def set_dir(self, dir_path):
'''
Initialize the FileManager
Creates the directories
Set global variables with absolute paths to the directories
Parameters
----------
dir_path: str
'''
self.dir = os.path.realpath(dir_path) # Absolute Path
self.cache_dir = os.path.join(self.dir, 'cached')
self.file_manager = FileManager(dir_path)
# Create paths if it doesn't exist
if not (os.access(self.dir, os.F_OK)):
os.makedirs(self.dir)
if not (os.access(self.cache_dir, os.F_OK)):
os.makedirs(self.cache_dir)
def set_dir(self, dir_path):
'''
Initialize the FileManager
Creates the directories
Set global variables with absolute paths to the directories
'''
self.dir = os.path.realpath(dir_path) # Absolute Path
self.cache_dir = os.path.join(self.dir, 'cached')
self.file_manager = FileManager(dir_path)
# Create paths if it doesn't exist
if not (os.access(self.dir, os.F_OK)):
os.makedirs(self.dir)
if not (os.access(self.cache_dir, os.F_OK)):
os.makedirs(self.cache_dir)
class DataAccess(object):
path = ''
'''
Class to manage the Access to the Data
Features
--------
1. Easy access to the data
2. Serialization of data
How to use
----------
Use one: Note: Option 2 overwrites option 1
1. Set the enviroment variable: FINANCEPATH
2. Set the Static Variable DataAccess.path
'''
def __init__(self):
if self.path != '':
self.set_dir(self.path)
else:
env_var = os.getenv("FINANCEPATH")
if env_var is not None:
self.set_dir(env_var)
else:
raise Exception('No path defined')
def set_dir(self, dir_path):
'''
Initialize the FileManager
Creates the directories
Set global variables with absolute paths to the directories
Parameters
----------
dir_path: str
'''
self.dir = os.path.realpath(dir_path) # Absolute Path
self.cache_dir = os.path.join(self.dir, 'cached')
self.file_manager = FileManager(dir_path)
# Create paths if it doesn't exist
if not (os.access(self.dir, os.F_OK)):
os.makedirs(self.dir)
if not (os.access(self.cache_dir, os.F_OK)):
os.makedirs(self.cache_dir)
def empty_dir(self, delete=True):
'''
Empty the directory of csv files. Do not delete the cache files/folder
Parameters
----------
delete:boolean, True if want to delete the folder too
'''
self.file_manager.empty_dir(delete)
def empty_cache(self, delete=True):
'''
Empty the directory of cached files. Does not delete the csv files/folder
Parameters
----------
delete: boolean, True if want to delete the folder too
'''
list_files = os.listdir(self.cache_dir) # Get the list of files
for f in list_files:
try:
os.remove(os.path.join(self.cache_dir, f))
except:
pass
if delete:
os.rmdir(self.cache_dir)
def empty_dirs(self, delete=True):
'''
Delete both cached and csv files.
Parameters
----------
delete:booelan, True if want to delete the folders too
'''
self.empty_cache(delete)
self.empty_dir(delete)
def save(self, data, name, extension='.data'):
'''
Saves a serialized (pickle) version of the data to the cache directory
Parameters
----------
data: object
name: str, identifier of the data
extension: str, extension of the filename
'''
h = hashlib.md5()
h.update(name.encode('utf8'))
filename = h.hexdigest() + extension
f = os.path.join(self.cache_dir, filename)
data.save(f)
def load(self, name, extension='.data'):
'''
Checks for an existing file name and if exists returns the data saved
Parameters
----------
name: str, identifier of the data
extension: str, extension of the filename
Returns
-------
data: object (usually pandas.DataFrame), if file was available; None otherwise.
