def dump(path, strip, meta, content):
"""Dump Tika extraction results for PATH (file or url path)."""
media_type = mime_clean(mime_guess(path))
if media_type not in SUPPORTED_MIME_TYPES:
click.echo("Unsupported media type {}.".format(media_type))
click.echo("Please request support at https://github.com/iscc/iscc-cli/issues")
if media_type == "application/x-mobipocket-ebook":
tempdir, epub_filepath = mobi.extract(path)
tika_result = parser.from_file(epub_filepath)
shutil.rmtree(tempdir)
else:
tika_result = parser.from_file(path)
if all([meta, content]):
raise UsageError("Use either --meta or --content for selective output.")
if strip:
tika_result["content"] = tika_result.get("content", "")[:strip]
if meta:
click.echo(json.dumps(tika_result.get("metadata", ""), indent=2))
elif content:
click.echo(json.dumps(tika_result.get("content", ""), indent=2))
else:
click.echo(json.dumps(tika_result, indent=2))
def extract_mobi_folder(bookdir, force=False):
# used lists
# mobi extracts the mobis into some temp dicts and this list will hold
# the paths
templist = []
# convlist will be holding directionary for after the conversion
convlist = []
# create list of all mobis inside the bookdir
mobilist = [
f.path for f in os.scandir(bookdir)
if f.is_file() and os.path.splitext(f)[1] in (
".mobi", ) and not os.path.isdir(os.path.splitext(f)[0])
]
if force:
mobilist = glob(bookdir + "/*.mobi")
# extract the mobis
for f in mobilist:
tempdir, _ = mobi.extract(f)
templist.append(tempdir + "\\mobi7")
# dictiorary names after conversion is just the filename
# minus the extension
for f in mobilist:
convlist.append(os.path.splitext(f)[0])
# copy over the mobi file structure
for i in range(len(templist)):
copy_delcopy(templist[i], convlist[i])
# clean up
for f in templist:
shutil.rmtree(os.path.dirname(f))
return convlist
def test_extract():
for fname in os.listdir(TEST_DIR):
ext = splitext(fname)[-1].upper()
if ext in [".MOBI", ".PRC", ".AZW", ".AZW3", ".AZW4"]:
tempdir, filepath = mobi.extract(join(TEST_DIR, fname))
assert exists(tempdir)
assert exists(filepath)
shutil.rmtree(tempdir)
def main(argv):
input = sys.argv[1]
if len(sys.argv) <= 2:
output = input.replace('.mobi', '_unpacked_mobi')
else:
output = sys.argv[2]
print ('Unpacking MOBI to {}'.format(output))
tempdir, _ = mobi.extract(input)
shutil.move(tempdir, output)
def unpack_mobi(self, filename):
storage_area = self.get_mobi_storage_area()
filemd5 = generate_file_md5(filename)
for fname in storage_area.iterdir():
if fname.is_file() and (fname.stem == filemd5):
return str(fname)
with mute_stdout():
tempdir, extracted_file = mobi.extract(str(filename))
filetype = Path(extracted_file).suffix.strip(".")
if filetype == "html":
return self.create_valid_epub_from_epub_like_structure(
tempdir, storage_area.joinpath(f"{filemd5}.epub"))
dst_filename = storage_area.joinpath(f"{filemd5}.{filetype}")
shutil.copy(extracted_file, dst_filename)
TemporaryDirectory._rmtree(tempdir)
return dst_filename
def parse(dir, outPath):
count = 0
for root, dirs, files in os.walk(dir):
for file in files:
try:
#获取文件路径
path = os.path.join(root, file)
# print("文件路径为"+path)
if path.endswith(".mobi"):
tempdir, filepath = mobi.extract(path)
# filepath里面是一个html
book = open(filepath, 'r', encoding='utf-8')
soup = BeautifulSoup(book.read(), "html.parser")
# 解析好的文件路径
filename = path.split("/")[-1]
currentPath = os.getcwd()
# print("当前路径为"+currentPath)
# savePath = join(currentPath,"out",filename+".txt")
savePath = join(outPath, filename + ".txt")
# print("解析好的文件路径"+savePath)
if os.path.exists(outPath):
pass
else:
os.makedirs(join(currentPath, "out"))
a = open(savePath, 'w', encoding='utf-8')
a.write(soup.text)
a.close()
# print("删除临时目录"+tempdir)
shutil.rmtree(tempdir)
count = count + 1
if count % 10 == 0:
print("处理了" + str(count))
except Exception as a:
print(a)
traceback.print_exc()
continue
def import_mobi_file(self, text_file_path):
tempdir, filepath = mobi.extract(text_file_path)
