getsizes.py

import re
import sys
import urllib2

revisions = dict()

urls = [
        'http://iso.reactos.org/livecd/',
        'http://iso.reactos.org/bootcd/',
#       'http://iso.reactos.org/regtestcd/'
]
for url in urls:
        data = urllib2.urlopen(url)
        data = data.read()

        iter = re.finditer('<tr><td class="n"><a href="[^"]*">([^\-]+)-([^\-]+)((?:-(?:[^\-]+))?).(7z|iso)</a></td>.*<td class="s">(.*)M</td>.*</tr>', data)
        foundany = False
        for match in iter:
                foundany = True
                g = match.groups()
                rev = g[1]
                type = g[0] + g[2] + ' (' + g[3] + ')'
                size = g[4]
                if 'patch' in type:
                        continue
                if not rev.isdigit(): #or int(rev) < 59000:
                        continue
                if not rev in revisions:
                        revisions[rev] = dict()
                revisions[rev][type] = size

revs = sorted(revisions.keys())
revList = '[]' #'["' + '","'.join(revs) + '"]'

types = set()
for r in revs:
        for k in revisions[r].keys():
                types.add(k)

typeSeries = []
for t in types:
        serdata = []
        for r in revs:
                if t in revisions[r]:
                        serdata.append('[' + r + ',' + revisions[r][t] + ']')
        series = '{"name":"' + t + '","data":[' + ','.join(serdata) + ']}'
        typeSeries.append(series)

print('{"revs":' + revList + ',')
print('"sizes":[' + ','.join(typeSeries) + ']}')