#!/usr/bin/env python3
# -*- coding: utf-8 -*-

# generates keywords.html
# (c) 2013 by the GRASS Development Team, Luca Delucchi

import os
import sys
import glob
from build_html import *

blacklist = ['Display', 'Database', 'General', 'Imagery', 'Misc', 'Postscript',
             'Raster', 'Raster3D', 'Temporal', 'Vector']

path = sys.argv[1]
year = os.getenv("VERSION_DATE")

keywords = {}

htmlfiles = glob.glob1(path, '*.html')

char_list = {}

for fname in htmlfiles:
    fil = open(os.path.join(path, fname))
    # TODO maybe move to Python re (regex)
    lines = fil.readlines()
    # remove empty lines
    lines = [x for x in lines if x != '\n']
    try:
        index_keys = lines.index('<h2>KEYWORDS</h2>\n') + 1
        index_desc = lines.index('<h2>NAME</h2>\n') + 1
    except:
        continue
    try:
        keys = lines[index_keys].split(',')
    except:
        continue
    for key in keys:
        key = key.strip()
        try:
            key = key.split('>')[1].split('<')[0]
        except:
            pass
        if not key:
            exit("Empty keyword from file %s line: %s"
                 % (fname, lines[index_keys]))
        if key not in keywords.keys():
            keywords[key] = []
            keywords[key].append(fname)
        elif fname not in keywords[key]:
            keywords[key].append(fname)

for black in blacklist:
    try:
        del keywords[black]
    except:
        try:
            del keywords[black.lower()]
        except:
            continue

for key in sorted(keywords.keys()):
    # this list it is useful to create the TOC using only the first
    # character for keyword
    firstchar = key[0].lower()
    if firstchar not in char_list.keys():
        char_list[str(firstchar)] = key
    elif firstchar in char_list.keys():
        if key.lower() < char_list[str(firstchar)].lower():
            char_list[str(firstchar.lower())] = key

keywordsfile = open(os.path.join(path, 'keywords.html'), 'w')
keywordsfile.write(header1_tmpl.substitute(title="GRASS GIS %s Reference "
                                           "Manual: Keywords index" % grass_version))
keywordsfile.write(headerkeywords_tmpl)
keywordsfile.write('<dl>')

sortedKeys = sorted(keywords.keys(), key=lambda s: s.lower())

for key in sortedKeys:
    keyword_line = '<dt><b><a name="%s" class="urlblack">%s</a></b></dt>' \
                   '<dd>' % (key, key)
    for value in sorted(keywords[key]):
        keyword_line += ' <a href="%s">%s</a>,' % (value,
                                                   value.replace('.html', ''))
    keyword_line = keyword_line.rstrip(',')
    keyword_line += '</dd>\n'
    keywordsfile.write(keyword_line)
keywordsfile.write("</dl>\n")
# create toc
toc = '<div class="toc">\n<h4 class="toc">Table of contents</h4><p class="toc">'
test_length = 0
all_keys = len(char_list.keys())
for k in sorted(char_list.keys()):
    test_length += 1
#    toc += '<li><a href="#%s" class="toc">%s</a></li>' % (char_list[k], k)
    if test_length % 4 == 0 and not test_length == all_keys:
        toc += '\n<a href="#%s" class="toc">%s</a>, ' % (char_list[k], k)
    elif test_length % 4 == 0 and test_length == all_keys:
        toc += '\n<a href="#%s" class="toc">%s</a>' % (char_list[k], k)
    elif test_length == all_keys:
        toc += '<a href="#%s" class="toc">%s</a>' % (char_list[k], k)
    else:
        toc += '<a href="#%s" class="toc">%s</a>, ' % (char_list[k], k)
toc += '</p></div>\n'
keywordsfile.write(toc)

write_html_footer(keywordsfile, "index.html", year)
keywordsfile.close()