
is_ignored, dom = se.get_dom_if_not_ignored(xhtml, ["toc"])
if not is_ignored and dom:
processed_xhtml = se.typography.hyphenate(dom, args.language, args.ignore_h_tags)
if processed_xhtml != xhtml:
file.seek(0)
file.write(processed_xhtml)
file.truncate()
if args.verbose:
console.print(" OK")
return 0
def renumber_endnotes(plain_output: bool) -> int:
"""
Entry point for `se renumber-endnotes`
"""
parser = argparse.ArgumentParser(
description=
"Renumber all endnotes and noterefs sequentially from the beginning, taking care to match noterefs and endnotes if possible."
)
parser.add_argument(
"-b",
"--brute-force",
action="store_true",
help=
"renumber without checking that noterefs and endnotes match; may result in endnotes with empty backlinks or noterefs without matching endnotes"
)
parser.add_argument("-v",
"--verbose",
action="store_true",
help="increase output verbosity")
parser.add_argument("directories",
metavar="DIRECTORY",
nargs="+",
help="a Standard Ebooks source directory")
args = parser.parse_args()
return_code = 0
for directory in args.directories:
try:
se_epub = SeEpub(directory)
except se.SeException as ex:
se.print_error(ex, plain_output=plain_output)
return_code = ex.code
return return_code
try:
if args.brute_force:
se_epub.recreate_endnotes()
else:
found_endnote_count, changed_endnote_count = se_epub.generate_endnotes(
)
if args.verbose:
print(
se.prep_output(
f"Found {found_endnote_count} endnote{'s' if found_endnote_count != 1 else ''} and changed {changed_endnote_count} endnote{'s' if changed_endnote_count != 1 else ''}.",
plain_output))
except se.SeException as ex:
se.print_error(ex, plain_output=plain_output)
return_code = ex.code
except FileNotFoundError:
se.print_error("Couldn’t find [path]endnotes.xhtml[/].",
plain_output=plain_output)
return_code = se.InvalidSeEbookException.code
return return_code
def semanticate(plain_output: bool) -> int:
"""
Entry point for `se semanticate`
"""
parser = argparse.ArgumentParser(
description=
"Automatically add semantics to Standard Ebooks source directories.")
parser.add_argument("-v",
"--verbose",
action="store_true",
help="increase output verbosity")
parser.add_argument(
"targets",
metavar="TARGET",
nargs="+",
help="an XHTML file, or a directory containing XHTML files")
args = parser.parse_args()
console = Console(
highlight=False,
theme=se.RICH_THEME,
force_terminal=se.is_called_from_parallel()
) # Syntax highlighting will do weird things when printing paths; force_terminal prints colors when called from GNU Parallel
return_code = 0
for filename in se.get_target_filenames(args.targets, ".xhtml"):
if args.verbose:
console.print(se.prep_output(
f"Processing [path][link=file://{filename}]{filename}[/][/] ...",
plain_output),
end="")
try:
with open(filename, "r+", encoding="utf-8") as file:
xhtml = file.read()
is_ignored, _ = se.get_dom_if_not_ignored(xhtml)
if not is_ignored:
processed_xhtml = se.formatting.semanticate(xhtml)
if processed_xhtml != xhtml:
file.seek(0)
file.write(processed_xhtml)
file.truncate()
except FileNotFoundError:
se.print_error(
f"Couldn’t open file: [path][link=file://{filename}]{filename}[/][/].",
plain_output=plain_output)
return_code = se.InvalidInputException.code
if args.verbose:
console.print(" OK")
return return_code
def lint(plain_output: bool) -> int:
"""
Entry point for `se lint`
"""
parser = argparse.ArgumentParser(description="Check for various Standard Ebooks style errors.")
