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

# # Sample workflow: 2000 block group parts to 2010 tracts
# 
# ## Starting from a subset of 2010 Delaware blocks
# 
# For further background information see:
# * **Schroeder, J. P**. 2007. *Target-density weighting interpolation and uncertainty evaluation for temporal analysis of census data*. Geographical Analysis 39 (3):311–335.
# 
# #### NHGIS [block crosswalks](https://www.nhgis.org/user-resources/geographic-crosswalks)

# In[1]:


get_ipython().run_line_magic('load_ext', 'watermark')
get_ipython().run_line_magic('watermark', '')


# In[2]:


import nhgisxwalk
import inspect
import pandas

get_ipython().run_line_magic('load_ext', 'autoreload')
get_ipython().run_line_magic('autoreload', '2')
get_ipython().run_line_magic('watermark', '-w')
get_ipython().run_line_magic('watermark', '-iv')


# ### Source and target years for the crosswalk

# In[3]:


source_year, target_year = "2000", "2010"


# ### Source-target building base

# In[4]:


subset_data_dir = "../testing_data_subsets"
base_xwalk_name = "/nhgis_blk%s_blk%s_gj.csv.zip" % (source_year, target_year)
base_xwalk_file = subset_data_dir + base_xwalk_name
data_types = nhgisxwalk.str_types(["GJOIN%s"%source_year, "GJOIN%s"%target_year])
base_xwalk = pandas.read_csv(base_xwalk_file, index_col=0, dtype=data_types)
base_xwalk.head()


# ### Convenience code shorthand/lookup

# In[5]:


print(inspect.getsource(nhgisxwalk.valid_geo_shorthand))


# In[6]:


nhgisxwalk.valid_geo_shorthand(shorthand_name=False)


# ### Instantiate an `nhgisxwalk.GeoCrossWalk` object
# ##### see [nhgisxwalk.GeoCrossWalk](https://github.com/jGaboardi/nhgisxwalk/blob/92b4fe55de0a9c53d0315dcda8ec121faaf20aef/nhgisxwalk/geocrosswalk.py#L19) for full details

# In[7]:


nhgisxwalk.desc_code_2000_SF1b


# In[8]:


input_vars = [
    nhgisxwalk.desc_code_2000_SF1b["Persons"]["Total"],
    nhgisxwalk.desc_code_2000_SF1b["Families"]["Total"],
    nhgisxwalk.desc_code_2000_SF1b["Households"]["Total"],
    nhgisxwalk.desc_code_2000_SF1b["Housing Units"]["Total"]
]
input_vars


# In[9]:


input_var_tags = ["pop", "fam", "hh", "hu"]


# In[10]:


subset_state = "10"
bgp2000_to_trt2010 = nhgisxwalk.GeoCrossWalk(
    base_xwalk,
    source_year=source_year,
    target_year=target_year,
    source_geo="bgp",
    target_geo="trt",
    base_source_table=subset_data_dir+"/2000_block.csv.zip",
    input_var=input_vars,
    weight_var=input_var_tags,
    stfips=subset_state,
    keep_base=True,
    add_geoid=True
)
bgp2000_to_trt2010.xwalk


# ### Write crosswalk to a `.csv`

# In[11]:


state_dir = "../../crosswalks/nhgis_bgp2000_trt2010_state/"
nhgisxwalk.xwalk_df_to_csv(
    cls=bgp2000_to_trt2010,
    path=state_dir
)


# ### Read crosswalk from a `.csv`

# In[12]:


fname = state_dir + bgp2000_to_trt2010.xwalk_name
bgp2000_to_trt2010_df = nhgisxwalk.xwalk_df_from_csv(fname)
bgp2000_to_trt2010_df


# -----------------------------------------------