steven-shi

    public static double getShannonEntropy_Image(BufferedImage actualImage){
        List<String> values= new ArrayList<String>();
        int n = 0;
        Map<Integer, Integer> occ = new HashMap<>();
        for(int i=0;i<actualImage.getHeight();i++){
            for(int j=0;j<actualImage.getWidth();j++){
                int pixel = actualImage.getRGB(j, i);
                int alpha = (pixel >> 24) & 0xff;
                int red = (pixel >> 16) & 0xff;

steven-shi

/*
 * Available context bindings:
 *   COLUMNS     List<DataColumn>
 *   ROWS        Iterable<DataRow>
 *   OUT         { append() }
 *   FORMATTER   { format(row, col); formatValue(Object, col) }
 *   TRANSPOSED  Boolean
 * plus ALL_COLUMNS, TABLE, DIALECT
 *
 * where:

steven-shi

SET QSQL_CACHE_CSV="/qfs/tmp/user/xshi/domain.txt";

SELECT bidrequestadsize, bidurl, cache_map_get(key)  into workdir.url_all_size from adtrain.last_2_days WHERE label = 'Impression.QC' AND bidurl LIKE 'http%' and getDomain(bidurl) not LIKE '%mail%' 
and CHAR_LENGTH(bidurl) < 1000 and
bidurl like cache_map_keyset();

SELECT
  bidrequestadsize,
  bidurl,
  count(1) cnt

steven-shi

import psycopg2
f=open('/Users/xshi/Downloads/domain.txt')
s=[l.strip() for l in f.readlines()]
f.close()

conn = psycopg2.connect("dbname='creativeapp' user='creativeapp' host='creative-db-staging.advertise.quantcast.com' password='creativeapp'")
cur = conn.cursor()

def domain_exists(domain):
    cur.execute("SELECT id FROM cpat.domain WHERE domain = %s", (domain,))

steven-shi

from xml.dom.minidom import parse,Document
import sys
import codecs

def escape_xml(file, lang_tag):
	xml = parse(file)
	doc = Document()
	elements=xml.getElementsByTagName(lang_tag)
	print len(elements)
	beg_tag = "<%s>"%lang_tag