parser.add_argument("-s", "--skip-lint-ignore", action="store_true", help="ignore rules in se-lint-ignore.xml file")
parser.add_argument("-v", "--verbose", action="store_true", help="increase output verbosity")
parser.add_argument("directories", metavar="DIRECTORY", nargs="+", help="a Standard Ebooks source directory")
args = parser.parse_args()
called_from_parallel = se.is_called_from_parallel(False)
force_terminal = True if called_from_parallel else None # True will force colors, None will guess whether colors are enabled, False will disable colors
first_output = True
return_code = 0
# Rich needs to know the terminal width in order to format tables.
# If we're called from Parallel, there is no width because Parallel is not a terminal. Thus we must export $COLUMNS before
# invoking Parallel, and then get that value here.
console = Console(width=int(os.environ["COLUMNS"]) if called_from_parallel and "COLUMNS" in os.environ else None, highlight=False, theme=se.RICH_THEME, force_terminal=force_terminal) # Syntax highlighting will do weird things when printing paths; force_terminal prints colors when called from GNU Parallel
for directory in args.directories:
directory = Path(directory).resolve()
messages = []
exception = None
table_data = []
has_output = False
try:
se_epub = SeEpub(directory)
messages = se_epub.lint(args.skip_lint_ignore)
except se.SeException as ex:
exception = ex
if len(args.directories) > 1:
return_code = se.LintFailedException.code
else:
return_code = ex.code
# Print a separator newline if more than one table is printed
if not first_output and (args.verbose or messages or exception):
console.print("")
elif first_output:
first_output = False
# Print the table header
if ((len(args.directories) > 1 or called_from_parallel) and (messages or exception)) or args.verbose:
has_output = True
if plain_output:
console.print(directory)
else:
console.print(f"[reverse][path][link=file://{directory}]{directory}[/][/][/reverse]")
if exception:
has_output = True
se.print_error(exception, plain_output=plain_output)
# Print the tables
if messages:
has_output = True
return_code = se.LintFailedException.code
if plain_output:
for message in messages:
label = "[Manual Review]"
if message.message_type == se.MESSAGE_TYPE_ERROR:
label = "[Error]"
# Replace color markup with `
message.text = se.prep_output(message.text, True)
message_filename = ""
if message.filename:
message_filename = message.filename.name
console.print(f"{message.code} {label} {message_filename} {message.text}")
if message.submessages:
for submessage in message.submessages:
# Indent each line in case we have a multi-line submessage
console.print(regex.sub(r"^", "\t", submessage, flags=regex.MULTILINE))
else:
for message in messages:
alert = "[bright_yellow]Manual Review[/bright_yellow]"
if message.message_type == se.MESSAGE_TYPE_ERROR:
alert = "[bright_red]Error[/bright_red]"
# Add hyperlinks around message filenames
message_filename = ""
if message.filename:
message_filename = f"[link=file://{message.filename.resolve()}]{message.filename.name}[/link]"
table_data.append([message.code, alert, message_filename, message.text])
if message.submessages:
for submessage in message.submessages:
# Brackets don't need to be escaped in submessages if we instantiate them in Text()
submessage_object = Text(submessage, style="dim")
table_data.append([" ", " ", Text("→", justify="right"), submessage_object])
table = Table(show_header=True, header_style="bold", show_lines=True, expand=True)
table.add_column("Code", width=5, no_wrap=True)
table.add_column("Severity", no_wrap=True)
table.add_column("File", no_wrap=True)
table.add_column("Message", ratio=10)
for row in table_data:
table.add_row(row[0], row[1], row[2], row[3])
console.print(table)
if args.verbose and not messages and not exception:
if plain_output:
console.print("OK")
else:
table = Table(show_header=False, box=box.SQUARE)
table.add_column("", style="white on green4 bold")
table.add_row("OK")
console.print(table)
# Print a newline if we're called from parallel and we just printed something, to
# better visually separate output blocks
if called_from_parallel and has_output:
console.print("")
return return_code
def typogrify(plain_output: bool) -> int:
"""
Entry point for `se typogrify`
"""
parser = argparse.ArgumentParser(
description=
"Apply some scriptable typography rules from the Standard Ebooks typography manual to XHTML files."
