#!/usr/bin/python
#
# Thrown together
# Currently only rescrapes/parses missing UK and NI from preivous day

import datetime, os, commands

base = '/home/parlparse/'
base_html = base + 'parldata/cmpages/'
base_xml = base + 'parldata/scrapedxml/'
base_scraper = base + 'parlparse/pyscraper/'

min = { 'debates': 'debates/debates', 'wms': 'wms/ministerial', 'wrans': 'wrans/answers', 'lords': 'lordspages/daylord' }
dirs = [ min.copy() for x in range(5) ]
for x in (0,1):
    dirs[x]['ni'] = 'ni/ni'
for x in (1,2,3):
    dirs[x]['westminhall'] = 'westminhall/westminster'

suffixes = { 'ni': '' }

def run_command(dir, cmd):
    #print 'Running', cmd
    if dir: print 'cd %s%s' % (base_scraper, dir)
    #os.chdir(base_scraper + dir)
    print cmd
    #status, output = commands.getstatusoutput(cmd)
    #print 'Exit status =', status
    #print 'Output', output

today = datetime.date.today()
wday = today.weekday()

if wday == 0: # Monday
    dirs = dirs[4]
    yesterday = (today - datetime.timedelta(days=3)).isoformat()
elif wday >= 1 and wday <= 5: # Tuesday to Saturday
    dirs = dirs[wday-1]
    yesterday = (today - datetime.timedelta(days=1)).isoformat()

rescrape = []
for type, dir in dirs.items():
    suffix = suffixes[type] if type in suffixes else 'a'
    file = base_html + dir + yesterday + suffix + '.html'
    if not os.path.exists(file):
        rescrape.append(type)

reparse = []
for type, dir in dirs.items():
    suffix = suffixes[type] if type in suffixes else 'a'
    file = base_xml + dir + yesterday + suffix + '.xml'
    if not os.path.exists(file):
        reparse.append(type)

# Scrape
if rescrape:
    run_command('', 'lazyrunall.py --date=' + yesterday + ' scrape ' + ' '.join(rescrape))

# Parse
if reparse:
    run_command('', 'lazyrunall.py --date=' + yesterday + ' parse ' + ' '.join(reparse) + ' --patchtool')
run_command('standing/', './parse.py --patchtool')

print '# Parse/fix any older things from the cron email now'

run_command('', 'other-sites-update 0')