choozmo
/
kw_tools


			
							123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337
							from pydoc import HTMLDoc
from fastapi import FastAPI
import dataset
import sys
import os
import time
from fastapi.middleware.cors import CORSMiddleware
from fastapi.staticfiles import StaticFiles
from pydantic import BaseModel
from fastapi import FastAPI, Form, Response, File, UploadFile, Request
import subprocess
import suggests
from typing import Optional
import networkx as nx
import pyvis
import time
from pyvis.network import Network
import pickle
import logging
import threading
import random
import string
from fastapi.responses import HTMLResponse,RedirectResponse, FileResponse
import dataset
import traceback
import time
from selenium import webdriver
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.common.by import By
from selenium.webdriver.chrome.service import Service
import networkx as nx
from pyvis.network import Network
import csv
import sys
import codecs
import difflib
import pymysql
pymysql.install_as_MySQLdb()
from pathlib import Path
from tempfile import NamedTemporaryFile
from typing import Callable
import shutil


driver = None

def id_generator(size=6, chars=string.ascii_uppercase + string.digits):
    return ''.join(random.choice(chars) for _ in range(size))


app = FastAPI()
origins = ["*"] 
app.add_middleware(
    CORSMiddleware,
    allow_origins=origins,
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"],
)

db = dataset.connect('mysql://choozmo:pAssw0rd@db.ptt.cx:3306/gtrends?charset=utf8mb4')

app.mount("/web", StaticFiles(directory="/Users/zooeytsai/kw_tools/web/static"), name="static")
# app.mount("/web", StaticFiles(directory="/root/src/kw_tools/web/static"), name="static")

def thread_function(kw):
    global db
    print(kw)
    G = nx.Graph()
    for k in kw:
        s = suggests.suggests.get_suggests(k, source='google')

        for sg in s['suggests']:
            G.add_edge(k,sg,weight=1)
            print(sg)
            time.sleep(1)
            s2 = suggests.suggests.get_suggests(k, source='google')
            for elmt in s2['suggests']:
                G.add_edge(sg,elmt,weight=1)

#    G.remove_nodes_from(list(nx.isolates(G)))
    G.remove_edges_from( list(nx.selfloop_edges(G)))

#    pickle.dump( G, open( "gs2.p", "wb" ) )

    pyG = Network(height="750px", width="100%",bgcolor="#333333",font_color="white")
    pyG.from_nx(G)
    id=id_generator()
    db['gen_graph'].insert({'filename':str(id),'kw':str(kw)})
#    pyG.save_graph('gstest')
#    pyG.show('static/gs/'+str(id)+'.html')
    pyG.save_graph('static/gs/'+str(id)+'.html')

@app.get("/tree_list/",response_class=HTMLResponse)
async def tree_list():
#    global db
    db = dataset.connect('mysql://choozmo:pAssw0rd@db.ptt.cx:3306/gtrends?charset=utf8mb4')
    html="<html><body><h2>清單</h2></br>請一分鐘後refresh </br></br>"
    html+="<table border='1'>"
    cursor=db.query('select filename,kw from gen_graph order by id desc')
    cnt=0
    for c in cursor:
        html+="<tr><td>"+c['kw']+"</td>"
        html+="<td><a href='/web/gs/"+c['filename']+".html'>"+c['filename']+"</a></td></tr>"
        cnt+=1
        if cnt > 10:
            break

    html+="</table></body></html>"
    return html

@app.post("/proj_kw/",response_class=HTMLResponse)
async def proj_kw(proj: str = Form(...),kws:Optional[str] = Form(None)):
    db = dataset.connect('mysql://choozmo:pAssw0rd@db.ptt.cx:3306/seo?charset=utf8mb4')
    table=db['serp_jobs']
    for kw in kws:
        table.insert({'proj':proj,'kw':kw})
    return "OK請稍後"


#response_class=RedirectResponse
@app.post("/gen_tree/",response_class=HTMLResponse)
async def func_expand(kw: str = Form(...),kw2:Optional[str] = Form(None),kw3:Optional[str] = Form(None),kw4:Optional[str] = Form(None) ):
    kwlst=[]
    if len(kw)>1:
        kwlst.append(kw)    
    if kw2 is not None:
        kwlst.append(kw2)    
    if kw3 is not None:
        kwlst.append(kw3)    
    if kw4 is not None:
        kwlst.append(kw4)    

    x = threading.Thread(target=thread_function, args=(kwlst,))
    x.start()
#    return "ok"
    return RedirectResponse(url="/tree_list",status_code=302)
#    return HTMLResponse('<html><head><meta http-equiv="refresh" content="0; URL="/tree_list" /></head></html>')


def restart_browser():
    global driver
    if driver is not None:
        print('closing')
        driver.quit()
        driver = None
    try: 
        options = webdriver.ChromeOptions()
        options.add_argument("--no-sandbox")
        options.add_argument("--disable-dev-shm-usage")
        options.add_argument('--headless')
        #options.add_argument('--remote-debugging-port=9222')
        #options.add_experimental_option("debuggerAddress", "127.0.0.1:9922")
        options.add_argument("--incognito")
        try:
            driver = webdriver.Chrome(options=options)
            str1 = driver.capabilities['chrome']['chromedriverVersion'].split(' ')[0]
            print('這裡',str1)
            #driver = webdriver.Remote(command_executor='http://127.0.0.1:'+str(portnum)+'/wd/hub',options=options)
        except:
            return None
    except:
       print('開啟失敗')
       driver=None
       return None
    return driver
    
