def _get_handle(path, mode, encoding=None, compression=None, memory_map=False):
"""Gets file handle for given path and mode.
"""
if compression is not None:
if encoding is not None and not compat.PY3:
msg = 'encoding + compression not yet supported in Python 2'
raise ValueError(msg)
if compression == 'gzip':
import gzip
f = gzip.GzipFile(path, mode)
elif compression == 'bz2':
import bz2
f = bz2.BZ2File(path, mode)
elif compression == 'zip':
import zipfile
zip_file = zipfile.ZipFile(path)
zip_names = zip_file.namelist()
if len(zip_names) == 1:
file_name = zip_names.pop()
f = zip_file.open(file_name)
elif len(zip_names) == 0:
raise ValueError(
'Zero files found in ZIP file {}'.format(path))
else:
raise ValueError(
'Multiple files found in ZIP file.'
' Only one file per ZIP :{}'.format(zip_names))
elif compression == 'xz':
lzma = compat.import_lzma()
f = lzma.LZMAFile(path, mode)
else:
raise ValueError('Unrecognized compression type: %s' % compression)
if compat.PY3:
from io import TextIOWrapper
f = TextIOWrapper(f, encoding=encoding)
return f
else:
if compat.PY3:
if encoding:
f = open(path, mode, encoding=encoding)
else:
f = open(path, mode, errors='replace')
else:
f = open(path, mode)
if memory_map and hasattr(f, 'fileno'):
try:
g = MMapWrapper(f)
f.close()
f = g
except Exception:
# we catch any errors that may have occurred
# because that is consistent with the lower-level
# functionality of the C engine (pd.read_csv), so
# leave the file handler as is then
pass
return f
class Snakefile:
def __init__(self, path, rulecount=0):
self.path = path
try:
self.file = open(self.path, encoding="utf-8")
except FileNotFoundError as e:
try:
self.file = TextIOWrapper(urllib.request.urlopen(self.path),
encoding="utf-8")
except (HTTPError, URLError, ContentTooShortError, ValueError):
raise WorkflowError("Failed to open {}.".format(path))
self.tokens = tokenize.generate_tokens(self.file.readline)
self.rulecount = rulecount
self.lines = 0
def __next__(self):
return next(self.tokens)
def __iter__(self):
return self
def __enter__(self):
return self
def __exit__(self, *args):
self.file.close()
def _write_table(profile_dir,
table_name,
rows,
fields,
append=False,
gzip=False):
# don't gzip if empty
rows = iter(rows)
try:
first_row = next(rows)
except StopIteration:
gzip = False
else:
rows = chain([first_row], rows)
if gzip and append:
logging.warning('Appending to a gzip file may result in '
'inefficient compression.')
if not os.path.exists(profile_dir):
raise ItsdbError(
'Profile directory does not exist: {}'.format(profile_dir))
tbl_filename = os.path.join(profile_dir, table_name)
mode = 'a' if append else 'w'
if gzip:
# text mode only from py3.3; until then use TextIOWrapper
#mode += 't' # text mode for gzip
f = TextIOWrapper(gzopen(tbl_filename + '.gz', mode=mode))
else:
f = open(tbl_filename, mode=mode)
for row in rows:
f.write(make_row(row, fields) + '\n')
f.close()
def _generate(self):
""" Generate the Python code. """
needs_close = False
if sys.hexversion >= 0x03000000:
if self._opts.output == '-':
from io import TextIOWrapper
pyfile = TextIOWrapper(sys.stdout.buffer, encoding='utf8')
else:
pyfile = open(self._opts.output, 'wt', encoding='utf8')
needs_close = True
else:
if self._opts.output == '-':
pyfile = sys.stdout
else:
pyfile = open(self._opts.output, 'wt')
needs_close = True
import_from = self._opts.import_from
if import_from:
from_imports = True
elif self._opts.from_imports:
from_imports = True
import_from = '.'
else:
from_imports = False
compileUi(self._ui_file, pyfile, self._opts.execute, self._opts.indent,
from_imports, self._opts.resource_suffix, import_from)
if needs_close:
pyfile.close()
def readline(config: LanguageIndexingConfiguration, default_config: TextIOWrapper) -> None:
title = Title()
name = None
for line in default_config:
# Do nothing if it is a comment or empty line
if line.startswith("#") or line.startswith("\n"):
continue
# Save the name for the section
elif line.startswith("["):
# When reach a new section, save everything before it into config
if name is not None:
setattr(config, name, title)
name = line[1:-2]
title = Title()
# Save each individual data under the section
else:
word = line.split(" : ")
# If there is no value for the key
if len(word) == 1:
setattr(title, word[0].rstrip(), None)
# If there is value for the key
else:
setattr(title, word[0], word[1].rstrip())
setattr(config, name, title)
default_config.close()
setDefault(config)
def get_curlbomb(self, args, script=None, override_defaults={}):
"""Prepare curlbomb to run in a thread
Assumes args has a '{script}' formatter in it to replace a temporary path with
If no '{script}' formatter is found, stdin is mocked through settings['stdin']
Returns tuple(curlbomb_thread, client_command)
"""
if type(script) == str:
script = bytes(script, "utf-8")
stdin = "{script}" not in args and script is not None
try:
log.info("Using stdin: {}".format(stdin))
if stdin:
s = TextIOWrapper(BytesIO(script))
override_defaults['stdin'] = s
else:
s = NamedTemporaryFile()
if script is not None:
s.write(script)
s.flush()
args = args.format(script=s.name)
args = shlex.split(args)
log.warn("starting curlbomb: {}".format(args))
settings = curlbomb.get_settings(args, override_defaults)
client_cmd = settings['get_curlbomb_command'](settings)
curlbomb_thread = CurlbombThread(settings)
curlbomb_thread.start()
return (curlbomb_thread,
client_cmd)
finally:
s.close()
def _write_table(profile_dir, table_name, rows, fields,
append=False, gzip=False):
# don't gzip if empty
rows = iter(rows)
try:
first_row = next(rows)
except StopIteration:
gzip = False
else:
rows = chain([first_row], rows)
if gzip and append:
logging.warning('Appending to a gzip file may result in '
'inefficient compression.')
if not os.path.exists(profile_dir):
raise ItsdbError('Profile directory does not exist: {}'
.format(profile_dir))
tbl_filename = os.path.join(profile_dir, table_name)
mode = 'a' if append else 'w'
if gzip:
# text mode only from py3.3; until then use TextIOWrapper
#mode += 't' # text mode for gzip
f = TextIOWrapper(gzopen(tbl_filename + '.gz', mode=mode))
else:
f = open(tbl_filename, mode=mode)
for row in rows:
f.write(make_row(row, fields) + '\n')
f.close()
class S3GZipCSVReader:
"""Reads a Gzipped CSV file from S3
Downloads and decompresses on-the-fly, so the entire file doesn't have
to be loaded into memory
"""
def __init__(self, s3_config, s3_path, **kwargs):
self.s3file = s3_config.fs_open(_strip_schema(s3_path))
self.gzfile = TextIOWrapper(
GzipFile(fileobj=self.s3file, mode='rb'),
encoding='utf-8',
newline='',
)
self.reader = get_csv_reader(self.gzfile, **kwargs)
def __enter__(self):
return self
def __exit__(self, exc_type, exc_val, exc_tb):
self.close()
def __iter__(self):
return self.reader.__iter__()
def next(self):
return self.reader.next()
def close(self):
self.gzfile.close()
self.s3file.close()
def cpplint_tests(self, filename):
# Run checks directly from cpplint
from io import TextIOWrapper, BytesIO
output_string = ''
real_stderr = sys.stderr
fake_out = TextIOWrapper(BytesIO(), sys.stdout.encoding)
try:
sys.stderr = fake_out
cpplint.ProcessFile(filename, 0)
except TypeError:
print(
'TODO: Fix TypeError: must be unicode, not str in Python 2.x')
finally:
# get output
sys.stderr.seek(0) # jump to the start
output_string = sys.stderr.read() # read output
# restore stderr
sys.stderr = real_stderr
fake_out.close()
# Parse output from cpplint
for line in output_string.splitlines():
result = re.search(filename + r'\:(\d+)\:\s+(.*)', line)
if result:
line_num = int(result.group(1))
self.add_error(label="CPPLINT_ERROR",
line=line_num,
data={'message': result.group(2)})
def query_to_csv(**kwargs):
query = kwargs["query"]
parameters = kwargs.get("parameters", {})
dest_file = kwargs["dest_file"]
chunksize = 5000
oracle = OracleSqlAlchemyHook()
df = oracle.get_pandas_df(query,
parameters=parameters,
chunksize=chunksize)
access_key = s3_conn.login
secret = s3_conn.password
bucket_name = s3_conn.extra_dejson["bucket_name"]
s3 = s3fs.S3FileSystem(anon=False, key=access_key, secret=secret)
logger = kwargs["ti"].log
logger.info(
f"Running {kwargs['task_instance_key_str']} for {kwargs['ds']}")
if dest_file:
path = f"{bucket_name}/{kwargs['ds']}/{kwargs['task_instance_key_str']}/{dest_file}"
with s3.open(
path,
"wb",
) as csvfile:
first_chunk = True
tcsv = TextIOWrapper(csvfile)
chunk: DataFrame
for idx, chunk in enumerate(df):
logger.info(f"Processing chunk {idx} of {dest_file}")
chunk.to_csv(tcsv,
index=False,
chunksize=chunksize,
header=first_chunk)
first_chunk = False
tcsv.close()
return f"s3://{path}"
def generate_domain_list(pattern_str: str, wordlist_file: TextIOWrapper,
skip_files: list, chunk: int, chunks: int, l: Logger):
if pattern_str is not None:
d = pattern.generate_candiates(pattern_str)
else:
d = [l.rstrip('\n') for l in wordlist_file.readlines()]
wordlist_file.close()
l.log('{} domains to test (initially)'.format(len(d)))
if skip_files is not None:
domains_to_skip = []
for file in skip_files:
domains_to_skip.extend(l.rstrip('\n') for l in file.readlines())
file.close()
d = sorted(set(d) - set(domains_to_skip))
if chunk is not None and chunks is not None:
# find the chunk of the list
l.log("Take chunk {} of {}".format(chunk, chunks))
d = chunking.get(d, chunk, chunks)
length = len(d)
l.log('{} domains to test (after skipping and chunking)'.format(length))
return d, length
def load(self):
"""Load this table's data into Athena."""
data_file_names = self._get_file_names()
districts = sorted(data_file_names.keys())
for district in districts:
district_file_name = data_file_names[district]
with NamedTemporaryFile('w+b') as raw_file:
with gzip.open(raw_file, 'wb') as gzip_file:
text_gzip_file = TextIOWrapper(gzip_file, encoding='utf-8')
self._convert_raw_file(district_file_name, text_gzip_file)
text_gzip_file.close()
self._athena.upload_data(self.name,
raw_file,
district=district)
is_partitioned = None not in districts
ddl = self._generate_ddl(is_partitioned)
self._athena.execute_query(ddl)
self.logger.debug('Ensured table exists for {0}'.format(self.name))
if is_partitioned:
self._athena.execute_query('MSCK REPAIR TABLE {0};'.format(
self.name))
self.logger.debug('Repaired table for {0}'.format(self.name))
self.logger.info('Loaded normal table {0}'.format(self.name))
def load(self, config_file: TextIOWrapper):
if not config_file:
loc = os.path.join(os.path.dirname(__file__), 'pysguard.conf.yaml')
config_file = open(loc, 'r')
self._config = yaml.safe_load(config_file)
config_file.close()
return self
def download(self, outfile: io.TextIOWrapper) -> Result:
if outfile.closed:
result = Result(Status.FILE_CLOSED, outfile)
self.emit('download', result)
return result
if outfile.mode.startswith('r'):
self._req.add_header('Range',
'bytes=%i-' % length(outfile, restore=False))
elif outfile.mode.startswith('a'):
self._req.add_header('Range', 'bytes=%i-' % outfile.tell())
if self._conn is None or self.__conn.closed:
st = self.connect()
if not st:
outfile.close()
return st
else:
self.__conn = st.value
bytes_written = 0
self.emit("download", Result(Status.OK, outfile))
for chunk in self:
self.emit('data', chunk)
outfile.write(chunk)
bytes_written += len(chunk)
if bytes_written >= int(self.__conn.headers['Content-Length']):
result = Result(Status.OK, outfile)
else:
result = Result(
Status.DISCONNECTED, outfile, "The connection was aborted," +
"check your internet connection")
self.emit('end', result)
return result
def load_gff(gff):
"""Parses a single GFF file and returns a chromosome-indexed dict for
that file.
Arguments
---------
gff: str
Filepath to GFF
Returns
-------
dict: A dictionary representation of the GFF entries, indexed by
chromosome ID
"""
annotations = {}
if gff.endswith('.gz'):
import gzip
from io import TextIOWrapper
fp = TextIOWrapper(gzip.open(gff))
else:
fp = open(gff)
for entry in GFF.parse(fp):
if len(entry.features) > 0 and entry.features[0].type == 'chromosome':
annotations[entry.id] = entry
fp.close()
return annotations
class Context(IteratorContext[Target], metaclass=ABCMeta):
@staticmethod
def _setup(instance: 'TargetReader.Context',
makefile: Makefile) -> None:
instance._makefile = makefile
instance._process = None
def __init__(self) -> None:
super().__init__()
self._makefile = None # type: Makefile
self._process = None # type: Popen
self._stdout = None # type: TextIOBase
def __enter__(self) -> Iterator[Target]:
self._process = self._spawn_subprocess()
self._stdout = TextIOWrapper(self._process.stdout,
encoding='utf-8')
return DefaultParsePipelineFactory.make().build_parse_pipeline(
self._stdout, self._makefile)
def __exit__(self, exc_type, exc_val, exc_tb) -> bool:
if self._process is not None:
self._stdout.close()
self._process.wait()
self._check_returncode()
return False
def _check_returncode(self) -> None:
if self._process.returncode != 0:
raise CalledProcessError(self._process.returncode, ' '.join(
self._process.args)) # type: ignore
@abstractmethod
def _spawn_subprocess(self) -> Popen:
pass
def _generate(self):
""" Generate the Python code. """
needs_close = False
if sys.hexversion >= 0x03000000:
if self._opts.output == '-':
from io import TextIOWrapper
pyfile = TextIOWrapper(sys.stdout.buffer, encoding='utf8')
else:
pyfile = open(self._opts.output, 'wt', encoding='utf8')
needs_close = True
else:
if self._opts.output == '-':
pyfile = sys.stdout
else:
pyfile = open(self._opts.output, 'wt')
needs_close = True
compileUi(self._ui_file, pyfile, self._opts.execute, self._opts.indent,
self._opts.pyqt3_wrapper, self._opts.from_imports,
self._opts.resource_suffix)
if needs_close:
pyfile.close()
def run(input_file: TextIOWrapper, output_file: TextIOWrapper, jobs: int,
max_articles: Optional[int]) -> None:
log.info("Import done, now starting work")
current_articles = []
nb_batches = 0
for i, line in enumerate(input_file):
current_articles.append(line)
if (i + 1) % BATCH_SIZE == 0:
log.info("Starting batch %9d", nb_batches + 1)
result = Parallel(n_jobs=jobs)(delayed(_run)(a)
for a in current_articles)
current_articles = []
append_to(output_file, result, end="\n")
nb_batches += 1
log.info("%9d batches done", nb_batches)
if len(current_articles) > 0:
if nb_batches == 0:
log.info("There is only one batch to process")
result = Parallel(n_jobs=jobs)(delayed(_run)(a)
for a in current_articles)
if len(result) > 0:
append_to(output_file, result, end="")
input_file.close()
output_file.close()
class Snakefile:
def __init__(self, path):
self.path = path
try:
self.file = open(self.path, encoding="utf-8")
except FileNotFoundError as e:
try:
self.file = TextIOWrapper(urllib.request.urlopen(self.path),
encoding="utf-8")
except (HTTPError, URLError, ContentTooShortError, ValueError):
raise WorkflowError("Failed to open {}.".format(path))
self.tokens = tokenize.generate_tokens(self.file.readline)
self.rulecount = 0
self.lines = 0
def __next__(self):
return next(self.tokens)
def __iter__(self):
return self
def __enter__(self):
return self
def __exit__(self, *args):
self.file.close()
def _read_table(cls, tar_handle, filename, requied_columns=()):
requied_columns = frozenset(requied_columns) | frozenset(("ID", ))
handle = TextIOWrapper(tar_handle.extractfile(filename))
result = {}
try:
header = handle.readline().rstrip("\r\n").split("\t")
if len(header) != len(set(header)):
raise ZonkeyDBError(
"Table %r does contains duplicate columns!" % (filename, ))
if requied_columns - set(header):
raise ZonkeyDBError("Required columns are missign in table "
"%r: %s" % (filename, ", ".join()))
for linenum, line in enumerate(handle):
fields = line.rstrip("\r\n").split("\t")
if len(fields) != len(header):
raise ZonkeyDBError(
"Error reading %r at line %i; "
"expected %i columns, found %i "
"columns!" %
(filename, linenum, len(header), len(fields)))
row = dict(zip(header, fields))
if row["ID"] in result:
raise ZonkeyDBError("Duplicate IDs in %r: %s" %
(filename, row["ID"]))
result[row["ID"]] = row
finally:
handle.close()
return result
class bom_open():
"""Context manager to open a file or stdin/stdout. Encoding can be detected
with chardet. Pass additional arguments to `open()`.
Python writes BOM for utf-8-sig, utf-16, or utf-32. BOM is not written
when endianness is specified.
If `file=None` or `'-'`, open stdin (for reading) or stdout (for writing).
If `encoding=None` and `mode='r'` or `'w+'`, file encoding will be detected
using chardet."""
def __init__(self,
file,
mode='r',
buffering=-1,
encoding=None,
*args,
**kwargs):
if file == '-':
self.file = None
else:
self.file = file
self.mode = mode
self.buffering = buffering
self.encoding = encoding
self.args = args
self.kwargs = kwargs
def __enter__(self):
if self.file:
self._f = open(self.file, self.mode, self.buffering, self.encoding,
*self.args, **self.kwargs)
elif self.mode == 'r':
self._f = sys.stdin
elif self.mode == 'w':
if self.encoding:
sys.stdout = open(sys.stdout.fileno(),
'w',
encoding=self.encoding,
buffering=1)
self._f = sys.stdout
else:
raise StdIOError('No file specified, and mode not appropriate '
'for stdin (r) or stdout (w)')
if (self.encoding is None and 'b' not in self.mode
and ('r' in self.mode or '+' in self.mode)):
# run chardet on buffer without advancing file position
peek = self._f.buffer.peek()
detected = chardet.detect(peek)
self.encoding = detected['encoding']
# re-attach file with detected encoding
if self._f.encoding.lower() != self.encoding.lower():
self._f = TextIOWrapper(self._f.detach(),
encoding=self.encoding)
return self._f
def __exit__(self, type, value, traceback):
self._f.close()
def deserialize_workflow_spec(self, s_state, filename=None):
"""
:param s_state: a byte-string with the contents of the packaged
workflow archive, or a file-like object.
:param filename: the name of the package file.
"""
if isinstance(s_state, dict):
return super().deserialize_workflow_spec(s_state)
if isinstance(s_state, str):
return super().deserialize_workflow_spec(s_state)
if isinstance(s_state, bytes):
s_state = BytesIO(s_state)
package_zip = zipfile.ZipFile(s_state,
"r",
compression=zipfile.ZIP_DEFLATED)
config = configparser.ConfigParser()
ini_fp = TextIOWrapper(package_zip.open(Packager.METADATA_FILE),
encoding="UTF-8")
try:
config.read_file(ini_fp)
finally:
ini_fp.close()
parser_class = BpmnParser
parser_class_module = config.get('MetaData',
'parser_class_module',
fallback=None)
if parser_class_module:
mod = __import__(parser_class_module,
fromlist=[config.get('MetaData', 'parser_class')])
parser_class = getattr(mod, config.get('MetaData', 'parser_class'))
parser = parser_class()
for info in package_zip.infolist():
parts = os.path.split(info.filename)
if (len(parts) == 2 and not parts[0]
and parts[1].lower().endswith('.bpmn')):
# It is in the root of the ZIP and is a BPMN file
try:
svg = package_zip.read(info.filename[:-5] + '.svg')
except KeyError:
svg = None
bpmn_fp = package_zip.open(info)
try:
bpmn = etree.parse(bpmn_fp)
finally:
bpmn_fp.close()
parser.add_bpmn_xml(bpmn,
svg=svg,
filename='%s:%s' %
(filename, info.filename))
spec_name = config.get('MetaData', 'entry_point_process')
return parser.get_spec(spec_name)
def test_readline(rf, fn):
f = rf.open(fn)
tr = TextIOWrapper(BufferedReader(f))
while 1:
ln = tr.readline()
if not ln:
break
tr.close()
def test_readline(rf, fn):
f = rf.open(fn)
tr = TextIOWrapper(BufferedReader(f))
while 1:
ln = tr.readline()
if not ln:
break
tr.close()
def close_logs_file(file: TextIOWrapper):
"""Closes the logs file"""
print_prefix(prefix=0)
print("Finishing logs...")
file.write("\nEnd of log entry\n")
file.close()
print_prefix(prefix=0)
print("Done writing logs")
def urlopen(self, url):
if url in self.to_cache:
if url not in self._url_cache:
f = TextIOWrapper(urlopen(url, auto_deflate=True), errors="ignore")
self._url_cache[url] = f.read().splitlines()
f.close()
return self._url_cache[url]
return TextIOWrapper(urlopen(url, auto_deflate=True), errors="ignore")
def get_fasta_from_file(in_memory_file):
file = TextIOWrapper(in_memory_file, 'utf-8')
fasta_seq = SeqIO.parse(file, 'fasta')
records = []
for record in fasta_seq:
temp = record.id, str(record.seq).lower()
records.append(temp)
file.close()
return records
def extract_files(
filenames,
header_definitions,
transform_functions,
output_writers,
error_filename,
input_encoding="latin1",
censorship=True,
):
"""Extract files from a fixed-width file containing more than one row type
`filenames` is expected to be a list of ZIP files having only one file
inside each. The file is read and metadata inside `fobjs` is used to parse
it and save the output files.
