def test_sniffer(self):
output = romanesco.convert(
"table",
{
"format": "csv",
"url": "file://" + os.path.join("data", "test.csv")
},
{"format": "rows"}
)
self.assertEqual(len(output["data"]["fields"]), 32)
self.assertEqual(output["data"]["fields"][:3], [
"FACILITY", "ADDRESS", "DATE OF INSPECTION"
])
self.assertEqual(len(output["data"]["rows"]), 14)
flu = romanesco.load(os.path.join(
self.analysis_path, "xdata", "flu.json"))
output = romanesco.run(
flu,
inputs={},
outputs={"data": {"type": "table", "format": "rows"}}
)
self.assertEqual(output["data"]["data"]["fields"][:3], [
"Date", "United States", "Alabama"
])
def test_pgls(self):
pgls = romanesco.load(os.path.join(self.arbor_path, "pgls.json"))
tree_file = os.path.join("data", "anolis.phy")
table_file = os.path.join("data", "anolisDataAppended.csv")
romanesco.run(
pgls, {
"tree": {
"format": "newick",
"url": "file://" + tree_file
},
"table": {
"format": "csv",
"url": "file://" + table_file
},
"correlation": {
"format": "text",
"data": "BM"
},
"ind_variable": {
"format": "text",
"data": "SVL"
},
"dep_variable": {
"format": "text",
"data": "PCI_limbs"
}
})
def package_analyses(search_paths, output_dir):
if not os.path.exists(output_dir):
os.mkdir(output_dir)
for path in search_paths:
for filename in glob.glob(os.path.join(path, "*.json")):
analysis = romanesco.load(filename)
with open(os.path.join(output_dir, os.path.basename(filename)), 'w') as f:
json.dump(analysis, f)
def package_analyses(search_paths, output_dir):
if not os.path.exists(output_dir):
os.mkdir(output_dir)
for path in search_paths:
for filename in glob.glob(os.path.join(path, "*.json")):
analysis = romanesco.load(filename)
with open(os.path.join(output_dir, os.path.basename(filename)),
'w') as f:
json.dump(analysis, f)
def test_pgls(self):
pgls = romanesco.load(os.path.join(self.arbor_path, "pgls.json"))
tree_file = os.path.join("data", "anolis.phy")
table_file = os.path.join("data", "anolisDataAppended.csv")
romanesco.run(
pgls,
{
"tree": {"format": "newick", "url": "file://" + tree_file},
"table": {"format": "csv", "url": "file://" + table_file},
"correlation": {"format": "text", "data": "BM"},
"ind_variable": {"format": "text", "data": "SVL"},
"dep_variable": {"format": "text", "data": "PCI_limbs"}
}
)
def test_cont2disc(self):
cont2disc = romanesco.load(
os.path.join(self.arbor_path, "continuous_to_discrete.json"))
table_file = os.path.join("data", "anolisDataAppended.csv")
romanesco.run(
cont2disc,
{
"table": {"format": "csv", "url": "file://" + table_file},
"column": {"format": "text", "data": "SVL"},
"thresh": {"format": "number", "data": 3.5}
},
{
"newtable": {"format": "rows"}
}
)
def import_analyses(client, analyses_path):
# First get the minerva analysis folder
minerva_analyses_folder = client.get('/minerva_analysis/folder')
if not minerva_analyses_folder['folder']:
minerva_analyses_folder = client.post('/minerva_analysis/folder')
minerva_analyses_folder = minerva_analyses_folder['folder']
# Note: this will clobber any analyses that have the same name
items = {v['name']: v for v in
client.listItem(minerva_analyses_folder['_id'])}
# look for specific analysis subfolders
# e.g. analyses/bsve
for analysis_subfolder in os.listdir(analyses_path):
analysis_path = os.path.join(analyses_path, analysis_subfolder)
# If there is an analysis.json, it is a Romanesco analysis
romanesco_analysis = os.path.join(analysis_path, 'analysis.json')
metadata = {}
minerva_metadata = {}
if os.path.exists(romanesco_analysis):
analysis = romanesco.load(romanesco_analysis)
analysis_name = analysis['name']
metadata['analysis'] = analysis
# set the analysis_type based on folder name
minerva_metadata['analysis_type'] = analysis_path.split('/')[-1]
minerva_metadata['analysis_name'] = analysis_name
else:
# look for a minerva.json
minerva_metadata_path = os.path.join(analysis_path, 'minerva.json')
with open(minerva_metadata_path) as minerva_metadata_file:
minerva_metadata = json.load(minerva_metadata_file)
analysis_name = minerva_metadata['analysis_name']
if analysis_name not in items.keys():
items[analysis_name] = client.createItem(minerva_analyses_folder['_id'],
analysis_name,
analysis_name)
analysis_item = items[analysis_name]
# Set the minerva metadata
# add the item_id as the analysis_id
minerva_metadata['analysis_id'] = analysis_item['_id']
metadata['minerva'] = minerva_metadata
client.addMetadataToItem(analysis_item['_id'], metadata)
def import_analyses(client, analyses_path):
# First get the minerva analysis folder
minerva_analyses_folder = client.get('/minerva_analysis/folder')
if not minerva_analyses_folder['folder']:
minerva_analyses_folder = client.post('/minerva_analysis/folder')
minerva_analyses_folder = minerva_analyses_folder['folder']
# look for specific analysis subfolders
# e.g. analyses/bsve
for analysis_subfolder in os.listdir(analyses_path):
analysis_path = os.path.join(analyses_path, analysis_subfolder)
# If there is an analysis.json, it is a Romanesco analysis
romanesco_analysis = os.path.join(analysis_path, 'analysis.json')
metadata = {}
minerva_metadata = {}
if os.path.exists(romanesco_analysis):
analysis = romanesco.load(romanesco_analysis)
analysis_name = analysis['name']
metadata['analysis'] = analysis
# set the analysis_type based on folder name
minerva_metadata['analysis_type'] = analysis_path.split('/')[-1]
minerva_metadata['analysis_name'] = analysis_name
else:
# look for a minerva.json
minerva_metadata_path = os.path.join(analysis_path, 'minerva.json')
with open(minerva_metadata_path) as minerva_metadata_file:
minerva_metadata = json.load(minerva_metadata_file)
analysis_name = minerva_metadata['analysis_name']
# See if we already have an analysis with that name
items = client.listItem(minerva_analyses_folder['_id'], analysis_name)
if len(items) == 0:
analysis_item = client.createItem(minerva_analyses_folder['_id'], analysis_name, analysis_name)
elif len(items) > 1:
raise Exception('More than one item found with name: %s' % analysis_name)
else:
analysis_item = items[0]
# Set the minerva metadata
# add the item_id as the analysis_id
minerva_metadata['analysis_id'] = analysis_item['_id']
metadata['minerva'] = minerva_metadata
client.addMetadataToItem(analysis_item['_id'], metadata)
def test_mammal_tree(self):
mammal = romanesco.load(
os.path.join("data", "Mammal tree extraction.json"))
romanesco.run(
mammal,
{
"table": {
"format": "csv",
"url": "file://" +
os.path.join("data", "mammal_lnMass_tiny.csv")
},
"outRowCount": {"format": "number", "data": 19}
},
{
"tree": {"format": "nested"}
}
)
def test_mammal_tree(self):
mammal = romanesco.load(
os.path.join("data", "Mammal tree extraction.json"))
romanesco.run(
mammal, {
"table": {
"format":
"csv",
"url":
"file://" + os.path.join("data", "mammal_lnMass_tiny.csv")
},
"outRowCount": {
"format": "number",
"data": 19
}
}, {"tree": {
"format": "nested"
}})
def test_cont2disc(self):
cont2disc = romanesco.load(
os.path.join(self.arbor_path, "continuous_to_discrete.json"))
table_file = os.path.join("data", "anolisDataAppended.csv")
romanesco.run(
cont2disc, {
"table": {
"format": "csv",
"url": "file://" + table_file
},
"column": {
"format": "text",
"data": "SVL"
},
"thresh": {
"format": "number",
"data": 3.5
}
}, {"newtable": {
"format": "rows"
}})
def test_sniffer(self):
output = romanesco.convert(
"table", {
"format": "csv",
"url": "file://" + os.path.join("data", "test.csv")
}, {"format": "rows"})
self.assertEqual(len(output["data"]["fields"]), 32)
self.assertEqual(output["data"]["fields"][:3],
["FACILITY", "ADDRESS", "DATE OF INSPECTION"])
self.assertEqual(len(output["data"]["rows"]), 14)
flu = romanesco.load(
os.path.join(self.analysis_path, "xdata", "flu.json"))
output = romanesco.run(
flu,
inputs={},
outputs={"data": {
"type": "table",
"format": "rows"
}})
self.assertEqual(output["data"]["data"]["fields"][:3],
["Date", "United States", "Alabama"])