@app.post("/ranking/")
async def ranking(kw: str = Form(...), domain:str = Form(...),kw2:Optional[str] = Form(None),domain2:Optional[str] = Form(None),kw3:Optional[str] = Form(None),domain3:Optional[str] = Form(None),kw4:Optional[str] = Form(None),domain4:Optional[str] = Form(None),kw5:Optional[str] = Form(None),domain5:Optional[str] = Form(None)):
    kwlst = []
    kwlst.append([kw,domain])
    if kw2 is not None:
        kwlst.append([kw2,domain2])
    if kw3 is not None:
        kwlst.append([kw3,domain3])
    if kw4 is not None:
        kwlst.append([kw4,domain4])
    if kw5 is not None:
        kwlst.append([kw5,domain5])
    result = []
    for i in kwlst:
        driver = restart_browser()
        # escaped_search_term=urllib.parse.quote(term)
        googleurl = 'https://www.google.com/?num=100'
        driver.get(googleurl)
        time.sleep(6)
        send_kw_elmt = driver.find_element(By.XPATH,
                                           '/html/body/div[1]/div[3]/form/div[1]/div[1]/div[1]/div/div[2]/input')
        send_kw_elmt.send_keys(i[0])
        time.sleep(3)
        send_kw_elmt.send_keys(Keys.ENTER)
        time.sleep(6)
        elmts = driver.find_elements_by_xpath("//div[@class='yuRUbf']/a")

        cnt = 1
        datadict = {'搜尋詞': [], '結果標題': [], '結果網址': [], '結果名次': []}
        domain_name = i[1]
        for elmt in elmts:
            try:
                href = elmt.get_attribute('href')
                if domain_name in href:
                    datadict['搜尋詞'].append(i[0])
                    datadict['結果標題'].append(elmt.text)
                    datadict['結果網址'].append(href)
                    datadict['結果名次'].append(str(cnt))
                cnt += 1
            except:
                print('href2 exception')
                traceback.print_exc()
        result.append(datadict)
        print(domain_name)
        print(datadict)
        driver.quit()
        print('數量',len(elmts))
        time.sleep(90)

    # return "ok"
    # return RedirectResponse(url="/ranking_result",)
    html = f"<html><body>{result}</body></html>"
    return html

@app.get("/ranking_result/")
async def tree_list():
    html = "<table border='1'>"

    # html += "<tr><td>" + c['kw'] + "</td>"
    
    return html

kwdict={}
G = nx.Graph()
def gcm0(strings):
    clusters = {}
    for string in (x.strip() for x in strings):
        match = difflib.get_close_matches(string, clusters.keys(), 8, 0.65)
        if match:
            clusters[match[0]].append(string)
        else:
            clusters[string] = [ string ]
    return clusters


def proc_row(row):
    elmts=row.split(' ')
    for elmt in elmts:
        if kwdict.get(elmt) is None:
            kwdict[elmt]=1
        else:
            kwdict[elmt]+=1


def save_upload_file_tmp(file: UploadFile) -> Path:
    try:
        suffix = Path(file.filename).suffix
        with NamedTemporaryFile(delete=False, suffix=suffix) as tmp:
            shutil.copyfileobj(file.file, tmp)
            tmp_path = Path(tmp.name)
    finally:
        file.file.close()
    return tmp_path

    
@app.post("/kwtree")
async def kwtree(file: UploadFile = File(...)):
    destination_file_path = "/root/src/kw_tools/web" + file.filename
    with codecs.open(destination_file_path,'r','utf-16') as out_file:
    # with codecs.open(file.file) as csvfile:
        # csv_reader = csv.reader(codecs.iterdecode(file.file, 'utf-8'))
        csvfile = csv.reader(out_file, delimiter='\t', quotechar='|')
        # spamreader = csv.reader(csvfile, delimiter='\t', quotechar='|')
        kwdict = {}
        addict = {}
        head = True
        rowlst = []
        for row in csvfile:
            if head:
                head = False
                continue
            ll = len(row)
            proc_row(row[0])
            if row not in rowlst:
                rowlst.append(row[0])
        head = True

    clusters = gcm0(rowlst)
    keys = []
    for k, v in clusters.items():
        # if len(v) > 20:
        keys.append(k)
        for x in v:
            G.add_edge(k, x, weight=1, label='')

    already_dict = {}
    from strsimpy.qgram import QGram
    qgram = QGram(2)
    for k1 in keys:
        for k2 in keys:
            if k1 != k2:
                if qgram.distance(k1, k2) <= 12:
                    if already_dict.get(k1) is None and already_dict.get(k2) is None:
                        already_dict[k1] = 1
                        already_dict[k2] = 1
                        G.add_edge(k1, k2, weight=1, label='')

    pyG = Network(height="100%", width="100%", bgcolor="#444444", font_color="white")
    pyG.set_options("""
        const options = {
        "nodes" : {
                "font" : {
                    "size" : "30",
                    "color" : "#ffffff"
                }
            },
          "physics": {
            "forceAtlas2Based": {
              "springLength": 100
            },
            "maxVelocity": 150,
            "minVelocity": 0.28,
            "solver": "forceAtlas2Based"
          }
        }
        """)
    G.remove_edges_from(nx.selfloop_edges(G))
    pyG.from_nx(G)

    # pyG.show_buttons(filter_=['physics'])
    news_file = random.randint(0,100)
    pyG.show(f'news{news_file}.html')
    check_file = False
    # while
    # print(clusters)

    # sys.exit()

    return FileResponse(f'/root/kw_tools/web/news{news_file}.html',media_type='text/html')