"""
error_fobj = open_compressed(error_filename, mode="w", encoding="latin1")
error_writer = CsvLazyDictWriter(error_fobj)
for filename in filenames:
# TODO: use another strategy to open this file (like using rows'
# open_compressed when archive support is implemented)
if os.path.isdir(filename):
continue
if not str(filename).endswith('.zip'):
continue
zf = ZipFile(filename)
inner_filenames = zf.filelist
assert (
len(inner_filenames) == 1
), f"Only one file inside the zip is expected (got {len(inner_filenames)})"
# XXX: The current approach of decoding here and then extracting
# fixed-width-file data will work only for encodings where 1 character is
# represented by 1 byte, such as latin1. If the encoding can represent one
# character using more than 1 byte (like UTF-8), this approach will make
# incorrect results.
fobj = TextIOWrapper(zf.open(inner_filenames[0]), encoding=input_encoding)
for line in tqdm(fobj, desc=f"Extracting {filename}"):
row_type = line[0]
try:
row = parse_row(header_definitions[row_type], line)
except ParsingError as exception:
error_writer.writerow(
{"error": exception.error, "line": exception.line}
)
continue
data = transform_functions[row_type](row)
for row in data:
if censorship: # Clear sensitive information
censor(row_type, row)
output_writers[row_type].writerow(row)
fobj.close()
zf.close()
error_fobj.close()
def parse_post_body(request):
text_io = TextIOWrapper(request.rfile, encoding="UTF-8")
splitted_data = text_io.readline().split("&")
print("Splitted data:" + str(splitted_data))
text_io.close()
key_value = dict()
for data in splitted_data:
temp = data.split("=")
key_value.update({temp[0]: temp[1]})
return key_value
def enqueue_output(out: TextIOWrapper, queue: Queue[Any]) -> bool:
try:
for line in out.readlines():
queue.put_nowait(line)
finally:
if out and not out.closed:
try:
out.close()
return True
except:
return False
def put(self, request, filename, format=None):
file_obj = TextIOWrapper(request.FILES['file'].file,
encoding=request.encoding)
csv_importer = CSVImporter()
import_status = csv_importer.importFromFile(
file_obj).get_import_status()
data = {'Import status': import_status}
response = Response(data, status=204)
response.data = data
file_obj.close()
return Response(data)
def make(self, gnuplot="gnuplot"): """ Execute gnuplot to make the plot output. There is no return from this call; gnuplot should generate its output to a file (or something). """ p = Popen(gnuplot, stdin=PIPE) try: wrapper = TextIOWrapper(p.stdin) self.make_script(wrapper) finally: wrapper.close() p.stdin.close()
class Console():
def __init__(self):
self.console_output = TextIOWrapper(BytesIO(), sys.stdout.encoding)
def __enter__(self):
sys.stdout = self.console_output
return self
def __exit__(self, type, value, traceback):
self.console_output.close()
sys.stdout = sys.__stdout__
def get_output(self):
self.console_output.seek(0)
return self.console_output.read()
def deserialize_workflow_spec(self, s_state, filename=None):
"""
:param s_state: a byte-string with the contents of the packaged workflow archive, or a file-like object.
:param filename: the name of the package file.
"""
if isinstance(s_state, (str, bytes)):
s_state = BytesIO(s_state)
package_zip = zipfile.ZipFile(s_state, "r", compression=zipfile.ZIP_DEFLATED)
config = configparser.SafeConfigParser()
ini_fp = TextIOWrapper(package_zip.open(Packager.METADATA_FILE), encoding="UTF-8")
try:
config.readfp(ini_fp)
finally:
ini_fp.close()
parser_class = BpmnParser
try:
parser_class_module = config.get('MetaData', 'parser_class_module', fallback=None)
except TypeError:
# unfortunately the fallback= does not exist on python 2
parser_class_module = config.get('MetaData', 'parser_class_module', None)
if parser_class_module:
mod = __import__(parser_class_module, fromlist=[config.get('MetaData', 'parser_class')])
parser_class = getattr(mod, config.get('MetaData', 'parser_class'))
parser = parser_class()
for info in package_zip.infolist():
parts = os.path.split(info.filename)
if len(parts) == 2 and not parts[0] and parts[1].lower().endswith('.bpmn'):
#It is in the root of the ZIP and is a BPMN file
try:
svg = package_zip.read(info.filename[:-5]+'.svg')
except KeyError as e:
svg = None
bpmn_fp = package_zip.open(info)
try:
bpmn = ET.parse(bpmn_fp)
finally:
bpmn_fp.close()
parser.add_bpmn_xml(bpmn, svg=svg, filename='%s:%s' % (filename, info.filename))
return parser.get_spec(config.get('MetaData', 'entry_point_process'))
def load(path=None):
datafile = "names.pickle"
if path:
datafile = os.path.join(path, datafile)
if os.path.exists(datafile):
return pickle.load(open(datafile, "rb"))
_temp_file = BytesIO()
_temp_file.write(urllib.request.urlopen(NAMES_URL).read())
_zip_file = zipfile.ZipFile(_temp_file, "r")
names = dict()
for filename in _zip_file.namelist():
if ".txt" not in filename:
continue
_file = _zip_file.open(filename)
_file = TextIOWrapper(_file)
rows = csv.reader(_file, delimiter=",")
for row in rows:
if len(row) < 3:
continue
name = row[0].lower()
gender = row[1]
count = int(row[2])
if name not in names:
names[name] = dict(M=0, F=0)
names[name][gender] = names[name][gender] + count
_file.close()
for key, value in names.items():
count = value["M"] + value["F"]
if value["M"] > value["F"]:
value["probability"] = float(value["M"]) / count
value["gender"] = "M"
else:
value["probability"] = float(value["F"]) / count
value["gender"] = "F"
_datafile = open(datafile, "wb")
pickle.dump(names, _datafile, -1)
_datafile.close()
return names
def load(path=None):
datafile = 'names.pickle'
if path:
datafile = os.path.join(path, datafile)
if os.path.exists(datafile):
return pickle.load(open(datafile, 'rb'))
_temp_file = BytesIO()
_temp_file.write(urllib.request.urlopen(NAMES_URL).read())
_zip_file = zipfile.ZipFile(_temp_file, 'r')
names = dict()
for filename in _zip_file.namelist():
if '.txt' not in filename:
continue
_file = _zip_file.open(filename)
_file = TextIOWrapper(_file)
rows = csv.reader(_file, delimiter=',')
for row in rows:
if len(row) < 3:
continue
name = row[0].lower()
gender = row[1]
count = int(row[2])
if name not in names:
names[name] = dict(M=0, F=0)
names[name][gender] = names[name][gender] + count
_file.close()
for key, value in names.items():
count = value['M'] + value['F']
if value['M'] > value['F']:
value['probability'] = float(value['M']) / count
value['gender'] = 'M'
else:
value['probability'] = float(value['F']) / count
value['gender'] = 'F'
_datafile = open(datafile, 'wb')
pickle.dump(names, _datafile, -1)
_datafile.close()
return names
def read_csvfile_rest(csvfile_handle: TextIOWrapper) -> Iterator[FileRow]:
reader = csv.reader(csvfile_handle)
for row in reader:
if len(row) == 1:
yield CommandRow(command=row[0])
else:
command, hyps, goal, *predictions = row
yield TacticRow(command=command,
hyps=hyps,
goal=goal,
predictions=[
PredictionResult(predictions[i],
predictions[i + 1])
for i in range(0,
len(predictions) - 1, 2)
])
csvfile_handle.close()
def test_io_not_autoclose_textiowrapper(self):
fp = BytesIO(
b"\xc3\xa4\xc3\xb6\xc3\xbc\xc3\x9f\n\xce\xb1\xce\xb2\xce\xb3\xce\xb4"
)
resp = HTTPResponse(fp, preload_content=False, auto_close=False)
reader = TextIOWrapper(resp, encoding="utf8")
assert list(reader) == ["äöüß\n", "αβγδ"]
assert not reader.closed
assert not resp.closed
with pytest.raises(StopIteration):
next(reader)
reader.close()
assert reader.closed
assert resp.closed
with pytest.raises(ValueError, match="I/O operation on closed file.?"):
next(reader)
def build_fa(self, input_config: io.TextIOWrapper) -> FA.FA:
"""
build FA from your input configuration file
:param input_config: input configuration file
:return: FA build from given configuration file
"""
int(input_config.readline())
delta = int(input_config.readline())
input_config.readline()
transition_table = []
for i in range(0, delta):
transition_table.append([int(number) for number in input_config.readline().rstrip('\n').split(' ')])
input_config.readline()
final_states = {int(number) for number in input_config.readline().rstrip('\n').split(' ')}
input_config.close()
del input_config
return FA.FA(tuple(transition_table), final_states)
class SmtLibSolver(Solver):
"""Wrapper for using a solver via textual SMT-LIB interface.
The solver is launched in a subprocess using args as arguments of
the executable. Interaction with the solver occurs via pipe.
"""
OptionsClass = SmtLibOptions
def __init__(self, args, environment, logic, LOGICS=None, **options):
Solver.__init__(self,
environment,
logic=logic,
**options)
self.to = self.environment.typeso
if LOGICS is not None: self.LOGICS = LOGICS
self.args = args
self.declared_vars = set()
self.declared_sorts = set()
self.solver = Popen(args, stdout=PIPE, stderr=PIPE, stdin=PIPE,
bufsize=-1)
# Give time to the process to start-up
time.sleep(0.01)
self.parser = SmtLibParser(interactive=True)
if PY2:
self.solver_stdin = self.solver.stdin
self.solver_stdout = self.solver.stdout
else:
self.solver_stdin = TextIOWrapper(self.solver.stdin)
self.solver_stdout = TextIOWrapper(self.solver.stdout)
# Initialize solver
self.options(self)
self.set_logic(logic)
def set_option(self, name, value):
self._send_silent_command(SmtLibCommand(smtcmd.SET_OPTION,
[name, value]))
def set_logic(self, logic):
self._send_silent_command(SmtLibCommand(smtcmd.SET_LOGIC, [logic]))
def _debug(self, msg, *format_args):
if self.options.debug_interaction:
print(msg % format_args)
def _send_command(self, cmd):
"""Sends a command to the STDIN pipe."""
self._debug("Sending: %s", cmd.serialize_to_string())
cmd.serialize(self.solver_stdin, daggify=True)
self.solver_stdin.write("\n")
self.solver_stdin.flush()
def _send_silent_command(self, cmd):
"""Sends a command to the STDIN pipe and awaits for acknowledgment."""
self._send_command(cmd)
self._check_success()
def _get_answer(self):
"""Reads a line from STDOUT pipe"""
res = self.solver_stdout.readline().strip()
self._debug("Read: %s", res)
return res
def _get_value_answer(self):
"""Reads and parses an assignment from the STDOUT pipe"""
lst = self.parser.get_assignment_list(self.solver_stdout)
self._debug("Read: %s", lst)
return lst
def _declare_sort(self, sort):
cmd = SmtLibCommand(smtcmd.DECLARE_SORT, [sort])
self._send_silent_command(cmd)
self.declared_sorts.add(sort)
def _declare_variable(self, symbol):
cmd = SmtLibCommand(smtcmd.DECLARE_FUN, [symbol])
self._send_silent_command(cmd)
self.declared_vars.add(symbol)
def _check_success(self):
res = self._get_answer()
if res != "success":
raise UnknownSolverAnswerError("Solver returned: '%s'" % res)
def solve(self, assumptions=None):
assert assumptions is None
self._send_command(SmtLibCommand(smtcmd.CHECK_SAT, []))
ans = self._get_answer()
if ans == "sat":
return True
elif ans == "unsat":
return False
elif ans == "unknown":
raise SolverReturnedUnknownResultError
else:
raise UnknownSolverAnswerError("Solver returned: " + ans)
def reset_assertions(self):
self._send_silent_command(SmtLibCommand(smtcmd.RESET_ASSERTIONS, []))
return
def add_assertion(self, formula, named=None):
# This is needed because Z3 (and possibly other solvers) incorrectly
# recognize N * M * x as a non-linear term
formula = formula.simplify()
sorts = self.to.get_types(formula, custom_only=True)
for s in sorts:
if s not in self.declared_sorts:
self._declare_sort(s)
deps = formula.get_free_variables()
for d in deps:
if d not in self.declared_vars:
self._declare_variable(d)
self._send_silent_command(SmtLibCommand(smtcmd.ASSERT, [formula]))
def push(self, levels=1):
self._send_silent_command(SmtLibCommand(smtcmd.PUSH, [levels]))
def pop(self, levels=1):
self._send_silent_command(SmtLibCommand(smtcmd.POP, [levels]))
def get_value(self, item):
self._send_command(SmtLibCommand(smtcmd.GET_VALUE, [item]))
lst = self._get_value_answer()
assert len(lst) == 1
assert len(lst[0]) == 2
return lst[0][1]
def print_model(self, name_filter=None):
if name_filter is not None:
raise NotImplementedError
for v in self.declared_vars:
print("%s = %s" % (v, self.get_value(v)))
def get_model(self):
assignment = {}
for s in self.environment.formula_manager.get_all_symbols():
if s.is_term():
v = self.get_value(s)
assignment[s] = v
return EagerModel(assignment=assignment, environment=self.environment)
def _exit(self):
self._send_command(SmtLibCommand(smtcmd.EXIT, []))
self.solver_stdin.close()
self.solver_stdout.close()
self.solver.stderr.close()
self.solver.terminate()
return
def _get_handle(path_or_buf, mode, encoding=None, compression=None,
memory_map=False, is_text=True):
"""
Get file handle for given path/buffer and mode.
Parameters
----------
path_or_buf :
a path (str) or buffer
mode : str
mode to open path_or_buf with
encoding : str or None
compression : {'infer', 'gzip', 'bz2', 'zip', 'xz', None}, default None
If 'infer' and `filepath_or_buffer` is path-like, then detect
compression from the following extensions: '.gz', '.bz2', '.zip',
or '.xz' (otherwise no compression).
memory_map : boolean, default False
See parsers._parser_params for more information.
is_text : boolean, default True
whether file/buffer is in text format (csv, json, etc.), or in binary
mode (pickle, etc.)
Returns
-------
f : file-like
A file-like object
handles : list of file-like objects
A list of file-like object that were opened in this function.
"""
try:
from s3fs import S3File
need_text_wrapping = (BytesIO, S3File)
except ImportError:
need_text_wrapping = (BytesIO,)
handles = list()
f = path_or_buf
# Convert pathlib.Path/py.path.local or string
path_or_buf = _stringify_path(path_or_buf)
is_path = isinstance(path_or_buf, str)
if is_path:
compression = _infer_compression(path_or_buf, compression)
if compression:
# GZ Compression
if compression == 'gzip':
if is_path:
f = gzip.open(path_or_buf, mode)
else:
f = gzip.GzipFile(fileobj=path_or_buf)
# BZ Compression
elif compression == 'bz2':
if is_path:
f = bz2.BZ2File(path_or_buf, mode)
else:
f = bz2.BZ2File(path_or_buf)
# ZIP Compression
elif compression == 'zip':
zf = BytesZipFile(path_or_buf, mode)
# Ensure the container is closed as well.
handles.append(zf)
if zf.mode == 'w':
f = zf
elif zf.mode == 'r':
zip_names = zf.namelist()
if len(zip_names) == 1:
f = zf.open(zip_names.pop())
elif len(zip_names) == 0:
raise ValueError('Zero files found in ZIP file {}'
.format(path_or_buf))
else:
raise ValueError('Multiple files found in ZIP file.'
' Only one file per ZIP: {}'
.format(zip_names))
# XZ Compression
elif compression == 'xz':
f = lzma.LZMAFile(path_or_buf, mode)
# Unrecognized Compression
else:
msg = 'Unrecognized compression type: {}'.format(compression)
raise ValueError(msg)
handles.append(f)
elif is_path:
if encoding:
# Encoding
f = open(path_or_buf, mode, encoding=encoding, newline="")
elif is_text:
# No explicit encoding
f = open(path_or_buf, mode, errors='replace', newline="")
else:
# Binary mode
f = open(path_or_buf, mode)
handles.append(f)
# Convert BytesIO or file objects passed with an encoding
if is_text and (compression or isinstance(f, need_text_wrapping)):
from io import TextIOWrapper
f = TextIOWrapper(f, encoding=encoding, newline='')
handles.append(f)
if memory_map and hasattr(f, 'fileno'):
try:
g = MMapWrapper(f)
f.close()
f = g
except Exception:
# we catch any errors that may have occurred
# because that is consistent with the lower-level
# functionality of the C engine (pd.read_csv), so
# leave the file handler as is then
pass
return f, handles
class Reader(object):
""" Read SAM/BAM format file as an iterable. """
def __init__(self, f, regions=False, kind=None, samtools_path=None):
ext = None
if samtools_path is None:
samtools_path = "samtools" # Get from the PATH
self.samtools_path = samtools_path
self.spool = None # use this to catch alignment during reader scraping
self.type = 'sam'
try:
self._f_name = f.name
_, ext = os.path.splitext(f.name)
if f.name == '<stdin>': # stdin stream
self._sam_init(f)
elif ext == '.bam' or kind == 'bam':
self._bam_init(f, regions)
self.type = 'bam'
elif ext == '.sam' or kind == 'sam':
self._sam_init(f)
else:
self._sam_init(f)
if (regions and ext != '.bam' and kind is None) or (regions and kind is not None and kind != 'bam'):
self.__exit__()
raise ValueError("Region support requires bam file.")
except AttributeError:
self._f_name = None
if isinstance(f, Connection):
self._pipe_init(f)
else:
self._sam_init(f)
def _pipe_init(self, f):
header = []
for line in iter(f.recv, ''):
if line[0] == '@':
header.append(line.rstrip('\n\r'))
else:
self.spool = line
break
self.header_as_dict(header)
self.f = iter(f.recv, '')
self._conn = 'pipe'
def _sam_init(self, f):
header = []
self.f = f
for line in self.f:
if line[0] == '@':
header.append(line.rstrip('\n\r'))
else:
self.spool = line
break
self.header_as_dict(header)
self._conn = 'file'
def _bam_init(self, f, regions):
pline = [self.samtools_path, 'view', '-H', f.name]
try:
p = Popen(pline, bufsize=-1, stdout=PIPE,
stderr=PIPE)
except OSError:
raise OSError('Samtools must be installed for BAM file support!\n')
self.header_as_dict([line.decode('utf-8').rstrip('\n\r') for line in p.stdout])
p.wait()
if regions:
try:
open(''.join([f.name, '.bai']))
except EnvironmentError:
sys.stderr.write("BAM index not found. Attempting to index file.\n")
index_p = Popen([self.samtools_path, 'index', f.name], stdout=PIPE, stderr=PIPE)
_, err = index_p.communicate()
if index_p.returncode > 0 or re.search("fail", str(err)):
raise OSError("Indexing failed. Is the BAM file sorted?\n")
else:
sys.stderr.write("Index created successfully.\n")
pline = [self.samtools_path, 'view', f.name, regions]
else:
pline = [self.samtools_path, 'view', f.name]
self.p = Popen(pline, bufsize=-1, stdout=PIPE,
stderr=PIPE)
if PY3:
self.f = TextIOWrapper(self.p.stdout)
else:
self.f = self.p.stdout
self._conn = 'proc'
def next(self):
""" Returns the next :class:`.Sam` object """
try:
if self.spool: # this will be the first alignment in a SAM file or stream
line = self.spool.rstrip('\n\r')
self.spool = None
else:
line = next(self.f).rstrip('\n\r')
if line == '':
raise StopIteration
fields = line.split('\t')
required = fields[:11]
tags = fields[11:]
return Sam(*required, tags=tags)
except StopIteration:
raise StopIteration
def __next__(self):
return self.next()
def __iter__(self):
return self
def __len__(self):
""" Returns the number of reads in an indexed BAM file.
Not implemented for SAM files. """
if self.type != 'bam':
raise NotImplementedError("len(Reader) is only implemented for BAM files.")
elif self.type == 'bam':
return sum(bam_read_count(self._f_name, self.samtools_path))
def subsample(self, n):
""" Returns an interator that draws every nth read from
the input file. Returns :class:`.Sam`. """
for i, line in enumerate(self.f):
if i % n == 0:
fields = line.split('\t')
required = fields[:11]
tags = fields[11:]
yield Sam(*required, tags=tags)
def header_as_dict(self, header):
""" Parse the header list and return a nested dictionary. """
self.header = DefaultOrderedDict(OrderedDict)
for line in header:
line = line.split('\t')
key, fields = (line[0], line[1:])
try:
self.header[key][fields[0]] = fields[1:]
except IndexError:
self.header[key][fields[0]] = ['']
@property
def seqs(self):
""" Return just the sequence names from the @SQ library as a generator. """
for key in self.header['@SQ'].keys():
yield key.split(':')[1]
def tile_genome(self, width):
""" Return a generator of UCSC-style regions tiling ``width``. """
assert isinstance(width, int)
for k, v in self.header['@SQ'].items():
rname = k.split(':')[1]
seqlength = v[0].split(':')[1]
for region in tile_region(rname, 1, int(seqlength), width):
yield region
def close(self):
self.__exit__()
def __enter__(self):
return self
def __exit__(self, *args):
if self._conn == 'file':
self.f.close()
if self._conn == 'proc':
self.f.close()
self.p.terminate()
def _get_handle(path, mode, encoding=None, compression=None, memory_map=False):
"""Gets file handle for given path and mode.
"""
if compression is not None:
if encoding is not None and not compat.PY3:
msg = 'encoding + compression not yet supported in Python 2'
raise ValueError(msg)
if compression == 'gzip':
import gzip
f = gzip.GzipFile(path, mode)
elif compression == 'bz2':
import bz2
f = bz2.BZ2File(path, mode)
elif compression == 'zip':
import zipfile
zip_file = zipfile.ZipFile(path)
zip_names = zip_file.namelist()
if len(zip_names) == 1:
file_name = zip_names.pop()
f = zip_file.open(file_name)
elif len(zip_names) == 0:
raise ValueError('Zero files found in ZIP file {}'
.format(path))
else:
raise ValueError('Multiple files found in ZIP file.'
' Only one file per ZIP :{}'
.format(zip_names))
elif compression == 'xz':
lzma = compat.import_lzma()
f = lzma.LZMAFile(path, mode)
else:
raise ValueError('Unrecognized compression type: %s' %
compression)
if compat.PY3:
from io import TextIOWrapper
f = TextIOWrapper(f, encoding=encoding)
return f
else:
if compat.PY3:
if encoding:
f = open(path, mode, encoding=encoding)
else:
f = open(path, mode, errors='replace')
else:
f = open(path, mode)
if memory_map and hasattr(f, 'fileno'):
try:
g = MMapWrapper(f)
f.close()
f = g
except Exception:
# we catch any errors that may have occurred
# because that is consistent with the lower-level
# functionality of the C engine (pd.read_csv), so
# leave the file handler as is then
pass
return f
def _get_handle(source, mode, encoding=None, compression=None, memory_map=False):
"""Gets file handle for given path and mode.
"""
f = source
is_path = isinstance(source, compat.string_types)
# in Python 3, convert BytesIO or fileobjects passed with an encoding
if compat.PY3 and isinstance(source, compat.BytesIO):
from io import TextIOWrapper
return TextIOWrapper(source, encoding=encoding)
elif compression is not None:
compression = compression.lower()
if encoding is not None and not compat.PY3 and not is_path:
msg = 'encoding + compression not yet supported in Python 2'
raise ValueError(msg)
# GZ Compression
if compression == 'gzip':
import gzip
f = gzip.GzipFile(source, mode) \
if is_path else gzip.GzipFile(fileobj=source)
# BZ Compression
elif compression == 'bz2':
import bz2
if is_path:
f = bz2.BZ2File(source, mode)
else:
f = bz2.BZ2File(source) if compat.PY3 else StringIO(
bz2.decompress(source.read()))
# Python 2's bz2 module can't take file objects, so have to
# run through decompress manually
# ZIP Compression
elif compression == 'zip':
import zipfile
zip_file = zipfile.ZipFile(source)
zip_names = zip_file.namelist()
if len(zip_names) == 1:
f = zip_file.open(zip_names.pop())
elif len(zip_names) == 0:
raise ValueError('Zero files found in ZIP file {}'
.format(source))
else:
raise ValueError('Multiple files found in ZIP file.'
' Only one file per ZIP :{}'
.format(zip_names))
# XZ Compression
elif compression == 'xz':
lzma = compat.import_lzma()
f = lzma.LZMAFile(source, mode)
else:
raise ValueError('Unrecognized compression: %s' % compression)
if compat.PY3:
from io import TextIOWrapper
f = TextIOWrapper(f, encoding=encoding)
return f
elif is_path:
if compat.PY3:
if encoding:
f = open(source, mode, encoding=encoding)
else:
f = open(source, mode, errors='replace')
else:
f = open(source, mode)
if memory_map and hasattr(f, 'fileno'):
try:
g = MMapWrapper(f)
f.close()
f = g
except Exception:
# we catch any errors that may have occurred
# because that is consistent with the lower-level
# functionality of the C engine (pd.read_csv), so
# leave the file handler as is then
pass
return f
class SmtLibSolver(Solver):
"""Wrapper for using a solver via textual SMT-LIB interface.
The solver is launched in a subprocess using args as arguments of
the executable. Interaction with the solver occurs via pipe.
"""
def __init__(self, args, environment, logic, user_options=None,
LOGICS=None):
Solver.__init__(self,
environment,
logic=logic,
user_options=user_options)
# Flag used to debug interaction with the solver
self.dbg = False
if LOGICS is not None: self.LOGICS = LOGICS
self.args = args
self.declared_vars = set()
self.solver = Popen(args, stdout=PIPE, stderr=PIPE, stdin=PIPE)
self.parser = SmtLibParser(interactive=True)
if PY2:
self.solver_stdin = self.solver.stdin
self.solver_stdout = self.solver.stdout
else:
self.solver_stdin = TextIOWrapper(self.solver.stdin)
self.solver_stdout = TextIOWrapper(self.solver.stdout)
# Initialize solver
self.set_option(":print-success", "true")
if self.options.generate_models:
self.set_option(":produce-models", "true")
# Redirect diagnostic output to stdout
self.set_option(":diagnostic-output-channel", '"stdout"')
if self.options is not None:
for o,v in iteritems(self.options):
self.set_option(o,v)
self.set_logic(logic)
def set_option(self, name, value):
self._send_silent_command(SmtLibCommand(smtcmd.SET_OPTION,
[name, value]))
def set_logic(self, logic):
self._send_silent_command(SmtLibCommand(smtcmd.SET_LOGIC, [logic]))
def _send_command(self, cmd):
"""Sends a command to the STDIN pipe."""
if self.dbg: print("Sending: " + cmd.serialize_to_string())
cmd.serialize(self.solver_stdin, daggify=True)
self.solver_stdin.write("\n")
self.solver_stdin.flush()
def _send_silent_command(self, cmd):
"""Sends a command to the STDIN pipe and awaits for acknowledgment."""
self._send_command(cmd)
self._check_success()
def _get_answer(self):
"""Reads a line from STDOUT pipe"""
res = self.solver_stdout.readline().strip()
if self.dbg: print("Read: " + str(res))
return res
def _get_value_answer(self):
"""Reads and parses an assignment from the STDOUT pipe"""
lst = self.parser.get_assignment_list(self.solver_stdout)
if self.dbg: print("Read: " + str(lst))
return lst
def _declare_variable(self, symbol):
cmd = SmtLibCommand(smtcmd.DECLARE_FUN, [symbol])
self._send_silent_command(cmd)
self.declared_vars.add(symbol)
def _check_success(self):
res = self._get_answer()
if res != "success":
raise UnknownSolverAnswerError("Solver returned: '%s'" % res)
def solve(self, assumptions=None):
assert assumptions is None
self._send_command(SmtLibCommand(smtcmd.CHECK_SAT, []))
ans = self._get_answer()
if ans == "sat":
return True
elif ans == "unsat":
return False
elif ans == "unknown":
raise SolverReturnedUnknownResultError
else:
raise UnknownSolverAnswerError("Solver returned: " + ans)
def reset_assertions(self):
self._send_silent_command(SmtLibCommand(smtcmd.RESET_ASSERTIONS, []))
return
def add_assertion(self, formula, named=None):
deps = formula.get_free_variables()
for d in deps:
if d not in self.declared_vars:
self._declare_variable(d)
self._send_silent_command(SmtLibCommand(smtcmd.ASSERT, [formula]))
def push(self, levels=1):
self._send_silent_command(SmtLibCommand(smtcmd.PUSH, [levels]))
def pop(self, levels=1):
self._send_silent_command(SmtLibCommand(smtcmd.POP, [levels]))
def get_value(self, item):
self._send_command(SmtLibCommand(smtcmd.GET_VALUE, [item]))
lst = self._get_value_answer()
assert len(lst) == 1
assert len(lst[0]) == 2
return lst[0][1]
def print_model(self, name_filter=None):
if name_filter is not None:
raise NotImplementedError
for v in self.declared_vars:
print("%s = %s" % (v, self.get_value(v)))
def get_model(self):
assignment = {}
for s in self.environment.formula_manager.get_all_symbols():
if s.is_term():
v = self.get_value(s)
assignment[s] = v
return EagerModel(assignment=assignment, environment=self.environment)
def _exit(self):
self._send_command(SmtLibCommand(smtcmd.EXIT, []))
self.solver_stdin.close()
self.solver_stdout.close()
self.solver.stderr.close()
self.solver.terminate()
return
class MultiPageTextImporter:
def __init__(self, mainControl):
"""
mainControl -- Currently PersonalWikiFrame object
"""
self.mainControl = mainControl
def getImportTypes(self, guiparent):
"""
Return sequence of tuples with the description of import types provided
by this object. A tuple has the form (<imp. type>,
<human readable description>, <panel for add. options or None>)
If panels for additional options must be created, they should use
guiparent as parent
"""
if guiparent:
res = wx.xrc.XmlResource.Get()
mptPanel = res.LoadPanel(guiparent, "ImportSubMultipageText")
# ctrls = XrcControls(htmlPanel)
# config = self.mainControl.getConfig()
#
# ctrls.cbPicsAsLinks.SetValue(config.getboolean("main",
# "html_export_pics_as_links"))
# ctrls.chTableOfContents.SetSelection(config.getint("main",
# "export_table_of_contents"))
# ctrls.tfHtmlTocTitle.SetValue(config.get("main",
# "html_toc_title"))
else:
mptPanel = None
return (
("multipage_text", _("Multipage text"), mptPanel),
)
def getImportSourceWildcards(self, importType):
"""
If an export type is intended to go to a file, this function
returns a (possibly empty) sequence of tuples
(wildcard description, wildcard filepattern).
If an export type goes to a directory, None is returned
"""
if importType == "multipage_text":
return ((_("Multipage files (*.mpt)"), "*.mpt"),
(_("Text file (*.txt)"), "*.txt"))
return None
def getAddOptVersion(self):
"""
Returns the version of the additional options information returned
by getAddOpt(). If the return value is -1, the version info can't
be stored between application sessions.
Otherwise, the addopt information can be stored between sessions
and can later handled back to the doImport method of the object
without previously showing the import dialog.
"""
return 0
def getAddOpt(self, addoptpanel):
"""
Reads additional options from panel addoptpanel.
If getAddOptVersion() > -1, the return value must be a sequence
of simple string, unicode and/or numeric objects. Otherwise, any object
can be returned (normally the addoptpanel itself)
"""
if addoptpanel is None:
return (0,)
else:
ctrls = XrcControls(addoptpanel)
showImportTableAlways = boolToInt(ctrls.cbShowImportTableAlways.GetValue())
return (showImportTableAlways,)
def _collectContent(self):
"""
Collect lines from current position of importFile up to separator
or file end collect all lines and return them as list of lines.
"""
content = []
while True:
# Read lines of wikiword
line = self.importFile.readline()
if line == "":
# The last page in mpt file without separator
# ends as the real wiki page
# content = u"".join(content)
break
if line == self.separator:
if len(content) > 0:
# Iff last line of mpt page is empty, the original
# page ended with a newline, so remove last
# character (=newline)
content[-1] = content[-1][:-1]
# content = u"".join(content)
break
content.append(line)
return "".join(content)
def _skipContent(self):
"""
Skip content until reaching next separator or end of file
"""
while True:
# Read lines of wikiword
line = self.importFile.readline()
if line == "":
# The last page in mpt file without separator
# ends as the real wiki page
break
if line == self.separator:
break
def doImport(self, wikiDocument, importType, importSrc,
compatFilenames, addOpt, importData=None):
"""
Run import operation.
wikiDocument -- WikiDocument object
importType -- string tag to identify how to import
importSrc -- Path to source directory or file to import from
compatFilenames -- Should the filenames be decoded from the lowest
level compatible?
addOpt -- additional options returned by getAddOpt()
importData -- if not None contains data to import as bytestring.
importSrc is ignored in this case. Needed for trashcan.
returns True if import was done (needed for trashcan)
"""
if importData is not None:
self.rawImportFile = BytesIO(importData) # TODO bytes or string???
else:
try:
self.rawImportFile = open(pathEnc(importSrc), "rb")
except IOError:
raise ImportException(_("Opening import file failed"))
self.wikiDocument = wikiDocument
self.tempDb = None
showImportTableAlways = addOpt[0]
# wikiData = self.wikiDocument.getWikiData()
# TODO Do not stop on each import error, instead create error list and
# continue
try:
try:
# Wrap input file to convert format
bom = self.rawImportFile.read(len(BOM_UTF8))
if bom != BOM_UTF8:
self.rawImportFile.seek(0)
self.importFile = TextIOWrapper(self.rawImportFile,
MBCS_ENCODING, "replace")
else:
self.importFile = TextIOWrapper(self.rawImportFile,
"utf-8", "replace")
line = self.importFile.readline()
if line.startswith("#!"):
# Skip initial line with #! to allow execution as shell script
line = self.importFile.readline()
if not line.startswith("Multipage text format "):
raise ImportException(
_("Bad file format, header not detected"))
# Following in the format identifier line is a version number
# of the file format
self.formatVer = int(line[22:-1])
if self.formatVer > 1:
raise ImportException(
_("File format number %i is not supported") %
self.formatVer)
# Next is the separator line
line = self.importFile.readline()
if not line.startswith("Separator: "):
raise ImportException(
_("Bad file format, header not detected"))
self.separator = line[11:]
startPos = self.importFile.tell()
if self.formatVer == 0:
self._doImportVer0()
elif self.formatVer == 1:
# Create temporary database. It is mainly filled during
# pass 1 to check for validity and other things before
# actual importing in pass 2
# TODO Respect settings for general temp location!!!
self.tempDb = ConnectWrapSyncCommit(sqlite3.connect(""))
try: # TODO: Remove column "collisionWithPresent", seems to be unused
self.tempDb.execSql("create table entries("
"unifName text primary key not null, " # Unified name in import file
"seen integer not null default 0, " # data really exists
"dontImport integer not null default 0, " # don't import this (set between pass 1 and 2)
"missingDep integer not null default 0, " # missing dependency(ies)
"importVersionData integer not null default 0, " # versioning data present
# "neededBy text default '',"
# "versionContentDifferencing text default '',"
"collisionWithPresent text not null default ''," # Unif. name of present entry which collides with imported one (if any)
"renameImportTo text not null default ''," # Rename imported element to (if at all)
"renamePresentTo text not null default ''" # Rename present element in database to (if at all)
");"
)
# Dependencies. If unifName isn't imported (or faulty), neededBy shouldn't be either
self.tempDb.execSql("create table depgraph("
"unifName text not null default '',"
"neededBy text not null default '',"
"constraint depgraphpk primary key (unifName, neededBy)"
");"
)
# Recursive processing is not supported for this table
self.tempDb.execSql("create table renamegraph("
"unifName text not null default '',"
"dependent text not null default '',"
"constraint renamegraphpk primary key (unifName, dependent),"
"constraint renamegraphsingledep unique (dependent)"
");"
)
# Collect some initial information into the temporary database
self._doImportVer1Pass1()
# Draw some logical conclusions on the temp db
self._markMissingDependencies()
self._markHasVersionData()
self._markCollision()
# Now ask user if necessary
if showImportTableAlways or self._isUserNeeded():
if not self._doUserDecision():
# Canceled by user
return False
# Further logical processing after possible user editing
self._markNonImportedVersionsData()
self._markNonImportedDependencies()
self._propagateRenames()
# TODO: Remove version data without ver. overview or main data
# Back to start of import file and import according to settings
# in temp db
self.importFile.seek(startPos)
self._doImportVer1Pass2()
return True
finally:
self.tempDb.close()
self.tempDb = None
except ImportException:
raise
except Exception as e:
traceback.print_exc()
raise ImportException(str(e))
finally:
self.importFile.close()
def _markMissingDependencies(self):
"""
If a datablock wasn't present, all dependent data blocks are marked as
not to import
"""
while True:
self.tempDb.execSql("""
update entries set missingDep=1, dontImport=1 where (not missingDep) and
unifName in (select depgraph.neededBy from depgraph inner join
entries on depgraph.unifName = entries.unifName where
(not entries.seen) or entries.missingDep);
""")
if self.tempDb.rowcount == 0:
break
def _markHasVersionData(self):
"""
Mark if version data present
"""
self.tempDb.execSql("""
update entries set importVersionData=1 where (not importVersionData) and
unifName in (select substr(unifName, 21) from entries where
unifName glob 'versioning/overview/*' and not dontImport)
""") # TODO Take missing deps into account here?
# self.tempDb.execSql("insert or replace into entries(unifName, importVersionData) "
# "values (?, 1)", (depunifName,))
def _markCollision(self):
"""
Mark collisions between existing and data blocks and such to import
"""
# First find collisions with wiki words
for wikipageUnifName in self.tempDb.execSqlQuerySingleColumn(
"select unifName from entries where unifName glob 'wikipage/*' "
"and not dontImport"):
wpName = wikipageUnifName[9:]
if not self.wikiDocument.isDefinedWikiPageName(wpName):
continue
self.tempDb.execSql("update entries set collisionWithPresent = ? "
"where unifName = ?",
(wikipageUnifName, wikipageUnifName))
# (u"wikipage/" + collisionWithPresent, wikipageUnifName))
# Then find other collisions (saved searches etc.)
for unifName in self.tempDb.execSqlQuerySingleColumn(
"select unifName from entries where (unifName glob 'savedsearch/*' "
"or unifName glob 'savedpagesearch/*') and not dontImport"):
if self.wikiDocument.hasDataBlock(unifName):
self.tempDb.execSql("update entries set collisionWithPresent = ? "
"where unifName = ?", (unifName, unifName))
def _markNonImportedVersionsData(self):
"""
After user dialog: If importVersionData is false for some entries
the depending version data shouldn't be imported.
Only the versioning overview is marked for not importing. The next step
propagates this to the other data blocks
"""
self.tempDb.execSql("""
update entries set dontImport = 1 where
unifName in (select 'versioning/overview/' || unifName from
entries where not importVersionData)
""")
# # Vice versa the importVersionData column must be updated if
# self.tempDb.execSql("""
# update entries set importVersionData = 0 where importVersionData
# and ('versioning/overview/' || unifName) in (select unifName
# from entries where dontImport)
# """)
def _markNonImportedDependencies(self):
"""
After user dialog: If some data blocks where chosen not to import
mark all dependent blocks to not import also (especially version data)
"""
while True:
self.tempDb.execSql("""
update entries set dontImport=1 where (not dontImport) and
unifName in (select depgraph.neededBy from depgraph inner join
entries on depgraph.unifName = entries.unifName where
entries.dontImport);
""")
if self.tempDb.rowcount == 0:
break
def _propagateRenames(self):
"""
Write rename commands for imported items to all parts to import
if some parts need renaming. Renaming of present items is not propagated.
"""
for unifName, renImportTo in self.tempDb.execSqlQuery(
"select unifName, renameImportTo from entries "
"where renameImportTo != '' and not dontImport"):
for depUnifName in self.tempDb.execSqlQuerySingleColumn(
"select dependent from renamegraph where unifName = ? and "
"dependent in (select unifName from entries where "
"not dontImport)", (unifName,)):
if depUnifName.endswith(unifName):
newName = depUnifName[:-len(unifName)] + renImportTo
self.tempDb.execSql("""
update entries set renameImportTo=? where unifName = ?
""", (newName, depUnifName))
def _doUserDecision(self):
"""
Called to present GUI to user for deciding what to do.
This method is overwritten for trashcan GUI.
Returns False if user canceled operation
"""
return MultiPageTextImporterDialog.runModal(
self.mainControl, self.tempDb,
self.mainControl)
def _isUserNeeded(self):
"""
Decide if a dialog must be shown to ask user how to proceed.
Under some circumstances the dialog may be shown regardless of the result.
"""
if self.tempDb.execSqlQuerySingleItem("select missingDep from entries "
"where missingDep limit 1", default=False):
# Missing dependency
return True
if len(self.tempDb.execSqlQuerySingleItem("select collisionWithPresent "
"from entries where collisionWithPresent != '' limit 1",
default="")) > 0:
# Name collision
return True
# No problems found
return False
def _doImportVer0(self):
"""
Import wikiwords if format version is 0.
"""
langHelper = wx.GetApp().createWikiLanguageHelper(
self.wikiDocument.getWikiDefaultWikiLanguage())
while True:
# Read next wikiword
line = self.importFile.readline()
if line == "":
break
wikiWord = line[:-1]
errMsg = langHelper.checkForInvalidWikiWord(wikiWord,
self.wikiDocument)
if errMsg:
raise ImportException(_("Bad wiki word: %s, %s") %
(wikiWord, errMsg))
content = self._collectContent()
page = self.wikiDocument.getWikiPageNoError(wikiWord)
page.replaceLiveText(content)
def _doImportVer1Pass1(self):
while True:
tag = self.importFile.readline()
if tag == "":
# End of file
break
tag = tag[:-1]
if tag.startswith("funcpage/"):
self._skipContent()
elif tag.startswith("savedsearch/"):
self._skipContent()
elif tag.startswith("savedpagesearch/"):
self._skipContent()
elif tag.startswith("wikipage/"):
self._skipContent()
elif tag.startswith("versioning/overview/"):
self._doImportItemVersioningOverviewVer1Pass1(tag[20:])
elif tag.startswith("versioning/packet/versionNo/"):
self._skipContent()
else:
# Unknown tag -> Ignore until separator
self._skipContent()
continue
self.tempDb.execSql("insert or replace into entries(unifName, seen) "
"values (?, 1)", (tag,))
def _readHintedDatablockVer1(self):
"""
Reads datablock and preprocesses encoding if necessary.
Returns either (hintStrings, content) or (None, None) if either
an unknown important hint was found or if encoding had an error.
hintStrings is a list of hints (as unistrings) which were
not processed by the function (therefore encoding hint is removed).
content can be a bytestring or a unistring.
If (None, None) is returned, the remaining content of the entry
was skipped already by the function.
"""
hintLine = self.importFile.readline()[:-1]
hintStrings = hintLine.split(" ")
resultHintStrings = []
# Set default
useB64 = False
# Process hints
for hint in hintStrings:
if hint.startswith("important/encoding/"):
if hint[19:] == "text":
useB64 = False
elif hint[19:] == "base64":
useB64 = True
else:
# Unknown encoding: don't read further
self._skipContent()
return None, None
elif hint.startswith("important/"):
# There is something important we do not understand
self._skipContent()
return None, None
else:
resultHintStrings.append(hint)
content = self._collectContent()
if useB64:
try:
content = base64BlockDecode(content)
except TypeError:
# base64 decoding failed
self._skipContent()
return None, None
return (resultHintStrings, content)
def _doImportItemVersioningOverviewVer1Pass1(self, subtag):
hintStrings, content = self._readHintedDatablockVer1()
if content is None:
return
# Always encode to UTF-8 no matter what the import file encoding is
content = content.encode("utf-8")
try:
ovw = Versioning.VersionOverview(self.wikiDocument,
unifiedBasePageName=subtag)
ovw.readOverviewFromBytes(content)
ovwUnifName = ovw.getUnifiedName()
self.tempDb.execSql("insert or replace into depgraph(unifName, neededBy) "
"values (?, ?)", (subtag, ovwUnifName))
self.tempDb.execSql("insert or replace into renamegraph(unifName, dependent) "
"values (?, ?)", (subtag, ovwUnifName))
for depUnifName in ovw.getDependentDataBlocks(omitSelf=True):
# Mutual dependency between version overview and each version packet
self.tempDb.execSql("insert or replace into depgraph(unifName, neededBy) "
"values (?, ?)", (depUnifName, ovwUnifName))
self.tempDb.execSql("insert or replace into depgraph(unifName, neededBy) "
"values (?, ?)", (ovwUnifName, depUnifName))
self.tempDb.execSql("insert or replace into renamegraph(unifName, dependent) "
"values (?, ?)", (subtag, depUnifName))
# self.tempDb.execSql("insert or replace into entries(unifName, needed) "
# "values (?, 1)", (depUnifName,))
except VersioningException:
return
def _doImportVer1Pass2(self):
wikiDoc = self.wikiDocument
# We have to rename present items
# First wikipages because this automatically renames depending version data
for pageFrom, pageTo in self.tempDb.execSqlQuery(
"""
select substr(unifName, 10), substr(renamePresentTo, 10)
from entries where unifName glob 'wikipage/*' and
renamePresentTo glob 'wikipage/*'
"""):
if wikiDoc.isDefinedWikiPageName(pageFrom):
wikiDoc.renameWikiWords({pageFrom: pageTo}, Consts.ModifyText.off)
