#!/usr/bin/env python
# coding: utf-8

# # Load the RDF data dump

# In[4]:


get_ipython().run_cell_magic('time', '', '\nimport pandas as pd\nimport matplotlib.pyplot as plt\nimport matplotlib.dates as mdates\n\nfrom rdflib import ConjunctiveGraph\n\nns = {"nb": "http://bise-eu.info/core-ontology#",\n      "dc": "http://dcterms/",\n      "p-plan": "http://purl.org/net/p-plan#",\n      "edam": "http://purl.obolibrary.org/obo/edam#"}\n\ng = ConjunctiveGraph()\n\n### latest version of the RDF dataset dump\ng.parse("https://raw.githubusercontent.com/bio-tools/content/master/datasets/bise-ontology-biii-dump.ttl", format="turtle")\ng.parse("bise-linked-data-webapp/static/data/EDAM-bioimaging_alpha03.owl")\nprint(str(len(g)) + \' triples in Biii data graph\')\n')


# # Level-0 entries

# In[7]:


get_ipython().run_line_magic('matplotlib', 'inline')

plt.figure(figsize=(16, 10))

soft_dates = []
date_query = """
SELECT ?soft ?date WHERE {
    ?soft a <http://biii.eu/software> .
    ?soft dc:created ?date .
}
"""
results = g.query(date_query, initNs=ns)
for r in results:
    soft_dates.append({'soft':str(r['soft']), 'created':str(r['date'])})

print(len(soft_dates))

df = pd.DataFrame(soft_dates)
df["created"] = df["created"].astype("datetime64")

#df2 = df.groupby([df["created"].dt.year, df["created"].dt.month]).count()
df2 = df.groupby([df["created"].dt.year]).count()

df2['soft'].cumsum().plot(kind='bar')
#df2['soft'].cumsum()


# # Sparse entries

# In[8]:


#%%time
get_ipython().run_line_magic('matplotlib', 'inline')

plt.figure(figsize=(16, 10))
#plt.figure(figsize=(9, 6))

soft_dates = []
date_query = """
SELECT DISTINCT ?soft ?date WHERE {
    ?soft a <http://biii.eu/software> .
    ?soft dc:created ?date .
    ?soft dc:title ?title .
    ?soft rdfs:comment ?description .
    FILTER (
        NOT EXISTS {?soft nb:hasImplementation ?i} 
        || NOT EXISTS {?soft nb:hasLocation ?loc} 
        || NOT EXISTS {?soft nb:hasIllustration ?ill} 
        || NOT EXISTS {?soft nb:hasAuthor ?auth}
        || NOT EXISTS {?soft nb:hasFunction ?func}
        || NOT EXISTS {?soft nb:openess ?lic_op}
        || NOT EXISTS {?soft nb:hasLicense ?lic}
        || NOT EXISTS {?soft nb:hasSupportedImageDimension ?dim}
        || NOT EXISTS {?soft nb:requires ?dep}
        || NOT EXISTS {?soft nb:hasProgrammingLanguage ?prog}
        || NOT EXISTS {?soft nb:hasPlatform ?plat}
        || NOT EXISTS {?soft nb:hasTopic ?topic}
        || NOT EXISTS {?soft nb:hasReferencePublication ?ref}
        || NOT EXISTS {?soft nb:hasDocumentation ?doc}
        || NOT EXISTS {?soft nb:hasComparison ?comp}
        || NOT EXISTS {?soft nb:hasDOI ?doi}
        || NOT EXISTS {?soft nb:hasUsageExample ?usag}
        || NOT EXISTS {?soft nb:hasTrainingMaterial ?train}
    )
}
"""
results = g.query(date_query, initNs=ns)
for r in results:
    soft_dates.append({'soft':str(r['soft']), 'created':str(r['date'])})

print(len(soft_dates))
df = pd.DataFrame(soft_dates)
df["created"] = df["created"].astype("datetime64")

#df2 = df.groupby([df["created"].dt.year, df["created"].dt.month]).count()
df2 = df.groupby([df["created"].dt.year]).count()
#df2

df2['soft'].cumsum().plot(kind='bar')
#df2['soft'].cumsum()


# # Detailed entries

# In[9]:


get_ipython().run_line_magic('matplotlib', 'inline')

plt.figure(figsize=(16, 10))
#plt.figure(figsize=(9, 6))

soft_dates = []
date_query = """
SELECT DISTINCT ?soft ?date WHERE {
    ?soft a <http://biii.eu/software> .
    ?soft dc:created ?date .
    ?soft dc:title ?title .
    ?soft rdfs:comment ?description .
    ?soft nb:hasImplementation ?i .
    ?soft nb:hasLocation ?loc .
    ?soft nb:hasIllustration ?ill .
    ?soft nb:hasAuthor ?auth .
    ?soft nb:hasFunction ?func .
    ?soft nb:openess ?lic_op .
    ?soft nb:hasLicense ?lic .
    
    ?soft nb:hasSupportedImageDimension|nb:requires ?usage .
    
    ?soft nb:hasProgrammingLanguage ?prog .
    ?soft nb:hasPlatform ?plat .
    ?soft nb:hasTopic ?topic .

}
"""
results = g.query(date_query, initNs=ns)
for r in results:
    soft_dates.append({'soft':str(r['soft']), 'created':str(r['date'])})

print(len(soft_dates))

df = pd.DataFrame(soft_dates)
df["created"] = df["created"].astype("datetime64")

#df2 = df.groupby([df["created"].dt.year, df["created"].dt.month]).count()
df2 = df.groupby([df["created"].dt.year]).count()

df2['soft'].cumsum().plot(kind='bar')
#df2['soft'].cumsum()

df


# # Comprehensive entries

# In[11]:


get_ipython().run_line_magic('matplotlib', 'inline')

plt.figure(figsize=(16, 10))
#plt.figure(figsize=(9, 6))

soft_dates = []
date_query = """
SELECT DISTINCT ?soft ?date WHERE {
    ?soft a <http://biii.eu/software> .
    ?soft dc:created ?date .
    ?soft dc:title ?title .
    ?soft rdfs:comment ?description .
    ?soft nb:hasImplementation ?i .
    ?soft nb:hasLocation ?loc .
    ?soft nb:hasIllustration ?ill .
    ?soft nb:hasAuthor ?auth .
    ?soft nb:hasFunction ?func .
    ?soft nb:openess ?lic_op .
    ?soft nb:hasLicense ?lic .
    
    ?soft nb:hasSupportedImageDimension|nb:requires ?usage .
    
    ?soft nb:hasProgrammingLanguage ?prog .
    ?soft nb:hasPlatform ?plat .
    ?soft nb:hasTopic ?topic .
    
    ?soft nb:hasReferencePublication|nb:hasDocumentation|nb:hasComparison ?ref .
    ?soft nb:hasDOI ?doi .
    ?soft nb:hasUsageExample|nb:hasTrainingMaterial ?train
    
}
"""
results = g.query(date_query, initNs=ns)
for r in results:
    soft_dates.append({'soft':str(r['soft']), 'created':str(r['date'])})

print(len(soft_dates))

df = pd.DataFrame(soft_dates)
df["created"] = df["created"].astype("datetime64")

#df2 = df.groupby([df["created"].dt.year, df["created"].dt.month]).count()
df2 = df.groupby([df["created"].dt.year]).count()

df2['soft'].cumsum().plot(kind='bar')
#df2['soft'].cumsum()

df


# In[ ]: