def list_parser(task):
rule = task["rule"]
content = task["recv"].getvalue()
ret = []
dyn_items = re.findall('({.*?sell_price.*?}),', content)
for i in dyn_items:
try:
item = json.loads(i)
except ValueError:
continue
link = item_base % item["id"]
price = item["sell_price"]
ret.append((link, price, 1))
t = etree.HTML(content)
nodes = t.xpath(rule["rule"])
for node in nodes:
link = node.xpath("div[@class = 'cat-item-pic']/a/@href")
price = node.xpath(
"figcaption[@class = 'cat-item-inf']/p/span[@class = 'cat-pire-nub']/text()"
)
if not link or not price:
log_with_time("rule error: %s" % task["url"])
continue
ret.append((link[0], price[0], 1))
result = format_price(ret)
return result
def list_parser(task, rule):
t = etree.HTML(task["text"])
nodes = t.xpath(rule["node"])
if not nodes:
log_with_time("node rule error: %s" % task["url"])
return
dp = []
dps = {}
ret = []
now = int(time.time())
for node in nodes:
link = node.xpath(rule["link"])
gid = node.xpath(rule["gid"])
if not link or not gid:
log_with_time("rule error: %s" % task["url"])
continue
gid = gid[0]
dp.append((link[0], ""))
ret.append(gid)
dps[gid] = now
return {
"dps_log": dps,
"dp": dp,
"price": ret,
}
def stock1_parser(task, rule):
try:
j = demjson.decode(task['text'])
except:
log_with_time("bad response: %r"%task['url'])
return []
code = j['code']
message = j['message']
url = ""
ret = {"spider":[], 'stock2':[]}
if code == 3 and message:
try:
skuid = re.search("\d+", message).group()
url = surl2(task['gid'], skuid)
except:
return []
if url == "":
#print(task['text'])
stock = 1 if j.get('totalAmount') else 0
ret['spider'] = format_price([(itemurl+task['gid'], task['price'], stock)])
else:
ret['stock2'] = [(url, task['gid'], task['price'])]
return ret
def pager(task):
rule = task["rule"]
content = task["recv"].getvalue()
t = etree.HTML(content)
total = t.xpath(rule["rule"])
if "找到0个商品" in content:
log_with_time("search result 0: %s" % task["url"])
return
if not total:
log_with_time("rule error: %s" % rule["rule"])
return
total = int(total[0])
page = rule["page"]
num = total / page
if total % page:
num += 1
tasks = []
for i in range(num):
tasks.append({
"url":
task["url"] + "&order=normal&sort=desc&per_page=%s" % (i * page),
"old_url":
task["url"],
})
return tasks
def rt_parser(items):
pids = get_pids(items)
if not pids:
log_with_time("got nothing: %s" % entries)
return
purl = price_url % (",".join(["J_" + i for i in pids]),
random.randint(1000000, 10000000), int(time.time() * 1000))
surl = stock_url % (async_http.quote(",".join([i for i in pids])),
random.randint(1000000, 10000000), int(time.time() * 1000))
price_res = simple_http.get(purl)
stock_res = simple_http.get(surl)
if price_res["status"] != 200 or stock_res["status"] != 200:
log_with_time("not200: %s" % price["res"])
return
try:
price_json = jsonp_json(price_res["text"])
stock_json = jsonp_json(stock_res["text"].decode("gbk"))
except:
traceback.print_exc()
return
prices = {}
for i in price_json:
prices[i["id"].split("_")[1]] = i["p"]
stocks = {}
for k,v in stock_json.items():
s = v["StockStateName"]
if u"有货" in s or u"现货" in s:
stocks[k] = 1
else:
stocks[k] = 0
ret = []
for pid in prices:
ret.append((str(pid), str(prices[pid]), stocks[pid]))
return format_price(ret)
def cats_parser(url, res, rule):
content = res['text']
t = etree.HTML(content)
ret = set()
items = t.xpath(rule)
for v in items:
#pdb.set_trace()
if '/c0-0/' in v:
continue
if '/ctg/s2/' in v:
r = "(?<=/ctg/s2/).+"
cat = re.search(r, v)
if not cat:
log_with_time("bad regex: %r %r" % (r, v))
continue
cat = cat.group().split('-')[0]
ret.add(ctgurl % cat)
elif 'list.yhd.com' in v:
# http://list.yhd.com/.../
r = "(?<=yhd\.com\/).+"
cat = re.search(r, v)
if not cat:
log_with_time("bad regex: %r %r" % (r, v))
continue
cat = cat.group().split('-')[0]
ret.add(lsturl % cat)
return ret
def list_parser(task, rule):
#pr.enable()
try:
t = etree.HTML(task['text'].decode('gbk', 'replace'))
except:
log_with_time("bad response %s"%task['url'])
return
ret = []
if ebookurl in task['url']:
r = __book_list_parser1(t, task, rule)
elif 'cp' in task['url']:
r = __book_list_parser2(t, task, rule)
else:
r = __norm_list_parser1(t, task, rule)
ret, comments, shop, dp = r
#pr.disable()
#pr.print_stats()
fret = format_price(ret)
dps = {}
for i in fret:
dps[i[1]] = int(time.time())
return {
"result": fret,
"dps": dps,
"shop": shop,
"comment": comments,
"dp": dp,
}
def dp_parser(task, rule):
desc_url = re.findall("desc: '(http.*?desc/[0-9]+)'", task["text"])
if not desc_url:
log_with_time("no desc: %s" % task["url"])
return
crc = urlcrc.get_urlcrc(3, task["url"])
return [(desc_url[0], str(crc), "")]
def pager(task, rule):
try:
tree = etree.HTML(task["text"])
except:
log_with_time("bad response %s" % task['url'])
return
page = re.findall("/([0-9]+)", " ".join(tree.xpath(rule)))
if not page:
log_with_time("page rule error")
return
cat = re.findall("/(v?c[0-9]+-[0-9]+-[0-9]+)", task["url"])[0]
ret = []
for i in range(1, int(page[0]) + 1):
if cat.startswith("vc"):
url_a = virta_base.format(cat=cat,
page=i,
cb=int(time.time() * 1000))
url_b = virtb_base.format(cat=cat,
page=i,
cb=int(time.time() * 1000))
else:
url_a = parta_base.format(cat=cat,
page=i,
cb=int(time.time() * 1000))
url_b = partb_base.format(cat=cat,
page=i,
cb=int(time.time() * 1000))
ret.append(url_a)
ret.append(url_b)
return ret
def list_parser(task, rule):
t = etree.HTML(task['text'])
nodes = t.xpath(rule['nodes'])
ret = []
dps = {}
for node in nodes:
gid = node.xpath(rule['gid'])
price = node.xpath(rule['price'])
if not gid or not price:
log_with_time("bad response: %r" % task['url'])
continue
gid = re_gid.search(gid[0]).group()
price = re_price.search(price[0].text).group()
ret.append({
"url": surl1,
"gid": gid,
"price": price,
"payload": {
"id": gid,
"type": "0",
"count": "1"
}
})
dps[gid] = time.time()
return {"stock": ret, "dps": dps}
def dp_parser(task, rule):
desc_url = re.findall("desc: '(http.*?desc/[0-9]+)'", task["text"])
if not desc_url:
log_with_time("no desc: %s" % task["url"])
return
crc = urlcrc.get_urlcrc(3, task["url"])
return [(desc_url[0], str(crc), "")]
def list_parser(task, rule):
t = etree.HTML(task['text'])
nodes = t.xpath(rule['nodes'])
prices = []
items = []
dps = {}
#pdb.set_trace()
for node in nodes:
gid = node.attrib['itemid']
buyinfo = node.xpath(rule['buyinfo'])
if not gid:
log_with_time("bad response: %r"%task['url'])
continue
if buyinfo:
buyinfo = buyinfo[0]
buycart = buyinfo.xpath(rule['buycart'])
stock = 1
if not buycart:
if buyinfo.xpath(rule['sellout']) or not node.xpath(rule['comment']):
stock = 0
prices.append((gid, stock))
else:
items.append(gid)
dps[gid] = int(time.time())
return {"prices": prices, "items": items, "dps": dps}
def extract_book(url, tree, rule):
result = []
dps = []
now = int(time.time())
dps_log = {}
nodes = tree.xpath(rule["book_node"])
comments = {}
lid = re.search("\d+", url.split('-')[-1]).group()
for node in nodes:
link_node = node.xpath(rule["book_title"])
stock = node.xpath(rule["book_stock"])
comment = node.xpath(rule["book_comment"])
if not link_node or not stock:
log_with_time("rule error: %s" % url)
continue
link_node = link_node[0]
link = link_node.attrib["href"]
gid = re_gid.search(link).group()
comments[gid] = comment[0]
title = link_node.text
if u"有货" in stock[0]:
s = 1
else:
s = 0
dps_log[gid] = now
dps.append((link, gid, title))
result.append((link, gid, lid, s))
return {
"book_price": result,
#"dp": dps,
"dps_log": dps_log,
"comment": comments
}
def cats_parser(url, res, rule):
content = res['text']
t = etree.HTML(content)
ret = set()
items = t.xpath(rule)
for v in items:
