def get_lists():
# get a list of category tags from all scripts
from retriever import SCRIPT_LIST
SCRIPT_LIST = SCRIPT_LIST()
SCRIPT_LIST.sort(key=attrgetter('name'))
full_tags = set()
tag_heads = set()
for script in SCRIPT_LIST:
for tag in script.tags:
full_tags.update([tag])
if len(tag_tree(tag)) > 0:
tag_heads.update([tag_tree(tag)[0]])
full_tags = sorted(list(full_tags))
tag_heads = sorted(list(tag_heads))
lists = []
for head in tag_heads:
valid_scripts = [
script for script in SCRIPT_LIST if len([
tag for tag in script.tags
if len(tag_tree(tag)) > 0 and tag_tree(tag)[0] == head
]) > 0
]
lists.append(
Category(head,
valid_scripts,
children=children(head, valid_scripts, [
tag_tree(tag)[1:]
for tag in full_tags if tag_tree(tag)[0] == head
], 1)))
# Get list of additional datasets from scripts.config file
if os.path.isfile("scripts.config"):
other_scripts = []
config = open("scripts.config", 'rb')
for line in config:
if line:
try:
new_dataset = eval(line)
other_scripts.append(new_dataset)
except:
pass
other_scripts.sort(key=attrgetter('name'))
if len(other_scripts) > 0:
lists.append(Category("Custom", other_scripts))
for script in other_scripts:
lists[0].scripts.append(script)
choice_tree = Category("All Datasets", SCRIPT_LIST, children=lists)
return choice_tree
def get_lists():
# get a list of category tags from all scripts
from retriever import SCRIPT_LIST
SCRIPT_LIST = SCRIPT_LIST()
SCRIPT_LIST.sort(key=attrgetter('name'))
full_tags = set()
tag_heads = set()
for script in SCRIPT_LIST:
for tag in script.tags:
full_tags.update([tag])
if len(tag_tree(tag)) > 0:
tag_heads.update([tag_tree(tag)[0]])
full_tags = sorted(list(full_tags))
tag_heads = sorted(list(tag_heads))
lists = []
for head in tag_heads:
valid_scripts = [script for script in SCRIPT_LIST
if len([tag for tag in script.tags
if len(tag_tree(tag)) > 0 and
tag_tree(tag)[0] == head]) > 0]
lists.append(Category(head, valid_scripts,
children=children(head, valid_scripts,
[tag_tree(tag)[1:]
for tag in full_tags
if tag_tree(tag)[0] == head],
1))
)
# Get list of additional datasets from scripts.config file
if os.path.isfile("scripts.config"):
other_scripts = []
config = open("scripts.config", 'rb')
for line in config:
if line:
try:
new_dataset = eval(line)
other_scripts.append(new_dataset)
except:
pass
other_scripts.sort(key=attrgetter('name'))
if len(other_scripts) > 0:
lists.append(Category("Custom", other_scripts))
for script in other_scripts:
lists[0].scripts.append(script)
choice_tree = Category("All Datasets", SCRIPT_LIST,
children=lists)
return choice_tree
def download_public_data(datasets, data_dir='./data/'):
"""Download public datasets using the EcoData Retriever"""
from retriever import VERSION, SCRIPT_LIST, ENGINE_LIST
from retriever.lib.tools import choose_engine, get_opts
for dataset in datasets:
script_list = SCRIPT_LIST()
opts = get_opts(script_list,
args=[
'install', dataset, '-e', 's', '-f',
'downloaded_data.sqlite'
])
script = opts["script"]
engine = choose_engine(opts)
if isinstance(script, list):
for dataset in script:
print "=> Installing", dataset.name
dataset.download(engine, debug=debug)
else:
script.download(engine)
print "Datasets successfully downloaded."
