A python script for scraping the TOC from a tex file with color values gathered from metadata hidden in latex comments.

prelim_status.py

import os, subprocess
from pyPdf import PdfFileReader
import traceback


status_colors = {"done":"green",
                 "in progress":"darkorange",
                 "pending" : "red",
                 "no status": "black"}

class html_generator:
    def __init__(self):
        pass

    def open_tag(self,tag_type, attribs=None):
        assert(type(tag_type) is str)
        attribs = "" if attribs is None else self.attribs_to_str(attribs)
        return "<"+tag_type+attribs+">"

    def close_tag(self,tag_type):
        assert(type(tag_type) is str)
        val = "</"+tag_type+">"
        val += "" if tag_type == "span" else "\n"
        return val

    def attribs_to_str(self,attribs):
        assert(type(attribs) is dict)
        val = ""
        for key,value in attribs.items():
            val+=" "+str(key)+"="+str(value)
        return val

    def paragraph(self,contents,attribs=None):
        return self.open_tag("p",attribs)+contents+self.close_tag("p")

    def lineitem(self,contents,attribs=None):
        return self.open_tag("li",attribs)+contents+self.close_tag("li")

    def span(self,contents,attribs=None):
        return self.open_tag("span",attribs)+contents+self.close_tag("span")
    
    def unorderedlist(self,contents,attribs=None):
        return self.open_tag("ul",attribs)+"\n"+contents+self.close_tag("ul")+"\n"
    
    def color_attrib(self,color):
        return {'style' : "color:"+color+";"}
    
def generate_html():

    try:
        subprocess.call(["bash","update_prelim.sh"])
    except:
        traceback.print_exc()
        print "Could not update prelim doc."

    for file in os.listdir("./"):
        if ".md" in file:
            f = open(file.split(".md")[0]+".html", 'wb')
            subprocess.call(["markdown",file], stdout=f)
            f.close()

    prelim_status_report()
    
def status_header(f):
    h = html_generator()
    header = h.paragraph("Prelim Table of Contents & Status Report")
    header += h.paragraph(h.span("Green",attribs=h.color_attrib("green"))+" - complete (for now)")
    header += h.paragraph(h.span("Orange",attribs=h.color_attrib("darkorange"))+" - in progress")
    header += h.paragraph(h.span("Red",attribs=h.color_attrib("red"))+" - haven't begun")
    f.write(header)
    return
        
def status_footer(f):
        #special output
    try:
        pages = PdfFileReader(open('prelim.pdf','rb')).getNumPages()
        f.write("<p>Number of Pages: " + str(pages)+"</p>")
    except:
        traceback.print_exc()
        f.write("<p>Number of Pages: Could not read pdf file </p>")

def html_from_tex():
    f = open("prelim.tex",'rb')
    lines = f.readlines()
    f.close()
    lines.append(None)
    return gen_html_toc(lines,"section")


def parse_line(l):
    #title should be inside the brackets
    title = l[l.find("{")+1:l.find("}")]
    #if there's no status
    if 'Status' not in l:
        #indicate no status found
        status = "no status"
    #if there is a status
    else:
        #format %%Status:<status_val>%%
        #parse that status value out
        status = l[l.find("%%")+2:]
        status = status[:status.find("%%")]
        status = status.split(":")[-1]
    #return the title of this item and status
    return title, status

def gen_html_toc(lines,search_key,return_key=None):
    #list of hierarchical keys in document
    search_keys = ["chapter","section","subsection","subsubsection"]
    assert(search_key in search_keys)
    
    #set downward recursion key w/ extra characters
    down_key = None if search_key == search_keys[-1] else "\\"+search_keys[search_keys.index(search_key)+1]+"{"
    #set search key w/ extra characters
    search_key = "\\"+search_key+"{"

    #for tracking first lineitem occurance
    first = True

    #create an html generator
    h = html_generator()
    html_out = ""
    #open unorderedlist
    html_out += "\n"+h.open_tag("ul")+"\n"
    while True:
        line = lines.pop(0)

        # if we find the down key in this line
        if down_key is not None and line is not None and down_key in line:
            #put this line back in so we don't skip something
            lines.insert(0,line)
            # make recursive call w/ down key and search key as return key            
            out,line = gen_html_toc(lines,down_key[1:-1],return_key=search_key)
            #add output to current html
            html_out+=out
                    
        #if we find our return key or hit the end of the file
        if line is None or (return_key is not None and return_key in line):
            # a) close the previous lineitem and current list            
            html_out += h.close_tag("li")            
            html_out += h.close_tag("ul")
            # b) return the html and current line            
            return html_out, line
        #if we find the key we're currently looking for
        elif search_key in line:
            #if this is the first item found
            if first:
                #we're now no longer on our first item
                first = not first
            #otherwise close the previous line item at this level
            else:
                html_out += h.close_tag("li")
                
            #parse title and status from the new item
            title,status = parse_line(line)
            #create a new lineitem w/ title and status color
            html_out += h.open_tag("li")+h.span(title,attribs=h.color_attrib(status_colors[status.lower()]))
        
def prelim_status_report():
    print "Opening file for writing..."
    report_file = open("prelim_status.html",'w')
    print "Generating html and writing to file..."
    status_header(report_file)
    print "Done"
    html = html_from_tex()
    report_file.write(html[0])
    #write num pages to html
    status_footer(report_file)
    report_file.close()

if __name__=="__main__":
    prelim_status_report()