#pdb.set_trace()
if '/c0-0/' in v:
continue
if '/ctg/s2/' in v:
r = "(?<=/ctg/s2/).+"
cat = re.search(r, v)
if not cat:
log_with_time("bad regex: %r %r" % (r, v))
continue
cat = cat.group().split('-')[0]
ret.add(ctgurl % cat)
elif 'list.yhd.com' in v:
# http://list.yhd.com/.../
r = "(?<=yhd\.com\/).+"
cat = re.search(r, v)
if not cat:
log_with_time("bad regex: %r %r" % (r, v))
continue
cat = cat.group().split('-')[0]
ret.add(lsturl % cat)
return ret
def page_parser(task, rule):
try:
t = etree.HTML(task["text"])
except:
traceback.print_exc()
return
tot = t.xpath(rule[0])
total = re.findall("/([0-9]+)", "".join(tot))
if not total:
log_with_time("rule error: %s" % rule)
return
total = int(total[0])
if total == 1:
return [{
"url": task['url'],
}]
bas = t.xpath(rule[1])[0]
base = re.findall("(^.*page=?)\d*", bas)[0]
tasks = []
for i in range(1, total + 1):
tasks.append({
"url": "http://www.likeface.com" + base + str(i),
})
return tasks
def extract_book(url, tree, rule):
result = []
dps = []
now = int(time.time())
dps_log = {}
nodes = tree.xpath(rule["book_node"])
comments = {}
lid = re.search("\d+", url.split('-')[-1]).group()
for node in nodes:
link_node = node.xpath(rule["book_title"])
stock = node.xpath(rule["book_stock"])
comment = node.xpath(rule["book_comment"])
if not link_node or not stock:
log_with_time("rule error: %s" % url)
continue
link_node = link_node[0]
link = link_node.attrib["href"]
gid = re_gid.search(link).group()
comments[gid] = comment[0]
title = link_node.text
if u"有货" in stock[0]:
s = 1
else:
s = 0
dps_log[gid] = now
dps.append((link, gid, title))
result.append((link, gid, lid, s))
return {
"book_price": result,
#"dp": dps,
"dps_log": dps_log,
"comment": comments
}
def list_parser(task, rule):
t = etree.HTML(task['text'])
nodes = t.xpath(rule['nodes'])
ret = []
dps = {}
dp = []
for node in nodes:
gid = node.xpath(rule['gid'])
if not gid:
log_with_time("bad rules: %r" % task['url'])
_gid = re.search("(?<=item/).+", gid[0])
if not _gid:
_gid = re.search("(?<=com/).+", gid[0])
if not id:
log_with_time("bad regex: %r" % task['url'])
continue
gid = _gid.group()
dp.append((dp_base % gid, ""))
ret.append(gid)
return {
"price": [
{
"url": priceurl,
"payload": {
"itemid": ','.join(ret)
}
}],
"dp": dp
}
def fix_url(url):
if "tuan" in url:
log_with_time("skip url: %s" % url)
return
x = re.findall("/([0-9\-]+)\.", url)
if not x:
return
return base + ",".join(x[0].split("-"))
def book_price(task, rule):
try:
j = json.loads(task['text'])
price = j['price'][0]['proPrice'] #if j['price'] else 0
except:
log_with_time("bad response: %s" % task['link'])
return
return format_price([[str(task['qid']), str(price), task['stock']]])
def fix_url(url):
if "tuan" in url:
log_with_time("skip url: %s" % url)
return
x = re.findall("/([0-9\-]+)\.", url)
if not x:
return
return base + ",".join(x[0].split("-"))
def book_price(task, rule):
try:
j = json.loads(task['text'])
price = j['price'][0]['proPrice'] #if j['price'] else 0
except:
log_with_time("bad response: %s" % task['link'])
return
return format_price([[str(task['qid']), str(price), task['stock']]])
def price_parser(task, rule):
try:
price = re.search("(?<=price\:)\d+\.\d+(?=\,)", task['text']).group()
except:
log_with_time("bad response: %r"%task['url'])
return []
ret = [(task['gid'], price, task['stock'])]
fret = format_price(ret)
return fret
def list_parser(task, rule):
j = json.loads(task['text'])
if not 'glist' in j:
log_with_time("bad response %r"%task['url'])
return []
ret = []
for g in j['glist']:
ret.append((surl%g['gcode'], g['gprice']))
return ret
def item_parser(task, rule):
try:
t = etree.HTML(task['text'])
btn = t.xpath(rule)[0]
stock = 0 if btn.attrib.get('disabled') else 1
except:
log_with_time("bad response: %s"%task['url'])
return
return [(task['gid'], stock)]
def list_parser(task, rule):
