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

# In[2]:


from IPython.core.display import HTML
HTML("""
<style>
div {font-family:Open Sans,sans-serif;}
h1 {color:white;background-color: #446a7f;padding:10px}
h2{color:black; background-color:#b3b6a0; padding:10px}
h3{color:#78BE21; background-color:#446a7f; padding:10px}
h4{color:white; background-color:#78BE21;padding:5px}
.info{background-color:}
img {border: 1px;border-color:black;}
</style>
""")


# #  Collections Data to Linked Art - Cleveland Museum of Art
# 
# This Jupyter notebook transforms collection data from the Cleveland Museum of Art to a Linked Art representation in a JSON-LD format.
# 
# ## Collection Data
# 
# The input data file is available from the Cleveland Museum of Art's GitHub repository:
# https://github.com/ClevelandMuseumArt/openaccess
# 
# ## Transformation Process
# 
# 1. Read CSV file
# 2. Convert CSV file to Python Dictionary
# 3. Create a field mapping between the CSV file fields and the Linked Art data model
# 4. Create an object property dictionary

# In[2]:


try:
    import IPython
except:
    get_ipython().run_line_magic('pip', 'install IPython')
    import IPython   
    
from IPython.display import display,IFrame,HTML, Javascript
 
try:
    import json
except:
    get_ipython().run_line_magic('pip', 'install json')
    import json 
  
try:
    import csv
except:
    get_ipython().run_line_magic('pip', 'install csv')
    import csv
    
import os

try:
    import cromulent 
except:
    get_ipython().run_line_magic('pip', 'install cromulent')
    import cromulent

from cromulent.model import factory
    
import lib.linkedart as la 

try:
    import pandas as pd
except:
    get_ipython().run_line_magic('pip', 'install pandas')
    import pandas as pd
    
    


# ### Parse Collection Data CSV File

# In[3]:


file = './data/cma/input/data.csv'
mpg = pd.read_csv(file,low_memory=False)
mpg.head()


# ### Read Collection Data CSV file into Python Dictionary
# 
# - Remove Byte Order Mark (BOM) from CSV file 
# - Use `csv.DictReader()` to create an object that operates like a regular reader but maps the information in each row to a dict whose keys are given by the optional fieldnames parameter. 
# 
# 

# In[4]:


# remove BOM see - https://stackoverflow.com/questions/8898294/convert-utf-8-with-bom-to-utf-8-with-no-bom-in-python
s = open(file, mode='r', encoding='utf-8-sig').read()
open(file, mode='w', encoding='utf-8').write(s)

# open file and read into Python dictionary
allObjects = csv.DictReader(open(file, mode='r',encoding='utf-8'))

display(HTML("<H3>Example Record in Python Dictionary</H3>"))
for obj in allObjects:
    print(json.dumps(obj,indent=2))
    break  


# #### Further Reading 
# 
# - Python CSV https://docs.python.org/3/library/csv.html
# - Byte Order Mark https://en.wikipedia.org/wiki/Byte_order_mark

# ### Create field mapping

# In[5]:


mapp =  {
    "id":"id",
    "accession_number":"accession_number",
    "accession_date": "",
    "classification" : "type",
    "title": "title",
    "alt_title": "title_in_original_language",
    "notes": "tombstone",
    "date_created":"creation_date",
    "date_created_earliest": "creation_date_earliest",
    "date_created_latest": "creation_date_latest",
    "created_period":"culture",
    "created_dynasty":"",
    "created_inscriptions":"inscriptions",
    "created_notes": "fun_fact",
    "creator":"creators",
    "physical_medium": "Medium",
    "physical_style": "",
    "physical_technique": "technique",
    "physical_description": "",
    "physical_dimensions": "measurements",
    "created_provenance": "provenance" ,
    "credit_line": "creditline",
    "collection" : "department",
    "current_status" : "current_location",
    "current_location": "current_location",
    "homepage": "url"
    
}



# ### Create object property dictionary

# In[6]:


def createObjProp(obj,mapp):
    objProp = {}
    csv_keys = list(obj.keys())
    for key in csv_keys:
        for prop in mapp:
            if key == mapp[prop]:
                if prop == "creator":
                    objProp[prop] = {"name": obj[key]}
                else:
                    objProp[prop] = obj[key]
    
    objProp["current_owner"] = {"name":"Cleveland Museum of Art",
                                "location":"Cleveland,Ohio",
                                "type": "http://vocab.getty.edu/aat/300312281" ,
                                "type_label": ""}
    return objProp


#  baseURI for JSON-LD document
baseURI = "https://clevelandart.org/art/"


for index,obj in enumerate(allObjects):
    if index >= 5:
        break
    objProp = createObjProp(obj,mapp)
    id = obj[mapp.get("id")]
    object_uri = baseURI + id
    
    objLA = None
    objLA = la.createObjDesc(objProp,la.objTypes,object_uri)
    # write to file 
    text_file = open("./data/cma/output/json/all/" + id + ".json", "wt")
    n = text_file.write(factory.toString(objLA, compact=False))
    text_file.close()


# In[7]:


from IPython.display import display,HTML,Javascript
HTML('<h1>Data Visualisation</h1><div id="example" style="width:3000px;height:100%"/>')


# In[8]:


from IPython.core.display import Javascript

code2 = "var file = './data/cma/output/json/all/" + id + ".json';"\
        "var selector = '#example';" \
        "visjsonld(file, selector); "  

with open('src/js/visld.js', 'r') as _jscript:
    code = _jscript.read() + code2

Javascript(code)




# In[9]:


from IPython.core.display import Javascript, HTML


def fn(fpath):       # 1.Get file names from directory
    file_list=os.listdir(r"" + fpath)
    cnt =1 
    for file in file_list:
        cnt = cnt+1
        
        display(HTML("<a target='_new' href='" + fpath + file + "'>" + file + "</a>"))


display(HTML("<h1>File list</h1><p>Click on a link to view created Linked Art JSON-LD file</p>"))
    
fn("./data/cma/output/json/all/")


  


# In[ ]:





# In[ ]: