'''
DVC SCRIPT

Brett Paufler
Copyright (c) 2021-04-23/24

A Python Script,
    which transforms dvc_transcript.txt
    into HTML Formatted Sections
    for use in a BrettRant

This is a one off script.

'''

#I have my preferred workflow
FILE_IN = "./input/dvc_transcript.txt"
FILE_OUT = "./output/dvc_formatted.txt"

#The input is read and saved to a working variable
with open(FILE_IN, 'r') as f:
    raw_text = f.read()
    #print raw_text

#Monolithic Text Split into Lines  
lines = raw_text.split('\n')
#print lines #Endless Prints
#print len(lines) #I check as I go

#Trims Off The Front Matter, leaving CSV Data
#I almost always get the number by Plugging & Chugging
lines = lines[24:]
#print lines #Leading & Ending Values are Correct
#print len(lines) #64 is the Correct Number of Courses

#Normally, this would go at the top
#But this is where I added it, so here it goes
from collections import namedtuple

#These are the Final Fields I want
#They differ from the CSV Fields
Course = namedtuple('Course',
         ['catalog', 'section', 'name',
         'grade', 'credit', 'term'])
#print Course

#Test Case without a Section Number
test_course_ns = Course(
    'ECON-220', '', 'Principles of MacRoeconomics',
    'A','3','1992SP')
#print test_course_ns

#Test Case with a Section Number
test_course = Course(
    'HSCI-126', '8931', 'Stress Management and Health',
    'A','3','2006SP')
#print test_course

#I'm going to Loop through the Lines
#And transform each line into a Course
#This is where I will store the Courses
course_list = []

#The Main Data Munging Loop
for line in lines: #What could be simpler?
    #print line #let's see what we are working with
    csv_values = line.split(',')
    #print csv_values
    
    #The End Values are Easier to Deal With
    #So, I'll work from the back-end, popping values
    
    #It seemed like a silly check at the time
    #But I have an Error in "Magic, Witchcraft, & Religion"
    #assert len(csv_values) == 4

    #This Handles My One Length Exception
    if len(csv_values) != 4:
        #print csv_values
        start_string = ','.join(csv_values[:3]) #Combine Three as One
        start_string = start_string.replace('"', '') #Kill the added "'s
        csv_values = [start_string] + csv_values[3:]
        #print csv_values
    assert len(csv_values) == 4

    #Now, where was I?
    #print csv_values
    #It All Looks Good
    
    #There are easier ways to do this
    #But they require more thought
    term = csv_values.pop()
    credit = csv_values.pop()
    grade = csv_values.pop()
    #print grade, credit, term
    
    #All that remains in csv_values is the start string
    assert len(csv_values) == 1
    #Yep, I was right
    
    #I need to pull catalog, section, & name
    cat_string = csv_values.pop()
    assert len(csv_values) == 0 #Can't Hurt
    string_pieces = cat_string.split()
    catalog = string_pieces.pop(0) #easiest first
    
    #If at first you don't succeed
    #The exception code will run
    try:
        section = str(int(string_pieces[0])) #If Error (i.e. no Int Conversion)
        string_pieces = string_pieces[1:]    #Then this never fires
    except:
        section = ''
    
    #Now, we just rejoin the pieces
    name = ' '.join(string_pieces)
    #print catalog, section, name
    
    #It All Looks Good, So Let's Build a Named_Tupel
    this_course = Course(
        catalog, section, name,
         grade, credit, term)
    #print this_course #Looks Good (Well, it does now)
    course_list.append(this_course)
    
#Credits is a Keyword, listing Python Credits
#I'd wanted to use that in my Tuple
#But nope, not going to happen

#print len(course_list) #64, which is correct

# # #
#The Munging Done (Data to Types)
#I will break here
#Output is next
#But it's time for a break
# # #

#Starting This Section
#All Other Prints Commented Out
#print len(course_list)
#print course_list[0]
#All Types are Strings

#Let's see if this works
test_first = "Carry S Through: %s %%s" % "First"
test_second = test_first % "Second"
#print test_first
#print test_second
#So, that Works
#Normally, I'd do that in a "Workspace"
#Or Delete it for clarity
#But this file is about My Workflow

#Let's Build Up a Wall of Text
text = ""
for c in course_list:
    text += '\n\n<br>\n<br>\n<div class="course_info">\n'
    text += c.catalog
    
    if c.section:
        text += " (%s)" % c.section
    text += "<br>\n"
    
    #This is Confusing, But it's done
    #And for a Script, that is good enough
    text += c.term
    if len(c.credit) == 1:
        text += ": %s.0-%s" % (c.credit, c.grade)
    else:
        text += ": %s-%s" % (c.credit, c.grade)
    
    text += "<br>\n"
    text += c.name
    text += "<br>\n</div>\n<br>\n"
    text += '<div class="course_comments">\n<br>\n</div>\n'
    
    #These Prints Will Carry Down
    #They Started at the top of the Loop
    #And I built the Loop on Top of them
    #I run the Program at every additional change
    print text 
    print course_list[0]
    
#We have the desired text

#Output as Displayed
with open("./output/transcript_text.txt", "w") as f:
    f.write(text)

#But I also want to check my Output Right Now
html = "<html><body>%s</body><html>" % text
test_text = 50 * "Test Text! "
#print test_text #Yep, it works
#Once again, normally IO would delete
div_now = '<div class="course_comments">\n'
div_then = '<div class="course_comments">\n%s' % test_text
html = html.replace(div_now, div_then)
#print html #And now, the html has placeholding gibberish text

#Is This Formatted Correctly?  Let's see
with open("./output/transcript_html.html", "w") as f:
    f.write(html)
    
"""
And that's that.
What isn't shown is where and when I commented prints
And the extent to which the final text was reordered
Once I outputted the HTML, I used that as my goby
    And rejiggered the output
    
HSCI-126 (8931): 3.0-A (2006SP)
Stress Management and Health

or something like the above, became

HSCI-126 (8931)
2006SP: 3.0-A
Stress Management and Health

I will apply css styling to achieve the desired result, likely
font-weight: bold;

The timing is roughly an hour a day for two days

I will copy/paste transcript_text into Rant_Blank
And start to fill in the course_comments

I doubt I saved any time from doing the formatting manually
Two Hours is a lot of time
But the time was way more enjoyably spent

And for the most,
    All time spent programming
    Leads to better programming
    
Also, This Script has become Content
Yippie!
"""

"""
And here starts the endless comment strings or none at all.

Originally, I'd thought of solving this problem
by rearranging Cells in Excell (or more accurately, OpenCalc)
And then, using NPP (N++, and/or NotePad++, they all being the same to rearrange rows.
If I knew what I wanted from the start, that route would have been faster
But at the end, when I made a change, I lost little work.
I had my Python Object and formmating output went quick.

I guess I'm trying to justify those two hours.
It seems like a fair clip of time considering the output.
"""