def _to_geojson(df, d, **kw):
if _is_geojson_imported is True:
nkw = Struct.inspect_kwargs(kwargs, gpd.to_file)
df.to_file(d, driver='GeoJSON', **nkw)
else:
nkw = Struct.inspect_kwargs(kwargs, Frame.to_json)
res = Frame.to_json(df, **nkw)
with open(d, 'w', encoding=encoding) as f:
json.dump(res, f, ensure_ascii=False)
def _read_geojson(s, **kw):
if _is_geopandas_imported is True:
nkw = Struct.inspect_kwargs(kw, gpd.read_file)
nkw.update({'driver': 'GeoJSON'})
return gpd.read_file(s, **nkw)
else:
nkw = Struct.inspect_kwargs(kw, geojson.load)
# note that geojson.load is a wrapper around the core json.load function
# with the same name, and will pass through any additional arguments
return geojson.load(s, **nkw)
def _read_shapefile(s, **kw):
try:
assert osp.exists(s) is True # Misc.File.file_exists(s)
except:
warnings.warn(
"\n! GeoPandas reads URLs and files on disk only - set flags on_disk=True and ignore_buffer=True when loading sourc"
)
try:
p, f = osp.dirname(s), osp.basename(osp.splitext(s)[0])
except:
pass
try:
assert (osp.exists(osp.join(p,'%s.shx' % f)) or osp.exists(osp.join(p,'%s.SHX' % f))) \
and (osp.exists(osp.join(p,'%s.prj' % f)) or osp.exists(osp.join(p,'%s.PRJ' % f))) \
and (osp.exists(osp.join(p,'%s.dbf' % f)) or osp.exists(osp.join(p,'%s.DBF' % f)))
except AssertionError:
warnings.warn(
"\n! Companion files [.dbf, .shx, .prj] are required together with shapefile source"
" - add companion files to path, e.g. set flags fmt='csv' and 'infer_fmt'=False when loading source"
)
except:
pass
nkw = Struct.inspect_kwargs(kw, gpd.read_file)
nkw.update({'driver': 'shapefile'})
return gpd.read_file(s, **nkw)
def loads(cls, s, **kwargs):
"""
"""
serialize = kwargs.pop('serialize', False)
nkwargs = Struct.inspect_kwargs(kwargs, json.loads)
try:
assert serialize is True
except:
return json.loads(s, **kwargs)
else:
return json.loads(s, object_hook=cls.restore, **nkwargs)
def dumps(cls, data, **kwargs):
"""
"""
serialize = kwargs.pop('serialize', False)
nkwargs = Struct.inspect_kwargs(kwargs, json.dumps)
try:
assert serialize is True
except:
return json.dumps(data, **nkwargs)
else:
return json.dumps(cls.serialize(data), **nkwargs)
def dump(cls, data, f, **kwargs):
serialize = kwargs.pop('serialize', False)
# note: when is_order_preserved is False, this entire class can actually be
# ignored since the dump/load methods are exactly equivalent to the original
# dump/load method of the json package
nkwargs = Struct.inspect_kwargs(kwargs, json.dump)
try:
assert serialize is True
except:
json.dump(data, f, **nkwargs)
else:
json.dump(cls.serialize(data), f, **nkwargs)
def check_format(fmt, infer_fmt=False):
"""
>>> fmt = File.check_format(fmt, infer_fmt=False)
"""
try:
assert fmt is None or isinstance(fmt, string_types) \
or (isinstance(fmt, Sequence) and all([isinstance(f, string_types) for f in fmt]))
except:
raise IOError("Wrong format for FMT parameter: '%s'" % fmt)
if fmt is None: fmt = list(File.FORMATS.values())
elif isinstance(fmt, string_types): fmt = [
fmt,
]
try:
assert isinstance(infer_fmt, (bool, string_types)) \
or (isinstance(infer_fmt, Sequence) and all([isinstance(f, string_types) for f in infer_fmt]))
except:
raise IOError("Wrong format for INFER_FMT flag: '%s'" % infer_fmt)
if infer_fmt is True: # extend... with all besides those parsed
infer_fmt = File.FORMATS.keys() # default
elif isinstance(infer_fmt, string_types):
infer_fmt = [
infer_fmt,
]
if not infer_fmt is False: # extend... with all besides those parsed
fmt.extend(infer_fmt) # test all!
try:
fmt.insert(fmt.index('xlsx'), 'xls') or fmt.remove('xlsx')
except:
pass
fmt = Struct.uniq_items(fmt, items=File.FORMATS)
try:
assert fmt not in (None, [], '')
except:
raise IOError("Data format FMT not recognised: '%s'" % fmt)
if isinstance(fmt, string_types):
fmt = [
fmt,
]
return fmt
def _read_topojson(s, **kw):
nkw = Struct.inspect_kwargs(kw, gpd.read_file)
nkw.update({'driver': 'TopoJSON'})
#with fiona.MemoryFile(s) as f: #with fiona.ZipMemoryFile(s) as f:
# return gpd.GeoDataFrame.from_features(f, crs=f.crs, **nkw)
return gpd.read_file(s, **nkw)
def _read_table(s, **kw):
nkw = Struct.inspect_kwargs(kw, pd.read_table)
return pd.read_table(s, **nkw)
def _read_html(s, **kw):
nkw = Struct.inspect_kwargs(kw, pd.read_html)
return pd.read_html(s, **nkw)
def _read_sas(s, **kw):
nkw = Struct.inspect_kwargs(kw, pd.read_sas)
return pd.read_sas(s, **nkw)
def unzip(file, **kwargs):
"""Unzip file on-disk.
>>> res = File.unzip(file, **kwargs)
"""
# try:
# assert isinstance(file, (io.BytesIO,string_types))
# except:
# raise TypeError("Zip file '%s' not recognised" % file)
try:
assert zipfile.is_zipfile(file)
except:
raise IOError("Zip file '%s' not recognised" % file)
path = kwargs.pop(
'path') if 'path' in kwargs else SysEnv.default_cache()
operators = [op for op in ['open', 'extract', 'extractall', 'getinfo', 'namelist', 'read', 'infolist'] \
if op in kwargs.keys()]
try:
assert operators in ([], [None]) or sum([1
for op in operators]) == 1
except:
raise IOError("Only one operation supported per call")
else:
if operators in ([], [None]):
operator = 'extractall'
else:
operator = operators[0]
if operator in ('infolist', 'namelist'):
try:
assert kwargs.get(operator) not in (False, None)
except:
raise IOError("No operation parsed")
else:
members = kwargs.pop(operator, None)
#if operator.startswith('extract'):
# warn("\n! Data extracted from zip file will be physically stored on local disk !")
if isinstance(members, string_types):
members = [
members,
]
with zipfile.ZipFile(file) as zf:
namelist, infolist = zf.namelist(), zf.infolist()
if operator == 'namelist':
return namelist if len(namelist) > 1 else namelist[0]
elif operator == 'infolist':
return infolist if len(infolist) > 1 else infolist[0]
elif operator == 'extractall':
if members in (None, True): members = namelist
return zf.extractall(path=path, members=members)
if members is None and len(namelist) == 1:
members = namelist
elif members is not None:
for i in reversed(range(len(members))):
m = members[i]
try:
assert m in namelist
except:
try:
_mem = [n for n in namelist if n.endswith(m)]
assert len(_mem) == 1
except:
if len(_mem) > 1:
warn(
"\n! Mulitple files machting in zip source - ambiguity not resolved !"
% m)
else: # len(_mem) == 0 <=> _mem = []
warn(
"\n! File '%s' not found in zip source !" %
m)
members.pop(i)
else:
members[i] = _mem[0]
else:
pass # continue
# now: operator in ('extract', 'getinfo', 'read')
if members in ([], None):
raise IOError(
"Impossible to retrieve member file(s) from zipped data")
nkw = Struct.inspect_kwargs(kwargs, getattr(zf, operator))
if operator == 'extract':
nkw.update({'path': path})
results = {m: getattr(zf, operator)(m, **nkw) for m in members}
return results
def _to_json(df, d, **kw):
nkw = Struct.inspect_kwargs(kwargs, Frame.to_json)
res = cls.to_json(df, **nkw)
with open(d, 'w', encoding=encoding) as f:
json.dump(res, f, ensure_ascii=False)
def _to_excel(df, d, **kw):
nkw = Struct.inspect_kwargs(kwargs, pd.to_excel)
df.to_excel(d, **nkw)
def _to_csv(df, d, **kw):
nkw = Struct.inspect_kwargs(kw, pd.to_csv)
df.to_csv(d, **nkw)
def _read_geopackage(s, **kw):
nkw = Struct.inspect_kwargs(kw, gpd.read_file)
nkw.update({'driver': 'GPKG'})
return gpd.read_file(s, **nkw)
def _read_csv(s, **kw):
nkw = Struct.inspect_kwargs(kw, pd.read_csv)
try:
return pd.read_csv(s, **nkw)
except:
return SysEnv.chardet_decorate(pd.read_csv)(s, **nkw)
def _to_geopackage(df, d, **kw):
nkw = Struct.inspect_kwargs(kwargs, gpd.to_file)
df.to_file(d, driver='GPKG', **nkw)
def _read_json(s, **kw):
nkw = Struct.inspect_kwargs(kw, pd.read_json)
return pd.read_json(s, **nkw)
