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

# # Stock exchange – visualising page data

# In[1]:


import altair as alt
import pandas as pd


# In[2]:


df = pd.read_csv("complete_page_list.csv", parse_dates=["date"])


# In[3]:


df["session"].value_counts()


# In[4]:


df["year"] = df["date"].dt.year


# In[5]:


df_years = df["year"].value_counts().to_frame().reset_index()
df_years.columns = ["year", "count"]


# In[6]:


# Pages per year

alt.Chart(df_years).mark_bar().encode(x="year:N", y="count:Q")


# In[7]:


df_dates = pd.DataFrame(df["date"].unique())
df_dates["year"] = df_dates[0].dt.year
df_years_dates = df_dates["year"].value_counts().to_frame().reset_index()
df_years_dates.columns = ["year", "count"]


# In[8]:


# Days per year

alt.Chart(df_years_dates).mark_bar().encode(x="year:N", y="count:Q")


# In[9]:


orders = {"M": 1, "N": 2, "A": 3, "U": 4}
df_sessions = (
    df.groupby(by="year")["session"]
    .value_counts()
    .to_frame()
    .rename(columns={"session": "count"}, level=0)
    .reset_index()
)
df_sessions["order"] = df_sessions["session"].apply(lambda x: orders[x])


# In[10]:


# Days per year

alt.Chart(df_sessions).mark_bar().encode(
    x="year:N",
    y="count:Q",
    color="session:N",
    order=alt.Order(
        # Sort the segments of the bars by this field
        "order",
        sort="ascending",
    ),
)


# ---
# Created by [Tim Sherratt](https://timsherratt.org) for the [GLAM Workbench](https://glam-workbench.github.io/).