def Find(self, evt):
dlg = wx.TextEntryDialog(self, 'Enter the keyword(s) to search for',
'Find', '')
dlg.ShowModal()
result = dlg.GetValue().strip()
if result:
search_terms = [
term.strip() for term in result.split(' ') if term.strip()
]
scripts = []
for script in SCRIPT_LIST():
if script.matches_terms(search_terms):
scripts.append(script)
if len(scripts) > 0:
results = Category(
"Search results: " + ', '.join(search_terms), scripts)
self.cat_list.AddChild(results, select=True)
else:
wx.MessageBox("Your search returned no results.", "No results")
dlg.Destroy()
"json": {
'engine': 'json',
'table_name': 'output_file_{table}.json'
},
"csv": {
'engine': 'csv',
'table_name': 'output_file_{table}.csv'
},
"sqlite": {
'engine': 'sqlite',
'file': dbfile,
'table_name': '{db}_{table}'
}
}
SCRIPT_LIST = SCRIPT_LIST()
TEST_ENGINES = {}
IGNORE = [
"forest-inventory-analysis", "bioclim", "prism-climate", "vertnet", "NPN",
"mammal-super-tree"
]
IGNORE = [dataset.lower() for dataset in IGNORE]
for engine in ENGINE_LIST:
if engine.abbreviation in engine_test:
try:
opts = engine_test[engine.abbreviation]
TEST_ENGINES[engine.abbreviation] = choose_engine(opts)
except:
TEST_ENGINES[engine.abbreviation] = None
pass
def main():
"""This function launches the EcoData Retriever."""
if len(sys.argv) == 1 or (len(sys.argv) > 1 and sys.argv[1] == 'gui'):
# if no command line args are passed, launch GUI
check_for_updates(graphical=False if 'darwin' in platform.platform().lower() else True)
lists = get_lists()
from retriever.app.main import launch_app
launch_app(lists)
else:
# otherwise, parse them
script_list = SCRIPT_LIST()
args = parser.parse_args()
if args.quiet:
sys.stdout = open(os.devnull, 'w')
if args.command == 'help':
parser.parse_args(['-h'])
if hasattr(args, 'compile') and args.compile:
script_list = SCRIPT_LIST(force_compile=True)
if args.command == 'update':
check_for_updates(graphical=False)
script_list = SCRIPT_LIST()
return
elif args.command == 'citation':
if args.dataset is None:
citation_path = os.path.join(os.path.split(__file__)[0], '../CITATION')
print citation_path
with open(citation_path) as citation_file:
print citation_file.read()
else:
scripts = name_matches(script_list, args.dataset)
for dataset in scripts:
print dataset.description
return
elif args.command == 'gui':
lists = get_lists()
from retriever.app.main import launch_app
launch_app(lists)
return
elif args.command == 'new':
f = open(args.filename, 'w')
f.write(sample_script)
f.close()
return
if args.command == 'ls' or args.dataset is None:
import lscolumns
#If scripts have never been downloaded there is nothing to list
if not script_list:
print "No scripts are currently available. Updating scripts now..."
check_for_updates(graphical=False)
print "\n\nScripts downloaded.\n"
script_list = SCRIPT_LIST()
all_scripts = set([script.shortname for script in script_list])
all_tags = set(["ALL"] +
[tag.strip().upper() for script in script_list for tagset in script.tags for tag in tagset.split('>')])
print "Available datasets (%s):" % len(all_scripts)
lscolumns.printls(sorted(list(all_scripts), key=lambda s: s.lower()))
print "Groups:"
lscolumns.printls(sorted(list(all_tags)))
return
engine = choose_engine(args.__dict__)
if hasattr(args, 'debug') and args.debug: debug = True
else: debug = False
scripts = name_matches(script_list, args.dataset)
if scripts:
for dataset in scripts:
print "=> Installing", dataset.name
try:
dataset.download(engine, debug=debug)
dataset.engine.final_cleanup()
except KeyboardInterrupt:
pass
except Exception as e:
print e
if debug: raise
print "Done!"
else:
print "The dataset %s isn't currently available in the Retriever" % (args.dataset)
print "Run 'retriever -ls to see a list of currently available datasets"
from builtins import str
from retriever import VERSION, COPYRIGHT
from retriever.lib.repository import check_for_updates
from retriever import SCRIPT_LIST
# Create the .rst file for the available datasets
datasetfile = open("datasets.rst", "w")
datasetfile_title = """
==================
Datasets Available
==================
"""
check_for_updates()
script_list = SCRIPT_LIST()
# write the title of dataset rst file
datasetfile.write(datasetfile_title)
# get info from the scripts
for script_num, script in enumerate(script_list, start=1):
if script.ref.strip():
reference_link = script.ref
elif bool(script.urls.values()):
reference_link = script.urls.values()[0].rpartition('/')[0]
else:
reference_link = ""
datasetfile.write("| " + str(script_num) +
". **{}** \n| shortname: {}\n| reference: {}\n\n".format(
script.name, script.shortname, reference_link))
from __future__ import print_function
from builtins import input
import os
import json
from time import sleep
from retriever import SCRIPT_LIST, HOME_DIR
short_names = [script.shortname.lower() for script in SCRIPT_LIST()]
def is_empty(val):
"""Check if a variable is an empty string or an empty list"""
return val == "" or val == []
def clean_input(prompt="", split_char='', ignore_empty=False, dtype=None):
"""Clean the user-input from the CLI before adding it"""
while True:
val = input(prompt).strip()
# split to list type if split_char specified
if split_char != "":
val = [v.strip() for v in val.split(split_char) if v.strip() != ""]
# do not ignore empty input if not allowed
if not ignore_empty and is_empty(val):
print("\tError: empty input. Need one or more values.\n")
continue
# ensure correct input datatype if specified
if not is_empty(val) and dtype is not None:
try:
if not type(eval(val)) == dtype:
print("\tError: input doesn't match required type ", dtype,
def main():
"""This function launches the Data Retriever."""
if len(sys.argv) == 1:
# if no command line args are passed, show the help options
parser.parse_args(['-h'])
else:
# otherwise, parse them
script_list = SCRIPT_LIST()
args = parser.parse_args()
if args.command == "install" and not args.engine:
parser.parse_args(['install','-h'])
if args.quiet:
sys.stdout = open(os.devnull, 'w')
if args.command == 'help':
parser.parse_args(['-h'])
if hasattr(args, 'compile') and args.compile:
script_list = SCRIPT_LIST(force_compile=True)
if args.command == 'defaults':
for engine_item in engine_list:
print("Default options for engine ", engine_item.name)
for default_opts in engine_item.required_opts:
print(default_opts[0], " ", default_opts[2])
print()
return
if args.command == 'update':
check_for_updates()
script_list = SCRIPT_LIST()
return
elif args.command == 'citation':
if args.dataset is None:
print("\nCitation for retriever:\n")
print(CITATION)
else:
scripts = name_matches(script_list, args.dataset)
for dataset in scripts:
print("\nDataset: {}".format(dataset.name))
print("Citation: {}".format(dataset.citation))
print("Description: {}\n".format(dataset.description))
return
elif args.command == 'new':
f = open(args.filename, 'w')
f.write(sample_script)
f.close()
return
elif args.command == 'reset':
reset_retriever(args.scope)
return
elif args.command == 'new_json':
# create new JSON script
create_json()
return
elif args.command == 'edit_json':
# edit existing JSON script
for json_file in [filename for filename in
os.listdir(os.path.join(HOME_DIR, 'scripts')) if filename[-5:] == '.json']:
if json_file.lower().find(args.filename.lower()) != -1:
edit_json(json_file)
return
raise Exception("File not found")
elif args.command == 'delete_json':
# delete existing JSON script
for json_file in [filename for filename in
os.listdir(os.path.join(HOME_DIR, 'scripts')) if filename[-5:] == '.json']:
if json_file.lower().find(args.dataset.lower()) != -1:
confirm = input("Really remove " + json_file +
" and all its contents? (y/N): ")
if confirm.lower().strip() in ['y', 'yes']:
# raise Exception(json_file)
os.remove(os.path.join(HOME_DIR, 'scripts', json_file))
try:
os.remove(os.path.join(
HOME_DIR, 'scripts', json_file[:-4] + 'py'))
except:
# Not compiled yet
pass
return
raise Exception("File not found")
if args.command == 'ls':
# If scripts have never been downloaded there is nothing to list
if not script_list:
print("No scripts are currently available. Updating scripts now...")
check_for_updates()
print("\n\nScripts downloaded.\n")
script_list = SCRIPT_LIST()
all_scripts = []
for script in script_list:
if script.shortname:
if args.l is not None:
script_name = script.name + "\nShortname: " + script.shortname + "\n"
if script.tags:
script_name += "Tags: " + \
str([tag for tag in script.tags]) + "\n"
not_found = 0
for term in args.l:
if script_name.lower().find(term.lower()) == -1:
not_found = 1
break
if not_found == 0:
all_scripts.append(script_name)
else:
script_name = script.shortname
all_scripts.append(script_name)
all_scripts = sorted(all_scripts, key=lambda s: s.lower())
print("Available datasets : {}\n".format(len(all_scripts)))
if args.l is None:
from retriever import lscolumns
lscolumns.printls(sorted(all_scripts, key=lambda s: s.lower()))
else:
count = 1
for script in all_scripts:
print("%d. %s" % (count, script))
count += 1
return
engine = choose_engine(args.__dict__)
if hasattr(args, 'debug') and args.debug:
debug = True
else:
debug = False
sys.tracebacklimit = 0
if hasattr(args, 'debug') and args.not_cached:
use_cache = False
else:
use_cache = True
if args.dataset is not None:
scripts = name_matches(script_list, args.dataset)
else:
raise Exception("no dataset specified.")
if scripts:
for dataset in scripts:
print("=> Installing", dataset.name)
try:
dataset.download(engine, debug=debug, use_cache=use_cache)
dataset.engine.final_cleanup()
except KeyboardInterrupt:
pass
except Exception as e:
print(e)
if debug:
raise
print("Done!")
else:
print("The dataset {} isn't currently available in the Retriever".format(
args.dataset))
print("Run 'retriever ls to see a list of currently available datasets")
def main():
"""This function launches the EcoData Retriever."""
if len(sys.argv) == 1 or (len(sys.argv) > 1 and sys.argv[1] == 'gui'):
# if no command line args are passed, launch GUI
check_for_updates(graphical=False if current_platform == 'darwin' else True)
lists = get_lists()
from retriever.app.main import launch_app
launch_app(lists)
else:
# otherwise, parse them
script_list = SCRIPT_LIST()
args = parser.parse_args()
if args.quiet:
sys.stdout = open(os.devnull, 'w')
if args.command == 'help':
parser.parse_args(['-h'])
if hasattr(args, 'compile') and args.compile:
script_list = SCRIPT_LIST(force_compile=True)
if args.command == 'update':
check_for_updates(graphical=False)
script_list = SCRIPT_LIST()
return
elif args.command == 'citation':
if args.dataset is None:
citation_path = os.path.join(os.path.split(__file__)[0], '../CITATION')
print "\nCitation for retriever:\n"
with open(citation_path) as citation_file:
print citation_file.read()
else:
scripts = name_matches(script_list, args.dataset)
for dataset in scripts:
print ("\nCitation: {}".format(dataset.citation))
print ("Description: {}\n".format(dataset.description))
return
elif args.command == 'gui':
lists = get_lists()
from retriever.app.main import launch_app
launch_app(lists)
return
elif args.command == 'new':
f = open(args.filename, 'w')
f.write(sample_script)
f.close()
return
elif args.command == 'reset':
reset_retriever(args.scope)
return
if args.command == 'ls' or args.dataset is None:
# If scripts have never been downloaded there is nothing to list
if not script_list:
print "No scripts are currently available. Updating scripts now..."
check_for_updates(graphical=False)
print "\n\nScripts downloaded.\n"
script_list = SCRIPT_LIST()
all_scripts = []
for script in script_list:
if script.name:
if args.l!=None:
script_name = script.name + "\nShortname: " + script.shortname+"\n"
if script.tags:
script_name += "Tags: "+str([tag for tag in script.tags])+"\n"
not_found = 0
for term in args.l:
if script_name.lower().find(term.lower()) == -1:
not_found = 1
break
if not_found == 0:
all_scripts.append(script_name)
else:
script_name = script.shortname
all_scripts.append(script_name)
all_scripts = sorted(all_scripts, key=lambda s: s.lower())
print "Available datasets : {}\n".format(len(all_scripts))
if args.l==None:
import lscolumns
lscolumns.printls(sorted(all_scripts, key=lambda s: s.lower()))
else:
count = 1
for script in all_scripts:
print ("%d. %s"%(count, script))
count += 1
return
engine = choose_engine(args.__dict__)
if hasattr(args, 'debug') and args.debug:
debug = True
else:
debug = False
scripts = name_matches(script_list, args.dataset)
if scripts:
for dataset in scripts:
print "=> Installing", dataset.name
try:
dataset.download(engine, debug=debug)
dataset.engine.final_cleanup()
except KeyboardInterrupt:
pass
except Exception as e:
print e
if debug: raise
print "Done!"
else:
print "The dataset {} isn't currently available in the Retriever".format(args.dataset)
print "Run 'retriever ls to see a list of currently available datasets"