# If extracted MOBI file has extension TXT or HTML that means that everything worked properly.
if re.search(r'\S+.txt|\S+.html', filepath, re.IGNORECASE):
file = open(filepath, 'r', errors='ignore')
content = file.read()
new_text = html2text.html2text(content.replace('\\n', ''))
self.save_temp_data(new_text)
shutil.rmtree(tempdir, ignore_errors=True)
# In other case (for example extracted file has EPUB format) that means that MOBI file was encrypted and
# content will be corrupted.
else:
self.text_loading_dialog.dismiss()
self.show_instructions(
'Something went wrong :( The file provided cannot be processed. Please try another one.'
)
self.update_text_preview()
self.text_loading_dialog.dismiss()
def batch(path, recursive, guess, debug):
"""Create ISCC Codes for all files in PATH.
Example:
$ iscc batch ~/Documents
"""
if debug:
log.add(sys.stdout)
results = []
for f in get_files(path, recursive=recursive):
filesize = os.path.getsize(f)
if not filesize:
msg = "Cannot proccess empty file: {}".format(f)
log.warning(msg)
continue
media_type = mime_clean(mime_guess(f))
if media_type not in SUPPORTED_MIME_TYPES:
fname = basename(f)
msg = "Unsupported file {} with mime type: {},,,,".format(
fname, media_type)
log.warning(msg)
continue
if media_type == "application/x-mobipocket-ebook":
try:
tempdir, epub_filepath = mobi.extract(f)
tika_result = parser.from_file(epub_filepath)
shutil.rmtree(tempdir)
except Exception as e:
msg = "Error with mobi extraction %s"
log.error(msg)
continue
else:
tika_result = parser.from_file(f)
title = get_title(tika_result, guess=guess, uri=f)
mid, norm_title, _ = iscc.meta_id(title)
gmt = mime_to_gmt(media_type, file_path=f)
if gmt == GMT.IMAGE:
try:
cid = iscc.content_id_image(f)
except Exception as e:
msg = "Clould not proccess image: {} ({})".format(f, e)
log.error(msg)
continue
elif gmt == GMT.TEXT:
text = tika_result["content"]
if not text:
msg = "Could not extract text from {}".format(basename(f))
log.warning(msg)
continue
cid = iscc.content_id_text(tika_result["content"])
elif gmt == GMT.AUDIO:
if not fpcalc.is_installed():
fpcalc.install()
features = audio_id.get_chroma_vector(f)
cid = audio_id.content_id_audio(features)
elif gmt == GMT.VIDEO:
features = video_id.get_frame_vectors(abspath(f))
cid = video_id.content_id_video(features)
else:
log.error("Could not generate ISCC")
continue
did = iscc.data_id(f)
iid, tophash = iscc.instance_id(f)
iscc_code_cs = ",".join((mid, cid, did, iid))
click.echo("{iscc_code},{tophash},{fname},{gmt},{title}".format(
iscc_code=iscc_code_cs,
tophash=tophash,
fname=basename(f),
gmt=gmt,
title=norm_title,
))
iscc_code = "-".join((mid, cid, did, iid))
results.append(
dict(
iscc=iscc_code,
norm_title=norm_title,
tophash=tophash,
gmt=gmt,
file_name=basename(f),
))
return results
def gen(file, guess, title, extra, verbose):
"""Generate ISCC Code for FILE."""
filesize = os.path.getsize(file.name)
if not filesize:
raise click.BadParameter("Cannot proccess empty file: {}".format(
file.name))
media_type = mime_clean(mime_guess(file.name))
if media_type not in SUPPORTED_MIME_TYPES:
click.echo("Unsupported media type {}.".format(media_type))
click.echo(
"Please request support at https://github.com/iscc/iscc-cli/issues"
)
if media_type == "application/x-mobipocket-ebook":
tempdir, epub_filepath = mobi.extract(file.name)
tika_result = parser.from_file(epub_filepath)
shutil.rmtree(tempdir)
else:
tika_result = parser.from_file(file.name)
if not title:
title = get_title(tika_result, guess=guess, uri=file.name)
if not extra:
extra = ""
mid, norm_title, _ = iscc.meta_id(title, extra)
gmt = mime_to_gmt(media_type, file_path=file.name)
if gmt == GMT.IMAGE:
cid = iscc.content_id_image(file.name)
elif gmt == GMT.TEXT:
text = tika_result["content"]
if not text:
click.echo("Could not extract text from {}".format(file.name))
return
cid = iscc.content_id_text(tika_result["content"])
elif gmt == GMT.AUDIO:
if not fpcalc.is_installed():
fpcalc.install()
features = audio_id.get_chroma_vector(file.name)
cid = audio_id.content_id_audio(features)
elif gmt == GMT.VIDEO:
features = video_id.get_frame_vectors(abspath(file.name))
cid = video_id.content_id_video(features)
else:
click.echo("Could not generate ISCC")
return
did = iscc.data_id(file.name)
iid, tophash = iscc.instance_id(file.name)
if not norm_title:
iscc_code = "-".join((cid, did, iid))
else:
iscc_code = "-".join((mid, cid, did, iid))
click.echo("ISCC:{}".format(iscc_code))
if verbose:
if norm_title:
click.echo("Norm Title: %s" % norm_title)
click.echo("Tophash: %s" % tophash)
click.echo("Filepath: %s" % file.name)
click.echo("GMT: %s" % gmt)
return dict(iscc=iscc_code,
norm_title=norm_title,
tophash=tophash,
gmt=gmt)
def web(url, guess, title, extra, verbose):
"""Generate ISCC Code from URL."""
extra = extra or ""
try:
resp = requests.get(url, headers=HEADERS, stream=True)
except Exception as e:
raise click.BadArgumentUsage(e)
data = BytesIO(resp.content)
media_type = clean_mime(detector.from_buffer(data))
if media_type not in SUPPORTED_MIME_TYPES:
click.echo("Unsupported media type {}".format(media_type))
click.echo(
"Please request support at https://github.com/iscc/iscc-cli/issues"
)
return
if media_type == "application/x-mobipocket-ebook":
data.seek(0)
tempdir, filepath = mobi.extract(data)
tika_result = parser.from_file(filepath)
shutil.rmtree(tempdir)
else:
data.seek(0)
tika_result = parser.from_buffer(data)
if not title:
title = get_title(tika_result, guess=guess, uri=url)
mid, norm_title, _ = iscc.meta_id(title, extra)
gmt = mime_to_gmt(media_type)
if gmt == GMT.IMAGE:
data.seek(0)
cid = iscc.content_id_image(data)
elif gmt == GMT.TEXT:
text = tika_result["content"]
if not text:
click.echo("Could not extract text")
return
cid = iscc.content_id_text(tika_result["content"])
elif gmt == GMT.AUDIO:
if not fpcalc.is_installed():
fpcalc.install()
data.seek(0)
features = audio_id.get_chroma_vector(data)
cid = audio_id.content_id_audio(features)
elif gmt == GMT.VIDEO:
local_path = download_file(url, sanitize=True)
features = video_id.get_frame_vectors(local_path)