'''
h = hashlib.md5()
h.update(name.encode('utf8'))
filename = h.hexdigest() + extension
f = os.path.join(self.cache_dir, filename)
if os.access(f, os.F_OK):
return pd.load(f)
def get_data(self, symbols, start, end, fields='Adj Close', save=True, useCache=True,
downloadMissing=True, ignoreMissing=True):
'''
Returns a pandas DataFrame with the data of the symbols and field
fields between the specified dates with the fields specified
Optional:
1. Load a serialized version of the data
2. Saves a serialized version of the data
3. If data is not available download the missing data
Parameters
----------
symbols_s: str or list of str
start: datetime, with the initial date
end: datetime, with the final date
fields: str or list of str
save: boolean, True if want to save the cache version
useCache: boolean: True if want to load a cached version (if available)
downloadMissing: boolean, True if want to download unavailable data
ignoreMissing=True
Returns
-------
data: pandas.DataFrame
'''
# 0. If ask for only one symbols or field convert it to a list of one item
if type(symbols) == str:
symbols = [symbols]
if type(fields) == str:
fields = [fields]
# 1. Load the Data, if requested
filename_id = "%s_%s_%s_%s" % ('_'.join(symbols), start.strftime('%m-%d-%Y'),
end.strftime('%m-%d-%Y'), '-'.join(fields))
if useCache == True:
data = self.load(filename_id)
if data is not None:
# 1.1 Data was cached before and loaded => return
return data
# 1. Data was not cached before need to load the data from csv files
# 1.1 Get the list of filenames from the FileManager
files = self.file_manager.get_filenames(symbols, start, end, downloadMissing, ignoreMissing)
# 1.2 We are going to create a pd.DataFrame from a dictionary of pd.Series
data_dic = {}
for f, symbol in zip(files, symbols):
# Create DataFrame from the csv
new_data = pd.read_csv(os.path.join(self.dir, f))
# Change the index of the DataFrame to be the date
new_data = new_data.set_index(pd.to_datetime(new_data['Date']))
for field in fields:
# For each field in fields, creates a new column
colname = ''
if len(symbols) == 1 and len(fields) == 1:
# Single symbol and Single field
colname = field
elif len(symbols) > 1 and len(fields) == 1:
# Multiple symbols and single fields
colname = symbol
elif len(symbols) == 1 and len(fields) > 1:
# Single symbol and Multiple fields
colname = field
else:
# Multiple symbols and multiple fields
colname = "%s %s" % (symbol, field)
# Adds the pd.Series to the dictionary
data_dic[colname] = new_data[field]
# 1.4. Create, slice and sort the data
data = pd.DataFrame(data_dic)
data = data.sort()[start:end] # Sort because Yahoo Finance data comes reverse
# Save a cache version if requested
if save == True:
self.save(data, filename_id)
return data
def download(self, symbols, start, end):
self.get_data(symbols, start, end, useCache=False, save=False)
class DataAccess(object):
path = ''
'''
Class to manage the Access to the Data
Features
--------
1. Easy access to the data
2. Serialization of data
How to use
----------
Use one: Note: Option 2 overwrites option 1
1. Set the enviroment variable: FINANCEPATH
2. Set the Static Variable DataAccess.path
'''
def __init__(self):
if self.path != '':
self.set_dir(self.path)
else:
env_var = os.getenv("FINANCEPATH")
if env_var is not None:
self.set_dir(env_var)
else:
raise Exception('No path defined')
def set_dir(self, dir_path):
'''
Initialize the FileManager
Creates the directories
Set global variables with absolute paths to the directories
Parameters
----------
dir_path: str
'''
self.dir = os.path.realpath(dir_path) # Absolute Path
self.cache_dir = os.path.join(self.dir, 'cached')
self.file_manager = FileManager(dir_path)
# Create paths if it doesn't exist
if not (os.access(self.dir, os.F_OK)):
os.makedirs(self.dir)
if not (os.access(self.cache_dir, os.F_OK)):
os.makedirs(self.cache_dir)
def empty_dir(self, delete=True):
'''
Empty the directory of csv files. Do not delete the cache files/folder
Parameters
----------
delete:boolean, True if want to delete the folder too
'''
self.file_manager.empty_dir(delete)
def empty_cache(self, delete=True):
'''
Empty the directory of cached files. Does not delete the csv files/folder
Parameters
----------
delete: boolean, True if want to delete the folder too
'''
list_files = os.listdir(self.cache_dir) # Get the list of files
for f in list_files:
try:
os.remove(os.path.join(self.cache_dir, f))
except:
pass
if delete:
os.rmdir(self.cache_dir)
def empty_dirs(self, delete=True):
'''
Delete both cached and csv files.
Parameters
----------
delete:booelan, True if want to delete the folders too
'''
self.empty_cache(delete)
self.empty_dir(delete)
def save(self, data, name, extension='.data'):
'''
Saves a serialized (pickle) version of the data to the cache directory
Parameters
----------
data: object
name: str, identifier of the data
extension: str, extension of the filename
'''
h = hashlib.md5()
h.update(name.encode('utf8'))
filename = h.hexdigest() + extension
f = os.path.join(self.cache_dir, filename)
data.save(f)
def load(self, name, extension='.data'):
'''
Checks for an existing file name and if exists returns the data saved
Parameters
----------
name: str, identifier of the data
extension: str, extension of the filename
Returns
-------
data: object (usually pandas.DataFrame), if file was available; None otherwise.
'''
h = hashlib.md5()
h.update(name.encode('utf8'))
filename = h.hexdigest() + extension
f = os.path.join(self.cache_dir, filename)
if os.access(f, os.F_OK):
return pd.load(f)
def get_data(self,
symbols,
start,
end,
fields='Adj Close',
save=True,
useCache=True,
downloadMissing=True,
ignoreMissing=True):
'''
Returns a pandas DataFrame with the data of the symbols and field
fields between the specified dates with the fields specified
Optional:
1. Load a serialized version of the data
2. Saves a serialized version of the data
3. If data is not available download the missing data
Parameters
----------
symbols_s: str or list of str
start: datetime, with the initial date
end: datetime, with the final date
fields: str or list of str
save: boolean, True if want to save the cache version
useCache: boolean: True if want to load a cached version (if available)
downloadMissing: boolean, True if want to download unavailable data
ignoreMissing=True
Returns
-------
data: pandas.DataFrame
'''
# 0. If ask for only one symbols or field convert it to a list of one item
if type(symbols) == str:
symbols = [symbols]
if type(fields) == str:
fields = [fields]
# 1. Load the Data, if requested
filename_id = "%s_%s_%s_%s" % (
'_'.join(symbols), start.strftime('%m-%d-%Y'),
end.strftime('%m-%d-%Y'), '-'.join(fields))
if useCache == True:
data = self.load(filename_id)
if data is not None:
# 1.1 Data was cached before and loaded => return
return data
# 1. Data was not cached before need to load the data from csv files
# 1.1 Get the list of filenames from the FileManager
files = self.file_manager.get_filenames(symbols, start, end,
downloadMissing, ignoreMissing)
# 1.2 We are going to create a pd.DataFrame from a dictionary of pd.Series
data_dic = {}
for f, symbol in zip(files, symbols):
# Create DataFrame from the csv
new_data = pd.read_csv(os.path.join(self.dir, f))
# Change the index of the DataFrame to be the date
new_data = new_data.set_index(pd.to_datetime(new_data['Date']))
for field in fields:
# For each field in fields, creates a new column
colname = ''
if len(symbols) == 1 and len(fields) == 1:
# Single symbol and Single field
colname = field
elif len(symbols) > 1 and len(fields) == 1:
# Multiple symbols and single fields
colname = symbol
elif len(symbols) == 1 and len(fields) > 1:
# Single symbol and Multiple fields
colname = field
else:
# Multiple symbols and multiple fields
colname = "%s %s" % (symbol, field)
# Adds the pd.Series to the dictionary
data_dic[colname] = new_data[field]
# 1.4. Create, slice and sort the data
data = pd.DataFrame(data_dic)
data = data.sort()[
start:end] # Sort because Yahoo Finance data comes reverse
# Save a cache version if requested
if save == True:
self.save(data, filename_id)
return data
def download(self, symbols, start, end):
self.get_data(symbols, start, end, useCache=False, save=False)
class DataAccess(object):
'''
Class to manage the Access to the Data
Features:
1. Easy access to the data
2. Download data from Yahoo! Finance
3. Serialization of the data
'''
def __init__(self, dir_path='./data/'):
self.set_dir(dir_path)
def set_dir(self, dir_path):
'''
Initialize the FileManager
Creates the directories
Set global variables with absolute paths to the directories
'''
self.dir = os.path.realpath(dir_path) # Absolute Path
self.cache_dir = os.path.join(self.dir, 'cached')
self.file_manager = FileManager(dir_path)
# Create paths if it doesn't exist
if not (os.access(self.dir, os.F_OK)):
os.makedirs(self.dir)
if not (os.access(self.cache_dir, os.F_OK)):
os.makedirs(self.cache_dir)
def empty_dir(self, delete=True):
'''
Empty the directory of .csv files. Do not delete the cache files/folder
Parameters:
delete=True - True if want to delete the folder too
'''
self.file_manager.empty_dir(delete)
def empty_cache(self, delete=True):
'''
Empty the directory of cached files. Do not delete the .csv files/folder
Parameters:
delete=True - True if want to delete the folder too
'''
list_files = os.listdir(self.cache_dir) # Get the list of files
for f in list_files:
try:
os.remove(os.path.join(self.cache_dir, f))
except:
pass
if delete:
os.rmdir(self.cache_dir)
def empty_dirs(self, delete=True):
'''
Delete both cached files and .csv files.
Parameters:
delete=True - True if want to delete the folders too
'''
self.empty_cache(delete)
self.empty_dir(delete)
def get_data(self, symbol_s, start_date, end_date, field_s, save=True, useCache=True,
downloadMissing=True):
'''
1. Checks if the data requeted has been previously cached:
1.1 if it was saved, then load the cache version
1.2 if not loads the data from the .csv files
1.2.1 saves the cache version (optional)
Args:
symbols_s - symbol (str) or list of symbols
start_date - datetime with the initial date
end_date - datetime with the final date
field_s - field (str) or list of fields
save=True - True if want to save the cache version
useCache=True - True if want to load a cached version, if available
downloadMissing=True - True if want to download unavailable data
Returns:
pandas.DataFrame - with the data of the symbols and fields requested
index:
'''
# 0. Generate and string which represents the data requested
filename_large = "%s_%s_%s_%s" % ('_'.join(symbol_s), start_date.strftime('%m-%d-%Y'),
end_date.strftime('%m-%d-%Y'), '-'.join(field_s))
# 0. Generates hash key of the string to reduce filename length
h = hashlib.md5()
h.update(filename_large.encode('utf8'))
filename = h.hexdigest() + ".data"
# 1. Load the Data
data = self.load(filename)
if data is not None and useCache == True:
# 1.1 Data was cached so return it
return data
else:
# 1.2 Data was not cached before need to load the data from csv files
df = self.get_data_from_files(symbol_s, start_date, end_date, field_s, downloadMissing)
if save == True:
# 1.2.1 Saves the cache version
self.save(df, filename)
return df
def load(self, name):
'''
Checks for an existing file name and if exists returns the data saved
'''
f = os.path.join(self.cache_dir, name)
if os.access(f, os.F_OK):
return pd.load(f)
else:
return None
def save(self, data, name):
'''
Save a serialized (pickle) version of the data to the cache directory
'''
f = os.path.join(self.cache_dir, name)
data.save(f)
def get_data_from_files(self, symbol_s, start_date, end_date, field_s, downloadMissing=True):
'''
Gets the data directly from the csv files
Args:
symbols_s - symbol (str) or list of symbols
start_date - datetime with the initial date
end_date - datetime with the final date
field_s - field (str) or list of fields
downloadMissing=True - True if want to download missing information from the internet
Returs:
pandas.DataFrame - with the data of the symbols and fields requested
index: DatetimeIndex
'''
# 0. If ask for only one symbols or field convert it to a list of one item
if type(symbol_s) == str:
symbol_s = [symbol_s]
if type(field_s) == str:
field_s = [field_s]
# 1. We are going to create a pd.DataFrame from a dictionary of pd.Series
data_dic = {}
# 1.1 Save the Indexes of the data
# 2. Get the file names with the information needed from the FileManager
files = self.file_manager.get_data(symbol_s, start_date, end_date, downloadMissing)
if type(files) == str:
files = [files]
for f, symbol in zip(files, symbol_s):
# for each file in files and symbol in symbol_s
n_data = pd.read_csv(os.path.join(self.dir, f))
# Needs to convert the string to datetime so the index of the Series is DatetimeIndex
n_data = n_data.set_index(pd.to_datetime(n_data['Date']))
for field in field_s:
# For each field in fields
colname = ''
if len(symbol_s) == 1 and len(field_s) == 1:
# Single symbol and Single field
colname = field
elif len(symbol_s) > 1 and len(field_s) == 1:
# Multiple symbols and single fields
colname = symbol
elif len(symbol_s) == 1 and len(field_s) > 1:
# Single symbol and Multiple fields
colname = field
else:
# Multiple symbols and multiple fields
colname = "%s %s" % (symbol, field)
# Adds the pd.Series to the dictionary
data_dic[colname] = n_data[field]
# 3. Create and return a pd.DataFrame from the dictionary of pd.Series
df = pd.DataFrame(data_dic)
df = df.sort() # Sort because Yahoo Finance data comes reverse
return df.ix[start_date:end_date] # Slice by date to only return what is important
