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

# # Identifying 'odd' characters for feature 'after' (N1904LFT)

# ## Table of content <a class="anchor" id="TOC"></a>
# * <a href="#bullet1">1 - Introduction]</a>
# * <a href="#bullet2">2 - Load Text-Fabric app and data</a>
# * <a href="#bullet3">3 - Performing the queries</a>
#     * <a href="#bullet3x1">3.1 - Showing the issue</a>
#     * <a href="#bullet3x2">3.2 - Setting up a query to find them</a>
#     * <a href="#bullet3x3">3.3 - Explanation of the regular expression</a>
#     * <a href="#bullet3x4">3.4 - Bug</a>

# # 1 - Introduction <a class="anchor" id="bullet1"></a>
# ##### [Back to TOC](#TOC)
# 
# This Jupyter Notebook investigates the pressense of 'odd' values for feature 'after'. 

# # 2 - Load Text-Fabric app and data <a class="anchor" id="bullet2"></a>
# ##### [Back to TOC](#TOC)

# In[2]:


get_ipython().run_line_magic('load_ext', 'autoreload')
get_ipython().run_line_magic('autoreload', '2')


# In[3]:


# Loading the New Testament TextFabric code
# Note: it is assumed Text-Fabric is installed in your environment.

from tf.fabric import Fabric
from tf.app import use


# In[4]:


# load the app and data
N1904 = use ("tonyjurg/Nestle1904LFT:latest", hoist=globals())


# # 3 - Performing the queries <a class="anchor" id="bullet3"></a>
# ##### [Back to TOC](#TOC)

# ## 3.1 - Showing the issue <a class="anchor" id="bullet3x1"></a>
# ##### [Back to TOC](#TOC)

# The following shows the pressence of a few 'odd' cases for feature 'after':

# In[31]:


result = F.after.freqList()
print ('frequency: {0}'.format(result))


# ## 3.2 - Setting up a query to find them <a class="anchor" id="bullet3x2"></a>
# ##### [Back to TOC](#TOC)

# In[51]:


# Library to format table
from tabulate import tabulate

# The actual query
SearchOddAfters = '''
word after~^(?!([\s\.·—,;]))
    '''
OddAfterList = N1904.search(SearchOddAfters)

# Postprocess the query results
Results=[]
for tuple in OddAfterList:
    node=tuple[0]
    location="{} {}:{}".format(F.book.v(node),F.chapter.v(node),F.verse.v(node))
    result=(location,F.word.v(node),F.after.v(node))
    Results.append(result)
      
# Produce the table
headers = ["location","word","after"]
print(tabulate(Results, headers=headers, tablefmt='fancy_grid'))


# ## 3.3 - Explanation of the regular expression <a class="anchor" id="bullet3x3"></a>
# ##### [Back to TOC](#TOC)
# 
# The regular expression broken down in its components:
# 
# `^`: This symbol is called a caret and represents the start of a string. It ensures that the following pattern is applied at the beginning of the string.
# 
# `(?!...)`: This is a negative lookahead assertion. It checks if the pattern inside the parentheses does not match at the current position.
# 
# `[…]`: This denotes a character class, which matches any single character that is within the brackets.
# 
# `[\s\.·,—,;]`: This character class contains multiple characters enclosed in the brackets. Let's break down the characters within it:
# 
# * `\s`: This is a shorthand character class that matches any whitespace character, including spaces, tabs, and newlines.
# * `\.`: This matches a literal period (dot).
# * `·`: This matches a specific Unicode character, which is a middle dot.
# * `—`: This matches an em dash character.
# * `,`: This matches a comma.
# * `;`: This matches a semicolon.
# In summary, the character class `[\s\.·,—,;]` matches any single character that is either a whitespace character, a period, a middle dot, an em dash, a comma, or a semicolon.
# 
# The regular expression selects any string which does not starts with a whitespace character, period, middle dot, em dash, comma, or semicolon.

# The following site can be used to build and verify a regular expression: [regex101.com](https://regex101.com/) (choose the 'Pyton flavor') 

# ## 3.4 - Bug <a class="anchor" id="bullet3x4"></a>
# ##### [Back to TOC](#TOC)
# 
# The observed behaviour was due to a bug. [Issue tracker #76](https://github.com/Clear-Bible/macula-greek/issues/76) was opened. When the text of a node starts with punctuation, the @after attribute contains the last character of the word. This is a bug in the transformation to XML LowFat Tree data.

# In[ ]: