ASCIIsite/ASCIIsite.py

160 lines
7.3 KiB
Python
Raw Normal View History

2021-07-14 23:55:45 +02:00
#! /usr/bin/env python3
#takes directory, converts all .adoc files to html files, copying the resulting html files to an identical directory strucuture, and copies over all non .adoc files unchanged. Optionally outputs as a tar.gz file.
import subprocess, sys, argparse, logging, tempfile, shutil, os, glob
2021-07-14 23:55:45 +02:00
from pathlib import Path
logging.basicConfig(format='%(asctime)s:%(message)s', level=logging.INFO)
#logging.basicConfig(format='%(asctime)s:%(message)s', level=logging.DEBUG)
2021-07-14 23:55:45 +02:00
def parse_arguments():
parser=argparse.ArgumentParser(description='create a website directory structure by converting .adoc files in a directory strucutre to .html files.')
2021-07-14 23:55:45 +02:00
parser.add_argument('inputDir', type=Path, help='The directory of adoc files to be copied and converted.')
parser.add_argument('-o', '--output', type=Path, help='What to name the generated directory or tar file')
parser.add_argument('--stylesheet', type=Path, help='A custom CSS file to be applied to the output.')
parser.add_argument('--exclude-file', type=Path, help='A text file containing glob patterns to exclude, 1 per line.')
parser.add_argument('--exclude', nargs='+', help='A list of glob patterns to ignore. Remember to quote them so your shell doesnt escape them!')
parser.add_argument('-z', '--compress', action='store_true', help='whether to compress the resulting directory to a tar.gz file. can be usefull for scripting to transfer the site to a remote server.')
parser.add_argument('-v', '--verbose', action='store_true', help='outputs debug messages onto the console.')
2021-07-14 23:55:45 +02:00
args=parser.parse_args()
#setting log level
if args.verbose:
logging.info('setting log level to verbose')
logging.getLogger().setLevel(level=logging.DEBUG)
2021-08-15 17:10:25 +02:00
#set compress flag
2021-07-15 21:13:00 +02:00
if args.output != None and not args.compress:
2021-07-14 23:55:45 +02:00
#detect based on whether outFile has a .tar.gz filename.
if args.output.suffixes == ['.tar', '.gz']:
compress=True
2021-07-14 23:55:45 +02:00
else:
compress=False
2021-07-14 23:55:45 +02:00
else:
compress=args.compress
2021-07-14 23:55:45 +02:00
2021-08-15 17:10:25 +02:00
#If outfile was not set, set it.
2021-07-14 23:55:45 +02:00
if args.output == None:
baseName=args.inputDir.with_name(args.inputDir.name+'_compiled').name
outFile=Path(os.getcwd()).joinpath(baseName)
2021-07-14 23:55:45 +02:00
else:
outFile=args.output.resolve()
2021-07-14 23:55:45 +02:00
2021-08-15 17:10:25 +02:00
#add .tar.gz if compress is set and the outfile does not already have it.
2021-07-15 21:13:00 +02:00
if compress and outFile.suffixes != ['.tar', '.gz']:
2021-08-15 17:10:25 +02:00
logging.info(f'outFile was {outFile}, corrected because compress flag is set.')
outFile=outFile.with_suffix('.tar.gz').resolve()
2021-07-14 23:55:45 +02:00
2021-08-15 17:10:25 +02:00
if args.inputDir.resolve() == outFile.resolve():
raise FileExistsError('output file cannot have the same path as the input file!')
2021-07-15 21:13:00 +02:00
logging.debug(f'inputing from {args.inputDir.resolve()}')
logging.info(f'outputting to {outFile.resolve()}')
logging.debug(f'compress is {compress}')
exclude=[]
if args.exclude_file != None:
2021-09-27 21:52:02 +02:00
with open(args.exclude_file, 'r') as file:
exclude=[glob.strip() for glob in file]
if args.exclude != None:
exclude.extend(args.exclude)
2021-09-27 21:52:02 +02:00
if not args.inputDir.resolve().exists():
2021-09-27 21:52:02 +02:00
print(f'Inputdir {args.inputDir.resolve()} does not exist!')