)
parser.add_argument(
"-n",
"--no-quotes",
dest="quotes",
action="store_false",
help="don’t convert to smart quotes before doing other adjustments")
parser.add_argument("-v",
"--verbose",
action="store_true",
help="increase output verbosity")
parser.add_argument(
"targets",
metavar="TARGET",
nargs="+",
help="an XHTML file, or a directory containing XHTML files")
args = parser.parse_args()
console = Console(
highlight=False,
theme=se.RICH_THEME,
force_terminal=se.is_called_from_parallel()
) # Syntax highlighting will do weird things when printing paths; force_terminal prints colors when called from GNU Parallel
return_code = 0
for filename in se.get_target_filenames(args.targets, (".xhtml", ".opf")):
if args.verbose:
console.print(se.prep_output(
f"Processing [path][link=file://{filename}]{filename}[/][/] ...",
plain_output),
end="")
try:
with open(filename, "r+", encoding="utf-8") as file:
xhtml = file.read()
is_ignored, dom = se.get_dom_if_not_ignored(
xhtml, ["titlepage", "imprint", "copyright-page"])
if not is_ignored:
if dom:
# Is this a metadata file?
# Typogrify metadata except for URLs, dates, and LoC subjects
if dom.xpath("/package"):
for node in dom.xpath(
"/package/metadata/dc:*[normalize-space(.) and local-name() != 'subject' and local-name() != 'source' and local-name() != 'date']"
) + dom.xpath(
"/package/metadata/meta[normalize-space(.) and (not(contains(@property, 'se:url') or @property = 'dcterms:modified' or @property = 'se:production-notes'))]"
):
node.text = html.unescape(node.text)
node.text = se.typography.typogrify(node.text)
# Tweak: Word joiners and nbsp don't go in metadata
node.text = node.text.replace(
se.WORD_JOINER, "")
node.text = node.text.replace(
se.NO_BREAK_SPACE, " ")
# Typogrify escapes ampersands, and then lxml will also escape them again, so we unescape them
# before passing to lxml.
if node.get_attr(
"property") != "se:long-description":
node.text = node.text.replace(
"&", "&").strip()
processed_xhtml = dom.to_string()
else:
processed_xhtml = se.typography.typogrify(
xhtml, args.quotes)
# Tweak: Word joiners and nbsp don't go in the ToC
if dom.xpath(
"/html/body//nav[contains(@epub:type, 'toc')]"
):
processed_xhtml = processed_xhtml.replace(
se.WORD_JOINER, "")
processed_xhtml = processed_xhtml.replace(
se.NO_BREAK_SPACE, " ")
else:
processed_xhtml = se.typography.typogrify(
xhtml, args.quotes)
if processed_xhtml != xhtml:
file.seek(0)
file.write(processed_xhtml)
file.truncate()
if args.verbose:
console.print(" OK")
except FileNotFoundError:
se.print_error(
f"Couldn’t open file: [path][link=file://{filename}]{filename}[/][/].",
plain_output=plain_output)
return_code = se.InvalidInputException.code
return return_code
def build_images(plain_output: bool) -> int:
"""
Entry point for `se build-images`
"""
parser = argparse.ArgumentParser(
description=
"Build ebook covers and titlepages for a Standard Ebook source directory, and place the output in DIRECTORY/src/epub/images/."
)
parser.add_argument("-v",
"--verbose",
action="store_true",
help="increase output verbosity")
parser.add_argument("directories",
metavar="DIRECTORY",
nargs="+",
help="a Standard Ebooks source directory")
args = parser.parse_args()
console = Console(
highlight=False,
theme=se.RICH_THEME,
force_terminal=se.is_called_from_parallel()
) # Syntax highlighting will do weird things when printing paths; force_terminal prints colors when called from GNU Parallel
for directory in args.directories:
directory = Path(directory).resolve()
if args.verbose:
console.print(
se.prep_output(
f"Processing [path][link=file://{directory}]{directory}[/][/] ...",
plain_output))
try:
se_epub = SeEpub(directory)
if args.verbose:
console.print("\tCleaning metadata ...", end="")
# Remove useless metadata from cover source files
for file_path in directory.glob("**/cover.*"):
se.images.remove_image_metadata(file_path)
if args.verbose:
console.print(" OK")
console.print(se.prep_output(
f"\tBuilding [path][link=file://{directory / 'src/epub/images/cover.svg'}]cover.svg[/][/] ...",
plain_output),
end="")
se_epub.generate_cover_svg()
if args.verbose:
console.print(" OK")
console.print(se.prep_output(
f"\tBuilding [path][link=file://{directory / 'src/epub/images/titlepage.svg'}]titlepage.svg[/][/] ...",
plain_output),
end="")
se_epub.generate_titlepage_svg()
if args.verbose:
console.print(" OK")
except se.SeException as ex:
se.print_error(ex)
return ex.code
return 0
def build(plain_output: bool) -> int:
"""
Entry point for `se build`
"""
parser = argparse.ArgumentParser(description="Build compatible .epub and advanced .epub ebooks from a Standard Ebook source directory. Output is placed in the current directory, or the target directory with --output-dir.")
parser.add_argument("-b", "--kobo", dest="build_kobo", action="store_true", help="also build a .kepub.epub file for Kobo")
parser.add_argument("-c", "--check", action="store_true", help="use epubcheck to validate the compatible .epub file, and the Nu Validator (v.Nu) to validate XHTML5; if Ace is installed, also validate using Ace; if --kindle is also specified and epubcheck, v.Nu, or Ace fail, don’t create a Kindle file")
parser.add_argument("-k", "--kindle", dest="build_kindle", action="store_true", help="also build an .azw3 file for Kindle")
parser.add_argument("-o", "--output-dir", metavar="DIRECTORY", type=str, default="", help="a directory to place output files in; will be created if it doesn’t exist")
parser.add_argument("-p", "--proof", dest="proof", action="store_true", help="insert additional CSS rules that are helpful for proofreading; output filenames will end in .proof")
parser.add_argument("-v", "--verbose", action="store_true", help="increase output verbosity")
parser.add_argument("-y", "--check-only", action="store_true", help="run tests used by --check but don’t output any ebook files and exit after checking")
parser.add_argument("directories", metavar="DIRECTORY", nargs="+", help="a Standard Ebooks source directory")
args = parser.parse_args()
called_from_parallel = se.is_called_from_parallel(False)
force_terminal = True if called_from_parallel else None # True will force colors, None will guess whether colors are enabled, False will disable colors
first_output = True
return_code = 0
# Rich needs to know the terminal width in order to format tables.
# If we're called from Parallel, there is no width because Parallel is not a terminal. Thus we must export $COLUMNS before
# invoking Parallel, and then get that value here.
console = Console(width=int(os.environ["COLUMNS"]) if called_from_parallel and "COLUMNS" in os.environ else None, highlight=False, theme=se.RICH_THEME, force_terminal=force_terminal) # Syntax highlighting will do weird things when printing paths; force_terminal prints colors when called from GNU Parallel
if args.check_only and (args.check or args.build_kindle or args.build_kobo or args.proof or args.output_dir):
se.print_error("The [bash]--check-only[/] option can’t be combined with any other flags except for [bash]--verbose[/].", plain_output=plain_output)
return se.InvalidArgumentsException.code
for directory in args.directories:
directory = Path(directory).resolve()
messages = []
exception = None
table_data = []
has_output = False
try:
se_epub = SeEpub(directory)
se_epub.build(args.check, args.check_only, args.build_kobo, args.build_kindle, Path(args.output_dir), args.proof)
except se.BuildFailedException as ex:
exception = ex
messages = ex.messages
except se.SeException as ex:
se.print_error(ex, plain_output=plain_output)
# Print a separator newline if more than one table is printed
if not first_output and (args.verbose or messages or exception):
console.print("")
elif first_output:
first_output = False
# Print the table header
if ((len(args.directories) > 1 or called_from_parallel) and (messages or exception)) or args.verbose:
has_output = True
if plain_output:
console.print(directory)
else:
console.print(f"[reverse][path][link=file://{directory}]{directory}[/][/][/reverse]")
if exception:
has_output = True
se.print_error(exception, plain_output=plain_output)
# Print the tables
if messages:
has_output = True
return_code = se.BuildFailedException.code
if plain_output:
for message in messages:
# Replace color markup with `
message.text = se.prep_output(message.text, True)
message_filename = ""
if message.filename:
message_filename = message.filename.name
console.print(f"{message.source}: {message.code} {message_filename}{message.location if message.location else ''} {message.text}")
else:
for message in messages:
# Add hyperlinks around message filenames
message_filename = ""
if message.filename:
message_filename = f"[link=file://{message.filename}]{message.filename.name}[/link]{message.location if message.location else ''}"
table_data.append([message.source, message.code, message_filename, message.text])
if message.submessages:
for submessage in message.submessages:
# Brackets don't need to be escaped in submessages if we instantiate them in Text()
submessage_object = Text(submessage, style="dim")
table_data.append([" ", " ", Text("→", justify="right"), submessage_object])
table = Table(show_header=True, header_style="bold", show_lines=True, expand=True)
table.add_column("Source", width=9, no_wrap=True)
table.add_column("Code", no_wrap=True)
table.add_column("File", no_wrap=True)
table.add_column("Message", ratio=10)
for row in table_data:
table.add_row(row[0], row[1], row[2], row[3])
console.print(table)
if args.verbose and not messages and not exception:
if plain_output:
console.print("OK")
else:
table = Table(show_header=False, box=box.SQUARE)
table.add_column("", style="white on green4 bold")
table.add_row("OK")
console.print(table)
# Print a newline if we're called from parallel and we just printed something, to
# better visually separate output blocks
if called_from_parallel and has_output:
console.print("")
return return_code
def xpath(plain_output: bool) -> int:
"""
Entry point for `se xpath`
"""
parser = argparse.ArgumentParser(
description=
"Print the results of an xpath expression evaluated against a set of XHTML files. The default namespace is removed."
