def dump(path, strip, meta, content):
"""Dump Tika extraction results for PATH (file or url path)."""
media_type = clean_mime(detector.from_file(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 test_clean_mime():
assert utils.clean_mime("") == ""
assert utils.clean_mime("text/html ") == "text/html"
assert utils.clean_mime(["text/html", "audio/mp3"]) == "text/html"
assert utils.clean_mime([" text/html", "audio/mp3"]) == "text/html"
assert utils.clean_mime(
" text/plain; charset=windows-1252 ") == "text/plain"
assert (utils.clean_mime(
[" text/plain; charset=windows-1252 ", "audio/mp3"]) == "text/plain")
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 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 = clean_mime(detector.from_file(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 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 = clean_mime(detector.from_file(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