exit()
stylesheet=None
if args.stylesheet != None:
stylesheet =args.stylesheet.resolve()
logging.info(f'using stylesheet {stylesheet}')
return args.inputDir.resolve(), outFile, stylesheet, compress, exclude
2021-08-15 17:10:25 +02:00
#Doing it in a tmpDir first, as some distrubutions put temp files on a ramdisk. this should speed up the operation sigificantly.
class TmpDir:
def __init__(self, srcDir, exclude):
self.tmpDir=tempfile.TemporaryDirectory()
logging.debug(f'making tmp file from {srcDir} at {self.tmpDir.name}')
self.path=self.tmpDir.name+'/'+Path(srcDir).resolve().name
self.ignorePatterns=['*.adoc', '.gitignore', '.git/*']
self.ignorePatterns.extend(exclude)
self.ignorePattern=shutil.ignore_patterns(*self.ignorePatterns)
shutil.copytree(srcDir, self.path, ignore=self.ignorePattern, symlinks=False)
2021-08-15 17:10:25 +02:00
#copy out from tmpDir (which may be in RAM, depending on distrubution) to disk
def copy_self_to(self, destPath):
logging.debug(f'outputting to {Path(destPath).resolve()}')
shutil.copytree(self.path, destPath, symlinks=False)
2021-08-15 17:10:25 +02:00
#copy out from tmpDir (which may be in RAM, depending on distrubution) to a compressed file on disk
def compress_and_copy_self_to(self, destPath):
#shutil.make_archive wants destPath to be without file extentions for some godforsaken reason.
destPath=Path(destPath.with_name(destPath.name.split('.')[0])).resolve()
logging.debug(f'compressing to {Path(destPath).resolve()} from {Path(self.path).parent}')
tarFile=shutil.make_archive(destPath, 'gztar', Path(self.path).parent)
def cleanup(self):
self.tmpDir.cleanup()
#works on the current working directory
def find_paths_to_convert(fileNameGlob):
return glob.glob(f'**/{fileNameGlob}', recursive=True)
#finds the depth of a file relative to given directory
def find_relative_file_depth (subfile, parentDir):
subfile=Path(subfile).resolve()
parentDir=Path(parentDir).resolve()
return len(subfile.parts)-len(parentDir.parts)-1
#simple wrapper around the asciidoctor cli.
def convert_file(inDir: Path, outDir: Path, inFile: Path, stylesheet: Path):
#in order for the stylesdir and imagesdir to be linked to correctly, we need to know the relative depth between the two directories.
depth=find_relative_file_depth(inFile, inDir)
2021-09-04 20:11:31 +02:00
logging.info(f'converting {Path(inFile).resolve()}')
logging.debug(f'converting {inFile=}, {outDir=}, {inDir=}, {stylesheet=}')
depthstring= '../'*depth
arguments=['asciidoctor',
#makes the stylesheet linked, but still includes it in the output.
'--attribute=linkcss',
f'--attribute=stylesdir={depthstring}css',
#set imagesdir
f'--attribute=imagesdir={depthstring}images',
#specifies the source directory root.
f'--source-dir={inDir}',
#Destination dir. It takes the file from the subtree --source-dir and puts it in the equivilant location in the subtree --destination-dir. (talking about filesystem subtrees).
f'--destination-dir={outDir}',
inFile]
if stylesheet != None:
arguments.insert(1, f'--attribute=copycss={stylesheet}')
arguments.insert(1, f'--attribute=stylesheet={stylesheet.name}')
else:
arguments.insert(1, f'--attribute=copycss')
logging.debug(f'{arguments=}')
try:
#the destdir can be used instead of destfile in order to preserve the directory structure relative to the base dir. really useful.
subprocess.run(arguments, check=True)
except Exception as e:
logging.error(f'could not convert {inFile}!')
logging.error(f'{e}')
if __name__ == '__main__':
inFile, outFile, stylesheet, compress, exclude=parse_arguments()
os.chdir(inFile)
tmpDir=TmpDir('./', exclude)
pathsToConvert=find_paths_to_convert('*.adoc')
for i in pathsToConvert:
convert_file(inDir='./', outDir=tmpDir.path, inFile=i, stylesheet=stylesheet)
if compress:
tmpDir.compress_and_copy_self_to(outFile)
else:
tmpDir.copy_self_to(outFile)
tmpDir.cleanup()