j = json.loads(task['text'])
if not 'glist' in j:
log_with_time("bad response %r" % task['url'])
return []
ret = []
for g in j['glist']:
ret.append((surl % g['gcode'], g['gprice']))
return ret
def pager(task, rule):
j = json.loads(task['text'])
if not 'gpagecount' in j:
log_with_time("bad response %r"%task['url'])
return []
code = re.search("(?<=code=)\d+(?=&)", task['url']).group()
ret = []
for i in range(1, j['gpagecount']+1):
ret.append(gurl%(code,i))
return ret
def pager(task, rule):
j = json.loads(task['text'])
if not 'gpagecount' in j:
log_with_time("bad response %r" % task['url'])
return []
code = re.search("(?<=code=)\d+(?=&)", task['url']).group()
ret = []
for i in range(1, j['gpagecount'] + 1):
ret.append(gurl % (code, i))
return ret
def price_parser(task, rule):
try:
items = jsonp_json(task["text"])
except ValueError as e:
log_with_time("price_parser: jsonp_json: %s" % task["text"])
return
d = {}
for item in items:
d[item["id"].split("_")[1]] = item["p"]
return [d]
def price_parser(task, rule):
try:
items = jsonp_json(task["text"])
except ValueError as e:
log_with_time("price_parser: jsonp_json: %s" % task["text"])
return
d = {}
for item in items:
d[item["id"].split("_")[1]] = item["p"]
return [d]
def stock_parser(task, rule):
try:
j = json.loads(task['text'])
stock = 1 if j['havestock'] in ("true", "realstock") else 0
except:
log_with_time("bad response %s"%task['url'])
return
ret = [(itemurl % task['info'][0], str(task['info'][1]), stock)]
fret = format_price(ret)
return fret
def cats(url, res, rule):
content = res["text"]
try:
t = etree.HTML(content)
except:
log_with_time("bad response %s" % content.decode("utf-8", "replace"))
return
ret = []
for i in t.xpath(rule):
ret.append(yougou + i)
return ret
def cats(url, res, rule):
content = res['text']
try:
t = etree.HTML(content)
except:
log_with_time("bad response %s"%content.decode('utf-8', 'replace'))
return
ret = []
for i in t.xpath(rule):
ret.append(yougou + i)
return ret
def pager(task, rule):
t = etree.HTML(task["text"])
page = t.xpath(rule)
if not page:
log_with_time("page rule error")
return
ret = [task["url"]]
num = re.findall("\d+", " ".join(page))
cat = re.findall("cateID=(\d+)", task["url"])[0]
for i in range(2, int(num[0]) + 1):
ret.append(page_base.format(cat=cat, page=i))
return ret
def checkoffline(task, rule):
try:
j = json.loads(task['text'])
j = j['items']
except:
log_with_time("bad response %s"%task['url'])
return
ret = []
for k,v in j.items():
if not v['is_found']:
ret.append((str(k), str(-1), -1))
fret = format_price(ret)
return fret
def meizhuang_cats_parser(url, content, rule):
t = etree.HTML(content)
ret = []
for node in t.xpath(rule[0]):
#link
link = node.xpath(rule[1])
#price
price = node.xpath(rule[2])
if not link or not price:
log_with_time("rule error: %s" % url)
ret.append((link[0], price[0], 1))
result = format_price(ret)
return result
def price_parser(task, rule):
price = _tess.recognize(task['text'], _tess.IMAGE_PNG, 32)
try:
price = re.search("\d+\.\d+|\d+", price).group()
except:
log_with_time("bad price: %s" % task['url'])
return
ret = [(task['gid'], price, task['stock'])]
fret = format_price(ret)
dps = {}
for i in fret:
dps[i[1]] = int(time.time())
return {"result":fret, "dps": dps}
def list_parser(task, rule):
t = etree.HTML(task['text'])
nodes = t.xpath(rule['nodes'])
ret = []
for node in nodes:
gid = node.xpath(rule['gid'])
price = node.xpath(rule['price'])
if not gid or not price:
log_with_time("bad response: %r" % task['url'])
continue
gid = re.findall("id-([0-9]+)", gid[0])
ret.append((gid[0], price[0]))
return ret
def list_parser(task, rule):
t = etree.HTML(task["text"])
nodes = t.xpath(rule["node"])
ret = []
for node in nodes:
qid = node.attrib["id"].split("_")
link = node.xpath(rule["link"])
price = node.xpath(rule["price"])