)
parser.add_argument(
"-f",
"--only-filenames",
action="store_true",
help=
"only output filenames of files that contain matches, not the matches themselves"
)
parser.add_argument("xpath", metavar="XPATH", help="an xpath expression")
parser.add_argument(
"targets",
metavar="TARGET",
nargs="+",
help="an XHTML file, or a directory containing XHTML files")
args = parser.parse_args()
console = Console(highlight=True, theme=se.RICH_THEME)
for filepath in se.get_target_filenames(args.targets, ".xhtml"):
try:
with open(filepath, "r", encoding="utf-8") as file:
dom = se.easy_xml.EasyXmlTree(file.read())
nodes = dom.xpath(args.xpath)
if nodes:
console.print(se.prep_output(
f"[path][link=file://{filepath}]{filepath}[/][/]",
plain_output),
highlight=False)
if not args.only_filenames:
for node in nodes:
if isinstance(node, se.easy_xml.EasyXmlElement):
output = node.to_string()
else:
# We may select text() nodes as a result
output = str(node)
output = "".join(
[f"\t{line}\n" for line in output.splitlines()])
# We only have to escape leading [ to prevent Rich from converting
# it to a style. If we also escape ] then Rich will print the slash.
output = output.replace("[", "\\[")
console.print(output)
except etree.XPathEvalError as ex:
se.print_error("Invalid xpath expression.")
return se.InvalidInputException.code
except se.SeException as ex:
se.print_error(
f"File: [path][link=file://{filepath}]{filepath}[/][/]. Exception: {ex}",
plain_output=plain_output)
return ex.code
return 0
def clean(plain_output: bool) -> int:
"""
Entry point for `se clean`
"""
parser = argparse.ArgumentParser(
description=
"Prettify and canonicalize individual XHTML, SVG, or CSS files, or all XHTML, SVG, or CSS files in a source directory."
)
parser.add_argument("-v",
"--verbose",
action="store_true",
help="increase output verbosity")
parser.add_argument(
"targets",
metavar="TARGET",
nargs="+",
help=
"an XHTML, SVG, or CSS file, or a directory containing XHTML, SVG, or CSS files"
)
args = parser.parse_args()
console = Console(
highlight=False,
theme=se.RICH_THEME,
force_terminal=se.is_called_from_parallel()
) # Syntax highlighting will do weird things when printing paths; force_terminal prints colors when called from GNU Parallel
for filepath in se.get_target_filenames(
args.targets, (".xhtml", ".svg", ".opf", ".ncx", ".xml", ".css")):
if args.verbose:
console.print(se.prep_output(
f"Processing [path][link=file://{filepath}]{filepath}[/][/] ...",
plain_output),
end="")
if filepath.suffix == ".css":
with open(filepath, "r+", encoding="utf-8") as file:
css = file.read()
try:
processed_css = se.formatting.format_css(css)
if processed_css != css:
file.seek(0)
file.write(processed_css)
file.truncate()
except se.SeException as ex:
se.print_error(
f"File: [path][link=file://{filepath}]{filepath}[/][/]. Exception: {ex}",
args.verbose,
plain_output=plain_output)
return ex.code
else:
try:
se.formatting.format_xml_file(filepath)
except se.MissingDependencyException as ex:
se.print_error(ex, plain_output=plain_output)
return ex.code
except se.SeException as ex:
se.print_error(
f"File: [path][link=file://{filepath}]{filepath}[/][/]. Exception: {ex}",
args.verbose,
plain_output=plain_output)
return ex.code
if args.verbose:
console.print(" OK")
return 0
def british2american(plain_output: bool) -> int:
"""
Entry point for `se british2american`
"""
parser = argparse.ArgumentParser(
description=
"Try to convert British quote style to American quote style. Quotes must already be typogrified using the `typogrify` tool. This script isn’t perfect; proofreading is required, especially near closing quotes near to em-dashes."