cid = video_id.content_id_video(features)
os.remove(local_path)
data.seek(0)
did = iscc.data_id(data)
data.seek(0)
iid, tophash = iscc.instance_id(data)
if not norm_title:
iscc_code = "-".join((cid, did, iid))
else:
iscc_code = "-".join((mid, cid, did, iid))
click.echo("ISCC:{}".format(iscc_code))
if verbose:
if norm_title:
click.echo("Norm Title: %s" % norm_title)
click.echo("Tophash: %s" % tophash)
click.echo("Filepath: %s" % url)
click.echo("GMT: %s" % gmt)
return dict(iscc=iscc_code,
norm_title=norm_title,
tophash=tophash,
gmt=gmt)
def test_extract_file_like():
with open(join(TEST_DIR, "demo.mobi"), "rb") as infile:
tempdir, filepath = mobi.extract(infile)
assert exists(tempdir)
assert exists(filepath)
shutil.rmtree(tempdir)
from bs4 import BeautifulSoup
from weasyprint import HTML, CSS, default_url_fetcher
import mobi
import shutil
import sys
if __name__ == "__main__":
unzip_file_path=sys.argv[1]
tempdir, filepath = mobi.extract(unzip_file_path)
image_base=filepath[:-9]
html=HTML(filename=filepath,base_url=image_base,encoding="utf8")
filename=unzip_file_path.split("/")[-1]
html.write_pdf(filename+'.pdf')
shutil.rmtree(tempdir)
def from_file(file: UploadFile = File(...),
title: str = Form(""),
extra: str = Form("")):
"""Generate Full ISCC Code from Media File with optional explicit metadata."""
media_type = detector.from_buffer(file.file)
if media_type not in SUPPORTED_MIME_TYPES:
raise HTTPException(
HTTP_415_UNSUPPORTED_MEDIA_TYPE,
"Unsupported media type '{}'. Please request support at "
"https://github.com/iscc/iscc-service/issues.".format(media_type),
)
if media_type == "application/x-mobipocket-ebook":
file.file.seek(0)
tempdir, filepath = mobi.extract(file.file)
tika_result = parser.from_file(filepath)
shutil.rmtree(tempdir)
else:
file.file.seek(0)
tika_result = parser.from_buffer(file.file)
if not title:
title = get_title(tika_result, guess=True)
mid, norm_title, norm_extra = iscc.meta_id(title, extra)
gmt = mime_to_gmt(media_type)
if gmt == GMT.IMAGE:
file.file.seek(0)
cid = iscc.content_id_image(file.file)
elif gmt == GMT.TEXT:
text = tika_result["content"]
if not text:
raise HTTPException(HTTP_422_UNPROCESSABLE_ENTITY,
"Could not extract text")
cid = iscc.content_id_text(tika_result["content"])
elif gmt == GMT.AUDIO:
file.file.seek(0)
features = audio_id.get_chroma_vector(file.file)
cid = audio_id.content_id_audio(features)
elif gmt == GMT.VIDEO:
file.file.seek(0)
_, ext = splitext(file.filename)
fn = "{}{}".format(uuid.uuid4(), ext)
tmp_path = join(APP_DIR, fn)
with open(tmp_path, "wb") as buffer:
shutil.copyfileobj(file.file, buffer)
features = video_id.get_frame_vectors(tmp_path)
cid = video_id.content_id_video(features)
os.remove(tmp_path)
file.file.seek(0)
did = iscc.data_id(file.file)
file.file.seek(0)
iid, tophash = iscc.instance_id(file.file)
if not norm_title:
iscc_code = "-".join((cid, did, iid))
else:
iscc_code = "-".join((mid, cid, did, iid))
components = iscc_split(iscc_code)
result = dict(
iscc=iscc_code,
tophash=tophash,
gmt=gmt,
bits=[code_to_bits(c) for c in components],
)
if norm_title:
result["title"] = title
result["title_trimmed"] = norm_title
if norm_extra:
result["extra"] = extra
result["extra_trimmed"] = norm_extra
file.file.close()
return result