if not link or not price or not qid:
log_with_time("rule error: %s" % task["url"])
continue
ret.append(("http://www.miyabaobei.com" + link[0], qid[1], price[0]))
return ret
def meizhuang_cats_parser(url, content, rule):
t = etree.HTML(content)
ret = []
for node in t.xpath(rule[0]):
#link
link = node.xpath(rule[1])
#price
price = node.xpath(rule[2])
if not link or not price:
log_with_time("rule error: %s" % url)
ret.append((link[0], price[0], 1))
result = format_price(ret)
return result
def pager(task, rule):
c = task["text"].decode("utf-8")
item = jsonp_json(c)
if "pageBar" not in item:
log_with_time("no pageBar: %s" % task["url"])
return
m = item["pageBar"]
ret = []
if not m.get("totalCount", 0):
log_with_time("empty category: %s" % task["url"])
return ret
for i in range(1, m["totalPage"] + 1):
ret.append({"url": payload(task['cat'], i)})
return ret
def stock_parser(task, rule):
t = etree.HTML(task["text"])
stock = t.xpath(rule)
ret = []
if not stock:
log_with_time("bad response: %s" % task["url"])
return ret
if int(stock[0].text):
stock = 1
else:
stock = 0
ret.append((task['url'], task['price'], stock))
fret = format_price(ret)
return fret
def stock_parser(task, rule):
t = etree.HTML(task["text"])
stock = t.xpath(rule)
ret = []
if not stock:
log_with_time("bad response: %s" % task["url"])
return ret
if int(stock[0].text):
stock = 1
else:
stock = 0
ret.append((task['url'], task['price'], stock))
fret = format_price(ret)
return fret
def list_parser(task):
t = etree.HTML(task["recv"].getvalue())
nodes = t.xpath(task["rule"]["rule"])
ret = []
for node in nodes:
link = node.xpath("div/div[@class = 'proTit']/a/@href")
price = node.xpath("div/div[@class = 'proPrice']/text()")
if not link or not price:
log_with_time("rule error: %s" % task["old_url"])
continue
p = fix_price(price[0])
ret.append((link[0], p, 1))
result = format_price(ret)
return result
def list_parser(task, rule):
tree = etree.HTML(task["text"])
nodes = tree.xpath(rule["node"])
ret = []
for node in nodes:
gidurl = node.xpath(rule['gidurl'])
price = node.xpath(rule['price'])
if not gidurl or not price:
log_with_time("list parser err: %r" % task['url'])
continue
gidurl = gidurl[0]
price = price[0].text[1:]
ret.append((gidurl, price))
ret = list(set(ret))
return ret
def list_parser(task, rule):
t = etree.HTML(task['text'])
nodes = t.xpath(rule['nodes'])
ret = []
for node in nodes:
gid = node.xpath(rule['gid'])
price = node.xpath(rule['price'])
if not gid or not price:
log_with_time("bad response: %r" % task['url'])
continue
gid = burl + gid[0]
price = price[0].text
price = re_price.search(price).group()
ret.append((gid, price))
return ret
def list_parser(task, rule):
import pdb
t = etree.HTML(task["text"])
nodes = t.xpath(rule["node"])
ret = []
for node in nodes:
link = node.xpath(rule["link"])
price = node.xpath(rule["price"])
if not link or not price:
log_with_time("rule error: %s" % task["old_url"])
continue
p = fix_price(price[0])
ret.append((str(link[0]), str(p), 1))
result = format_price(ret)
return result
def list_parser(task, rule):
tree = etree.HTML(task["text"])
nodes = tree.xpath(rule["node"])
ret = []
for node in nodes:
gidurl = node.xpath(rule['gidurl'])
price = node.xpath(rule['price'])
if not gidurl or not price:
log_with_time("list parser err: %r" % task['url'])
continue
gidurl = gidurl[0]
price = price[0].text[1:]
ret.append((gidurl, price))
ret = list(set(ret))
return ret
def promo_filter(item):
url, sku = item
parts =re.findall("/([A-Za-z-0-9]+)\.h", url)
if not parts:
log_with_time("url rule error: %s" % url)
pid, sid = parts[0].split("-")
#if "A" in url:
# goodsNo = re.findall("([0-9]+)\.html", url)[0]
#else:
# goodsNo = sku
p = promo_url.format(time = int(time.time() * 1000), goodsNo = sku, sid = sid, pid = pid)
return {
"url": p,
"old": url
}