def __getitem__(self, key):
if isinstance(key, slice):
return self._values.__getitem__(key)
elif is_int(key):
return self._values.__getitem__(key)
elif key in self._keyhash:
index = self._keyhash[key]
return self._values.__getitem__(index)
else:
raise Exception("Couldn't find value {0} in IndexDict".format(key))
def __setitem__(self, key, value):
if is_int(key):
#'key' is actually an index,
#must be an already existing item
self._values[key] = value
else:
len_vals = len(self._values)
index = self._keyhash.get(key, len_vals)
if index >= len(self._values):
self._keyhash[key] = len(self._values)
self._values.append(value)
else:
self._values[index] = value
def process_dict_list_obj(dict_list_obj, path):
"""json-style nested objects consisting of
a (dictionary OR list) of (dictionary OR list) of (dictionary OR list) etc
"""
if not isinstance(path, list):
t = type(path)
raise Exception("process_dict_list_obj function argument path requires a list, received '{path}', type {t}".format(**vars()))
end_node = follow_path(dict_list_obj, path)
if isinstance(end_node, list):
cols = set()
for n in end_node:
if isinstance(n, dict):
cols = cols.union(six.viewkeys(n))
elif isinstance(n, (list,tuple)):
#write a csv from a list of lists
#by using a dummy header (eg 0,1,2,...)
#@example:
#[["a","b"],["c","d"],["e","f"]] ->
#0,1
#a,b
#c,d
#e,f
cols = cols.union(range(len(n)))
else:
raise Exception("Can't convert simple list to csv.")
# else isinstance(n, list):
# raise Exception("can't convert nested lists to csv -- try using a deeper path")
cols = sorted(list(cols))
yield cols
for n in end_node:
if isinstance(n, dict):
r = [n.get(c,"") for c in cols]
#don't use dumps on a string or it'll create an extra pair of quotes
r = [json.dumps(x,ensure_ascii = False) if isinstance(x,(tuple,list,dict)) else x for x in r]
yield r
elif isinstance(n,(list,tuple)):
r = [n[c] if (jtutils.is_int(c) and c < len(n)) else "" for c in cols]
#don't use dumps on a string or it'll create an extra pair of quotes
r = [json.dumps(x,ensure_ascii = False) if isinstance(x,(tuple,list,dict)) else x for x in r]
yield r
else:
raise Exception("Not a dict, list or tuple!")
elif isinstance(end_node, dict):
cols = list(six.viewkeys(end_node))
yield cols
r = [end_node.get(c,"") for c in cols]
#don't use dumps on a string or it'll create an extra pair of quotes
r = [json.dumps(x,ensure_ascii = False) if isinstance(x,(list,dict)) else x for x in r]
yield r
else:
raise_path_error()
def gen_outhdr(hdr, add_list, keep_list, drop_list):
outhdr = hdr[:]
if keep_list:
if not add_list:
add_list = [x for x in keep_list if x not in hdr and not is_int(x)]
#example tmp_dict = {0:"col1",1:"col2",2:"col3","col1":"col1","col2":"col2","col3":"col3"}
tmp_dict = dict(
list(enumerate(outhdr)) + list(zip(outhdr, outhdr)) +
list(zip(add_list, add_list)))
outhdr = [tmp_dict[x] for x in keep_list
] #first include already existing columns from -c argument
if add_list:
outhdr += [x for x in add_list if x not in outhdr]
if drop_list:
outhdr = [
x for ix, x in enumerate(outhdr)
if (ix not in drop_list and x not in drop_list)
]
return outhdr