Das deutsche Python-Forum

Code: Alles auswählen

str="#Polls #slipping, have HER camp on defense/lowering expectations, going into #Debates2016"
str3="#MaryBrigidMcManamon, Washington Post: Constitutionally speaking, #Cruz simply isnt eligible to be president"
laenge = 0
keyword=''
while laenge < len(str):
    if str[laenge]=='#':
        index_hashtag = laenge
        hash=True
        if str[index_hashtag+1]=='#' or str[index_hashtag+1]==' ':
            laenge= index_hashtag+2
            hash=False
    if str[laenge]==' ' or str[laenge]=='.' or str[laenge]==',' or str[laenge]==str[len(str)-1]:
        index_leerZ=laenge
        leerZ=True
        laenge += 1
        if hash==True and leerZ==True:
            keyword=keyword+str[index_hashtag:index_leerZ]
            hash = False
            leerZ = False
            print keyword
    laenge+=1

Code: Alles auswählen

# -*- coding: utf-8 -*-
import csv
with open('new.csv', 'wb') as newcsv:
    csvfile = open("american-election-tweets2.csv", "rb")
    reader = csv.DictReader(csvfile.read().decode('utf-8-sig').encode('utf-8').splitlines(), delimiter=';')
    fieldnames = ['handle', 'text', 'is_retweet', 'original_author', 'time', 'in_reply_to_screen_name',
                    'retweet_count', 'favorite_count', 'hashtag']
    writer = csv.DictWriter(newcsv, fieldnames=fieldnames, delimiter=';')
    writer.writeheader()
    num=0
    for row in reader:
        laenge = 0
        keywords = []
        row['text']=row['text'].strip('\xe2\x80\xa6')
        for laenge in range(len(row['text'])):
            if row['text'][laenge] == '#':
                index_hashtag = laenge
                hash = True
                if row['text'][index_hashtag + 1] == '#' or row['text'][index_hashtag + 1] == ' ':
                    laenge = index_hashtag + 2
                    hash = False
            if row['text'][laenge] == ' ' or row['text'][laenge] == '.' or row['text'][laenge] == ',' \
                    or row['text'][laenge] == row['text'][len(row['text']) - 1] or row['text'][laenge] == '"' \
                    or row['text'][laenge] == '!' or row['text'][laenge] == ':' or row['text'][laenge] == '-' or \
                            row['text'][laenge] == "'":
                if laenge == len(row['text']) - 1: laenge += 1
                leerZ = True
                if hash == True:
                    keywords.append(row['text'][index_hashtag:laenge])
                    hash = False
                    leerZ = False
                    print (row['text'][index_hashtag])
                    print (row['text'][laenge-1])
                    print keywords
        writer.writerow({'handle': row['handle'], 'text': row['text'], 'is_retweet': row['is_retweet'],'original_author': row['original_author'],
                         'time': row['time'],'in_reply_to_screen_name': row['in_reply_to_screen_name'],'retweet_count': row['retweet_count'],
                         'favorite_count': row['favorite_count'],'hashtag': keywords})
        keywords=[]

Code: Alles auswählen

import re

def durchsuche_csv(file):
    regex  = re.compile("#[A-Za-z]+")
    inhalt = open(file, "r")
    suche  = inhalt.read()
    inhalt.close
    return regex.findall(suche)

funde = durchsuche_csv("test.csv")

print("#Ausgabe:\n")

for alle in funde:
    print(alle)

#Ausgabe:

#Polls
#slipping
#Debates
#MaryBrigidMcManamon
#Cruz

Code: Alles auswählen

regex = re.compile("#[a-z]+", re.I)

Code: Alles auswählen

import csv
import re
#Suche nach den #Keyword mit RE
def durchsuche_csv(String):
    regex  = re.compile("#[A-Za-z]+", re.I)
    return regex.findall(String)
# öffnen der Dateien fürs Lesen und Schreiben
with open('new.csv', 'wb') as newcsv:
    csvfile = open("american-election-tweets2.csv", "rb")
    reader = csv.DictReader(csvfile.read().decode('utf-8-sig').encode('utf-8').splitlines(), delimiter=';')
    fieldnames = ['handle', 'text', 'is_retweet', 'original_author', 'time', 'in_reply_to_screen_name',
                    'retweet_count', 'favorite_count', 'hashtag']
    writer = csv.DictWriter(newcsv, fieldnames=fieldnames, delimiter=';')
    writer.writeheader()
#Durchlaufen jede eingelesene Zeile; die Suchfunktion aufrufen und text durchsuchen
    for row in reader:
        keywords= durchsuche_csv(row['text'])
        lowercaseKeywords=[]
        #jedes gefundene #Keyword wird klein geschrieben
        for word in keywords:
            lowercaseKeywords.append(word.lower())
        writer.writerow({'handle': row['handle'], 'text': row['text'], 'is_retweet': row['is_retweet'],
                     'original_author': row['original_author'],
                     'time': row['time'], 'in_reply_to_screen_name': row['in_reply_to_screen_name'],
                     'retweet_count': row['retweet_count'],
                     'favorite_count': row['favorite_count'], 'hashtag': lowercaseKeywords})
        lowercaseKeywords = []

Code: Alles auswählen

import csv
import re

FIELDNAMES = ['handle', 'text', 'is_retweet', 'original_author', 'time', 'in_reply_to_screen_name', 'retweet_count', 'favorite_count', 'hashtag']
find_hashtags =re.compile("#[A-Za-z]+").findall

with open('american-election-tweets2.csv', encoding="utf-8-sig") as csvfile:
    with open('new.csv', 'w', encoding="utf-8") as newcsv:
        reader = csv.DictReader(csvfile, delimiter=';')
        writer = csv.DictWriter(newcsv, fieldnames=FIELDNAMES, delimiter=';')
        writer.writeheader()
        for row in reader:
            row['hashtag'] = list(map(str.lower, find_hashtags(row['text'])))
            writer.writerow(row)

Code: Alles auswählen

    regex  = re.compile("\B#\w+")
    inhalt = open(file, "r", encoding="utf-8")