)
parser.add_argument("-f",
"--force",
action="store_true",
help="force conversion of quote style")
parser.add_argument("-v",
"--verbose",
action="store_true",
help="increase output verbosity")
parser.add_argument(
"targets",
metavar="TARGET",
nargs="+",
help="an XHTML file, or a directory containing XHTML files")
args = parser.parse_args()
return_code = 0
console = Console(
highlight=False,
theme=se.RICH_THEME,
force_terminal=se.is_called_from_parallel()
) # Syntax highlighting will do weird things when printing paths; force_terminal prints colors when called from GNU Parallel
for filename in se.get_target_filenames(args.targets, ".xhtml"):
# Skip the uncopyright, since it contains quotes but is a fixed file
if filename.name == "uncopyright.xhtml":
continue
if args.verbose:
console.print(se.prep_output(
f"Processing [path][link=file://{filename}]{filename}[/][/] ...",
plain_output),
end="")
try:
with open(filename, "r+", encoding="utf-8") as file:
xhtml = file.read()
new_xhtml = xhtml
convert = True
if not args.force:
if se.typography.guess_quoting_style(xhtml) == "american":
convert = False
if args.verbose:
console.print("")
se.print_error(
f"File appears to already use American quote style, ignoring. Use [bash]--force[/] to convert anyway.{f' File: [path][link=file://{filename}]{filename}[/][/]' if not args.verbose else ''}",
args.verbose,
plain_output=plain_output)
if convert:
new_xhtml = se.typography.convert_british_to_american(
xhtml)
if new_xhtml != xhtml:
file.seek(0)
file.write(new_xhtml)
file.truncate()
except FileNotFoundError:
se.print_error(
f"Couldn’t open file: [path][link=file://{filename}]{filename}[/][/].",
plain_output=plain_output)
return_code = se.InvalidInputException.code
return return_code
def prepare_release(plain_output: bool) -> int:
"""
Entry point for `se prepare-release`
"""
parser = argparse.ArgumentParser(
description=
"Calculate work word count, insert release date if not yet set, and update modified date and revision number."
)
parser.add_argument("-w",
"--no-word-count",
dest="word_count",
action="store_false",
help="don’t calculate word count")
parser.add_argument("-r",
"--no-revision",
dest="revision",
action="store_false",
help="don’t increment the revision number")
parser.add_argument("-v",
"--verbose",
action="store_true",
help="increase output verbosity")
parser.add_argument("directories",
metavar="DIRECTORY",
nargs="+",
help="a Standard Ebooks source directory")
args = parser.parse_args()
console = Console(
highlight=False,
theme=se.RICH_THEME,
force_terminal=se.is_called_from_parallel()
) # Syntax highlighting will do weird things when printing paths; force_terminal prints colors when called from GNU Parallel
for directory in args.directories:
directory = Path(directory).resolve()
if args.verbose:
console.print(
se.prep_output(
f"Processing [path][link=file://{directory}]{directory}[/][/] ...",
plain_output))
try:
se_epub = SeEpub(directory)
if args.word_count:
if args.verbose:
console.print("\tUpdating word count and reading ease ...",
end="")
se_epub.update_word_count()
se_epub.update_flesch_reading_ease()
if args.verbose:
console.print(" OK")
if args.revision:
if args.verbose:
console.print("\tUpdating revision number ...", end="")
se_epub.set_release_timestamp()
if args.verbose:
console.print(" OK")
except se.SeException as ex:
se.print_error(ex, plain_output=plain_output)
return ex.code
return 0