# TODO How to handle rename of home page?
# Then remaining data blocks
for oldUnifName, newUnifName in self.tempDb.execSqlQuery(
"""
select unifName, renamePresentTo
from entries where unifName not glob 'wikipage/*' and
renamePresentTo != ''
"""):
wikiDoc.renameDataBlock(oldUnifName, newUnifName)
# For wiki pages with versions to import, existing versions must be
# deleted
for wikiWord in self.tempDb.execSqlQuerySingleColumn(
"""
select substr(unifName, 10)
from entries where unifName glob 'wikipage/*' and
renameImportTo == '' and not dontImport and importVersionData
union
select substr(renameImportTo, 10)
from entries where unifName glob 'wikipage/*' and
renameImportTo glob 'wikipage/*' and not dontImport and
importVersionData
"""):
if not wikiDoc.isDefinedWikiPageName(wikiWord):
continue
page = wikiDoc.getWikiPage(wikiWord)
versionOverview = page.getExistingVersionOverview()
if versionOverview is not None:
versionOverview.delete()
while True:
tag = self.importFile.readline()
if tag == "":
# End of file
break
tag = tag[:-1] # Remove line end
try:
dontImport, renameImportTo = \
self.tempDb.execSqlQuery(
"select dontImport, renameImportTo from "
"entries where unifName = ?", (tag,))[0]
except IndexError:
# Maybe dangerous
traceback.print_exc()
self._skipContent()
continue
if dontImport:
self._skipContent()
continue
if renameImportTo == "":
renameImportTo = tag
if tag.startswith("wikipage/"):
self._importItemWikiPageVer1Pass2(renameImportTo[9:])
elif tag.startswith("funcpage/"):
self._importItemFuncPageVer1Pass2(tag[9:])
elif tag.startswith("savedsearch/"):
self._importB64DatablockVer1Pass2(renameImportTo)
elif tag.startswith("savedpagesearch/"):
self._importHintedDatablockVer1Pass2(renameImportTo)
elif tag.startswith("versioning/"):
self._importHintedDatablockVer1Pass2(renameImportTo)
else:
# Unknown tag -> Ignore until separator
self._skipContent()
for wikiWord in self.tempDb.execSqlQuerySingleColumn(
"""
select substr(unifName, 10)
from entries where unifName glob 'wikipage/*' and
renamePresentTo == '' and importVersionData
union
select substr(renamePresentTo, 10)
from entries where unifName glob 'wikipage/*' and
renamePresentTo glob 'wikipage/*' and importVersionData
"""):
if not wikiDoc.isDefinedWikiPageName(wikiWord):
continue
page = wikiDoc.getWikiPage(wikiWord)
versionOverview = page.getExistingVersionOverview()
if versionOverview is not None:
versionOverview.readOverview()
def _importItemWikiPageVer1Pass2(self, wikiWord):
timeStampLine = self.importFile.readline()[:-1]
timeStrings = timeStampLine.split(" ")
if len(timeStrings) < 3:
traceback.print_exc()
self._skipContent()
return # TODO Report error
timeStrings = timeStrings[:3]
try:
timeStrings = [str(ts) for ts in timeStrings]
except UnicodeEncodeError:
traceback.print_exc()
self._skipContent()
return # TODO Report error
try:
timeStamps = [timegm(time.strptime(ts, "%Y-%m-%d/%H:%M:%S"))
for ts in timeStrings]
except (ValueError, OverflowError):
traceback.print_exc()
self._skipContent()
return # TODO Report error
content = self._collectContent()
page = self.wikiDocument.getWikiPageNoError(wikiWord)
# TODO How to handle versions here?
page.replaceLiveText(content)
if page.getTxtEditor() is not None:
page.writeToDatabase()
page.setTimestamps(timeStamps)
def _importItemFuncPageVer1Pass2(self, subtag):
# The subtag is functional page tag
try:
# subtag is unicode but func tags are bytestrings
subtag = str(subtag)
except UnicodeEncodeError:
self._skipContent()
return
content = self._collectContent()
try:
page = self.wikiDocument.getFuncPage(subtag)
page.replaceLiveText(content)
except BadFuncPageTagException:
# This function tag is bad or unknown -> ignore
return # TODO Report error
def _importB64DatablockVer1Pass2(self, unifName):
# Content is base64 encoded
b64Content = self._collectContent()
try:
datablock = base64BlockDecode(b64Content)
self.wikiDocument.getWikiData().storeDataBlock(unifName, datablock,
storeHint=Consts.DATABLOCK_STOREHINT_INTERN)
except TypeError:
# base64 decoding failed
return # TODO Report error
def _importTextDatablockVer1Pass2(self, unifName):
content = self._collectContent()
try:
self.wikiDocument.getWikiData().storeDataBlock(unifName, content,
storeHint=Consts.DATABLOCK_STOREHINT_INTERN)
except TypeError:
return # TODO Report error
def _importHintedDatablockVer1Pass2(self, unifName):
"""
A hinted datablock starts with an extra line defining encoding
(text or B64) and storage hint. It was introduced later therefore
only versioning packets use this while saved searches don't.
"""
hintStrings, content = self._readHintedDatablockVer1()
if hintStrings is None:
return
# Set defaults
storeHint = Consts.DATABLOCK_STOREHINT_INTERN
# Process hints
for hint in hintStrings:
if hint.startswith("storeHint/"):
if hint[10:] == "extern":
storeHint = Consts.DATABLOCK_STOREHINT_EXTERN
elif hint[10:] == "intern":
storeHint = Consts.DATABLOCK_STOREHINT_INTERN
# No else. It is not vital to get the right storage hint
try:
if isinstance(content, str):
content = BOM_UTF8 + content.encode("utf-8")
self.wikiDocument.getWikiData().storeDataBlock(unifName, content,
storeHint=storeHint)
except TypeError:
traceback.print_exc()
return # TODO Report error
stimuliParser.add_argument('--ch4', type=PortPin, required=False, help='I/O Register bit to map CH4 to.')
stimuliParser.add_argument('-s', '--threshold', type=int, default=128, required=False, help='Logic level treshold value')
stimuliParser.add_argument('--clkFreq', type=int, required=True, help="Simulated AVR uC clock frequency in Hz")
actionMap = {
"info": info,
"csv": csv,
"plot": plot,
"json": json,
"vcd": vcd,
"stimuli": stimuli
}
args = parser.parse_args()
try:
with args.infile as f:
scopeData = wfm.parseRigolWFM(f, args.forgiving)
if isinstance(args.outfile, TextIOWrapper):
outputFile = args.outfile
else:
outputFile = TextIOWrapper(args.outfile, encoding="ascii")
actionMap[args.action](args, scopeData, outputFile)
outputFile.close()
except wfm.FormatError as e:
print("Format does not follow the known file format. Try the --forgiving option.", file=sys.stderr)
print("If you'd like to help development, please report this error:\n", file=sys.stderr)
print(e, file=sys.stderr)
sys.exit()
class FileObjectPosix(object):
"""
A file-like object that operates on non-blocking files.
.. seealso:: :func:`gevent.os.make_nonblocking`
"""
default_bufsize = io.DEFAULT_BUFFER_SIZE
def __init__(self, fobj, mode='rb', bufsize=-1, close=True):
"""
:param fobj: Either an integer fileno, or an object supporting the
usual :meth:`socket.fileno` method. The file will be
put in non-blocking mode.
"""
if isinstance(fobj, int):
fileno = fobj
fobj = None
else:
fileno = fobj.fileno()
if not isinstance(fileno, int):
raise TypeError('fileno must be int: %r' % fileno)
orig_mode = mode
mode = (mode or 'rb').replace('b', '')
if 'U' in mode:
self._translate = True
mode = mode.replace('U', '')
else:
self._translate = False
if len(mode) != 1:
# Python 3 builtin `open` raises a ValueError for invalid modes;
# Python 2 ignores in. In the past, we raised an AssertionError, if __debug__ was
# enabled (which it usually was). Match Python 3 because it makes more sense
# and because __debug__ may not be enabled
raise ValueError('mode can only be [rb, rU, wb], not %r' % (orig_mode,))
self._fobj = fobj
self._closed = False
self._close = close
self.fileio = GreenFileDescriptorIO(fileno, mode, closefd=close)
if bufsize < 0:
bufsize = self.default_bufsize
if mode == 'r':
if bufsize == 0:
bufsize = 1
elif bufsize == 1:
bufsize = self.default_bufsize
self.io = BufferedReader(self.fileio, bufsize)
elif mode == 'w':
if bufsize == 0:
bufsize = 1
elif bufsize == 1:
bufsize = self.default_bufsize
self.io = BufferedWriter(self.fileio, bufsize)
else:
# QQQ: not used
self.io = BufferedRandom(self.fileio, bufsize)
if self._translate:
self.io = TextIOWrapper(self.io)
@property
def closed(self):
"""True if the file is cloed"""
return self._closed
def close(self):
if self._closed:
# make sure close() is only ran once when called concurrently
return
self._closed = True
try:
self.io.close()
self.fileio.close()
finally:
self._fobj = None
def flush(self):
self.io.flush()
def fileno(self):
return self.io.fileno()
def write(self, data):
self.io.write(data)
def writelines(self, lines):
self.io.writelines(lines)
def read(self, size=-1):
return self.io.read(size)
def readline(self, size=-1):
return self.io.readline(size)
def readlines(self, sizehint=0):
return self.io.readlines(sizehint)
def readable(self):
return self.io.readable()
def writable(self):
return self.io.writable()
def seek(self, *args, **kwargs):
return self.io.seek(*args, **kwargs)
def seekable(self):
return self.io.seekable()
def tell(self):
return self.io.tell()
def truncate(self, size=None):
return self.io.truncate(size)
def __iter__(self):
return self.io
def __getattr__(self, name):
# XXX: Should this really be _fobj, or self.io?
# _fobj can easily be None but io never is
return getattr(self._fobj, name)
class FileObjectPosix(object):
"""
A file-like object that operates on non-blocking files but
provides a synchronous, cooperative interface.
.. note::
Random read/write (e.g., ``mode='rwb'``) is not supported.
For that, use :class:`io.BufferedRWPair` around two instance of this
class.
.. tip::
Although this object provides a :meth:`fileno` method and
so can itself be passed to :func:`fcntl.fcntl`, setting the
:data:`os.O_NONBLOCK` flag will have no effect; likewise, removing
that flag will cause this object to no longer be cooperative.
"""
#: platform specific default for the *bufsize* parameter
default_bufsize = io.DEFAULT_BUFFER_SIZE
def __init__(self, fobj, mode='rb', bufsize=-1, close=True):
"""
:keyword fobj: Either an integer fileno, or an object supporting the
usual :meth:`socket.fileno` method. The file *will* be
put in non-blocking mode using :func:`gevent.os.make_nonblocking`.
:keyword str mode: The manner of access to the file, one of "rb", "rU" or "wb"
(where the "b" or "U" can be omitted).
If "U" is part of the mode, IO will be done on text, otherwise bytes.
:keyword int bufsize: If given, the size of the buffer to use. The default
value means to use a platform-specific default, and a value of 0 is translated
to a value of 1. Other values are interpreted as for the :mod:`io` package.
Buffering is ignored in text mode.
"""
if isinstance(fobj, int):
fileno = fobj
fobj = None
else:
fileno = fobj.fileno()
if not isinstance(fileno, int):
raise TypeError('fileno must be int: %r' % fileno)
orig_mode = mode
mode = (mode or 'rb').replace('b', '')
if 'U' in mode:
self._translate = True
mode = mode.replace('U', '')
else:
self._translate = False
if len(mode) != 1 and mode not in 'rw': # pragma: no cover
# Python 3 builtin `open` raises a ValueError for invalid modes;
# Python 2 ignores it. In the past, we raised an AssertionError, if __debug__ was
# enabled (which it usually was). Match Python 3 because it makes more sense
# and because __debug__ may not be enabled.
# NOTE: This is preventing a mode like 'rwb' for binary random access;
# that code was never tested and was explicitly marked as "not used"
raise ValueError('mode can only be [rb, rU, wb], not %r' % (orig_mode,))
self._fobj = fobj
self._closed = False
self._close = close
self.fileio = GreenFileDescriptorIO(fileno, mode, closefd=close)
if bufsize < 0 or bufsize == 1:
bufsize = self.default_bufsize
elif bufsize == 0:
bufsize = 1
if mode == 'r':
self.io = BufferedReader(self.fileio, bufsize)
else:
assert mode == 'w'
self.io = BufferedWriter(self.fileio, bufsize)
#else: # QQQ: not used, not reachable
#
# self.io = BufferedRandom(self.fileio, bufsize)
if self._translate:
self.io = TextIOWrapper(self.io)
@property
def closed(self):
"""True if the file is cloed"""
return self._closed
def close(self):
if self._closed:
# make sure close() is only run once when called concurrently
return
self._closed = True
try:
self.io.close()
self.fileio.close()
finally:
self._fobj = None
def flush(self):
self.io.flush()
def fileno(self):
return self.io.fileno()
def write(self, data):
self.io.write(data)
def writelines(self, lines):
self.io.writelines(lines)
def read(self, size=-1):
return self.io.read(size)
def readline(self, size=-1):
return self.io.readline(size)
def readlines(self, sizehint=0):
return self.io.readlines(sizehint)
def readable(self):
return self.io.readable()
def writable(self):
return self.io.writable()
def seek(self, *args, **kwargs):
return self.io.seek(*args, **kwargs)
def seekable(self):
return self.io.seekable()
def tell(self):
return self.io.tell()
def truncate(self, size=None):
return self.io.truncate(size)
def __iter__(self):
return self.io
def __getattr__(self, name):
# XXX: Should this really be _fobj, or self.io?
# _fobj can easily be None but io never is
return getattr(self._fobj, name)
class FileObjectPosix:
default_bufsize = io.DEFAULT_BUFFER_SIZE
def __init__(self, fobj, mode='rb', bufsize=-1, close=True):
if isinstance(fobj, int):
fileno = fobj
fobj = None
else:
fileno = fobj.fileno()
if not isinstance(fileno, int):
raise TypeError('fileno must be int: %r' % fileno)
mode = (mode or 'rb').replace('b', '')
if 'U' in mode:
self._translate = True
mode = mode.replace('U', '')
else:
self._translate = False
assert len(mode) == 1, 'mode can only be [rb, rU, wb]'
self._fobj = fobj
self._closed = False
self._close = close
self.fileio = GreenFileDescriptorIO(fileno, mode, closefd=close)
if bufsize < 0:
bufsize = self.default_bufsize
if mode == 'r':
if bufsize == 0:
bufsize = 1
elif bufsize == 1:
bufsize = self.default_bufsize
self.io = BufferedReader(self.fileio, bufsize)
elif mode == 'w':
if bufsize == 0:
bufsize = 1
elif bufsize == 1:
bufsize = self.default_bufsize
self.io = BufferedWriter(self.fileio, bufsize)
else:
# QQQ: not used
self.io = BufferedRandom(self.fileio, bufsize)
if self._translate:
self.io = TextIOWrapper(self.io)
@property
def closed(self):
"""True if the file is cloed"""
return self._closed
def close(self):
if self._closed:
# make sure close() is only ran once when called concurrently
return
self._closed = True
try:
self.io.close()
self.fileio.close()
finally:
self._fobj = None
def flush(self):
self.io.flush()
def fileno(self):
return self.io.fileno()
def write(self, data):
self.io.write(data)
def writelines(self, list):
self.io.writelines(list)
def read(self, size=-1):
return self.io.read(size)
def readline(self, size=-1):
return self.io.readline(size)
def readlines(self, sizehint=0):
return self.io.readlines(sizehint)
def __iter__(self):
return self.io
class FileObjectPosix(object):
"""
A file-like object that operates on non-blocking files.
.. seealso:: :func:`gevent.os.make_nonblocking`
"""
default_bufsize = io.DEFAULT_BUFFER_SIZE
def __init__(self, fobj, mode='rb', bufsize=-1, close=True):
"""
:param fobj: Either an integer fileno, or an object supporting the
usual :meth:`socket.fileno` method. The file will be
put in non-blocking mode.
"""
if isinstance(fobj, int):
fileno = fobj
fobj = None
else:
fileno = fobj.fileno()
if not isinstance(fileno, int):
raise TypeError('fileno must be int: %r' % fileno)
mode = (mode or 'rb').replace('b', '')
if 'U' in mode:
self._translate = True
mode = mode.replace('U', '')
else:
self._translate = False
assert len(mode) == 1, 'mode can only be [rb, rU, wb]'
self._fobj = fobj
self._closed = False
self._close = close
self.fileio = GreenFileDescriptorIO(fileno, mode, closefd=close)
if bufsize < 0:
bufsize = self.default_bufsize
if mode == 'r':
if bufsize == 0:
bufsize = 1
elif bufsize == 1:
bufsize = self.default_bufsize
self.io = BufferedReader(self.fileio, bufsize)
elif mode == 'w':
if bufsize == 0:
bufsize = 1
elif bufsize == 1:
bufsize = self.default_bufsize
self.io = BufferedWriter(self.fileio, bufsize)
else:
# QQQ: not used
self.io = BufferedRandom(self.fileio, bufsize)
if self._translate:
self.io = TextIOWrapper(self.io)
@property
def closed(self):
"""True if the file is cloed"""
return self._closed
def close(self):
if self._closed:
# make sure close() is only ran once when called concurrently
return
self._closed = True
try:
self.io.close()
self.fileio.close()
finally:
self._fobj = None
def flush(self):
self.io.flush()
def fileno(self):
return self.io.fileno()
def write(self, data):
self.io.write(data)
def writelines(self, lines):
self.io.writelines(lines)
def read(self, size=-1):
return self.io.read(size)
def readline(self, size=-1):
return self.io.readline(size)
def readlines(self, sizehint=0):
return self.io.readlines(sizehint)
def seek(self, *args, **kwargs):
return self.io.seek(*args, **kwargs)
def seekable(self):
return self.io.seekable()
def tell(self):
return self.io.tell()
def truncate(self, size=None):
return self.io.truncate(size)
def __iter__(self):
return self.io
def __getattr__(self, name):
return getattr(self._fobj, name)
class FileObjectPosix(object):
"""
A file-like object that operates on non-blocking files but
provides a synchronous, cooperative interface.
.. caution::
This object is most effective wrapping files that can be used appropriately
with :func:`select.select` such as sockets and pipes.
In general, on most platforms, operations on regular files
(e.g., ``open('/etc/hosts')``) are considered non-blocking
already, even though they can take some time to complete as
data is copied to the kernel and flushed to disk (this time
is relatively bounded compared to sockets or pipes, though).
A :func:`~os.read` or :func:`~os.write` call on such a file
will still effectively block for some small period of time.
Therefore, wrapping this class around a regular file is
unlikely to make IO gevent-friendly: reading or writing large
amounts of data could still block the event loop.
If you'll be working with regular files and doing IO in large
chunks, you may consider using
:class:`~gevent.fileobject.FileObjectThread` or
:func:`~gevent.os.tp_read` and :func:`~gevent.os.tp_write` to bypass this
concern.
.. note::
Random read/write (e.g., ``mode='rwb'``) is not supported.
For that, use :class:`io.BufferedRWPair` around two instance of this
class.
.. tip::
Although this object provides a :meth:`fileno` method and
so can itself be passed to :func:`fcntl.fcntl`, setting the
:data:`os.O_NONBLOCK` flag will have no effect; however, removing
that flag will cause this object to no longer be cooperative.
.. versionchanged:: 1.1
Now uses the :mod:`io` package internally. Under Python 2, previously
used the undocumented class :class:`socket._fileobject`. This provides
better file-like semantics (and portability to Python 3).
"""
#: platform specific default for the *bufsize* parameter
default_bufsize = io.DEFAULT_BUFFER_SIZE
def __init__(self, fobj, mode='rb', bufsize=-1, close=True):
"""
:keyword fobj: Either an integer fileno, or an object supporting the
usual :meth:`socket.fileno` method. The file *will* be
put in non-blocking mode using :func:`gevent.os.make_nonblocking`.
:keyword str mode: The manner of access to the file, one of "rb", "rU" or "wb"
(where the "b" or "U" can be omitted).
If "U" is part of the mode, IO will be done on text, otherwise bytes.
:keyword int bufsize: If given, the size of the buffer to use. The default
value means to use a platform-specific default, and a value of 0 is translated
to a value of 1. Other values are interpreted as for the :mod:`io` package.
Buffering is ignored in text mode.
"""
if isinstance(fobj, int):
fileno = fobj
fobj = None
else:
fileno = fobj.fileno()
if not isinstance(fileno, int):
raise TypeError('fileno must be int: %r' % fileno)
orig_mode = mode
mode = (mode or 'rb').replace('b', '')
if 'U' in mode:
self._translate = True
mode = mode.replace('U', '')
else:
self._translate = False
if len(mode) != 1 and mode not in 'rw': # pragma: no cover
# Python 3 builtin `open` raises a ValueError for invalid modes;
# Python 2 ignores it. In the past, we raised an AssertionError, if __debug__ was
# enabled (which it usually was). Match Python 3 because it makes more sense
# and because __debug__ may not be enabled.
# NOTE: This is preventing a mode like 'rwb' for binary random access;
# that code was never tested and was explicitly marked as "not used"
raise ValueError('mode can only be [rb, rU, wb], not %r' % (orig_mode,))
self._fobj = fobj
self._closed = False
self._close = close
self.fileio = GreenFileDescriptorIO(fileno, mode, closefd=close)
if bufsize < 0 or bufsize == 1:
bufsize = self.default_bufsize
elif bufsize == 0:
bufsize = 1
if mode == 'r':
self.io = BufferedReader(self.fileio, bufsize)
else:
assert mode == 'w'
self.io = BufferedWriter(self.fileio, bufsize)
#else: # QQQ: not used, not reachable
#
# self.io = BufferedRandom(self.fileio, bufsize)
if self._translate:
self.io = TextIOWrapper(self.io)
@property
def closed(self):
"""True if the file is closed"""
return self._closed
def close(self):
if self._closed:
# make sure close() is only run once when called concurrently
return
self._closed = True
try:
self.io.close()
self.fileio.close()
finally:
self._fobj = None
def flush(self):
self.io.flush()
def fileno(self):
return self.io.fileno()
def write(self, data):
self.io.write(data)
def writelines(self, lines):
self.io.writelines(lines)
def read(self, size=-1):
return self.io.read(size)
def readline(self, size=-1):
return self.io.readline(size)
def readlines(self, sizehint=0):
return self.io.readlines(sizehint)
def readable(self):
"""
.. versionadded:: 1.1b2
"""
return self.io.readable()
def writable(self):
"""
.. versionadded:: 1.1b2
"""
return self.io.writable()
def seek(self, *args, **kwargs):
return self.io.seek(*args, **kwargs)
def seekable(self):
return self.io.seekable()
def tell(self):
return self.io.tell()
def truncate(self, size=None):
return self.io.truncate(size)
def __iter__(self):
return self.io
def __getattr__(self, name):
# XXX: Should this really be _fobj, or self.io?
# _fobj can easily be None but io never is
return getattr(self._fobj, name)