我的天

'''
***使用说明***
终端界面输入xbus2wp.py bus.xml xrspook。其中：
xbus2wp.py为脚本名字，bus.xml为BlogBus导出文件，xrspook为博主名字，3个参数以空格分开
若运行无误，输出的文件名为[原文件名_xbus2wp.xml]
脚本基于python3，适配WordPress 5.4.2（2020-06-18）
'''
 
import re, sys, getopt, datetime
from xml.dom import minidom
from time import time
 
def convert(inputFileName, owner, order='asc'):
    """"""
    try:
        xmldoc = minidom.parse(inputFileName)
    except Exception as e:
        print ('Fail.')
        print (e)
        print ('Please repair or delete invalid token like "& < >" there.')
        sys.exit(1)
 
    bus = xmldoc.documentElement
    logs = bus.getElementsByTagName('Log')
 
    dom = minidom.Document()
    rss = dom.createElement('rss') # rss是root，根元素
    dom.appendChild(rss)      
    rss.setAttribute('version', '2.0')
    rss.setAttribute('xmlns:content', 'http://purl.org/rss/1.0/modules/content/')
    rss.setAttribute('xmlns:wfw', 'http://wellformedweb.org/CommentAPI/')
    rss.setAttribute('xmlns:dc', 'http://purl.org/dc/elements/1.1/')
    rss.setAttribute('xmlns:wp', 'http://wordpress.org/export/1.0/')
    channel = dom.createElement('channel')
    rss.appendChild(channel)
    wxr_version = dom.createElement('wp:wxr_version') # 加入wxr戳，无戳无法进行WordPress导入
    channel.appendChild(wxr_version)
    wxr_version_node = dom.createTextNode('1.1')
    wxr_version.appendChild(wxr_version_node)
 
    busname = bus.getElementsByTagName('BlogName')[0] # 提取原BlogBus名字
    busname_text = getElementData(busname).replace(' ', '_')
 
    # create a list to contain items instead of appending them to
    # channel directly in order to sort them of lately according to order.
    if order == 'desc':
        item_list = []
    else:
        item_list = None
 
    for log in logs:
        title = log.getElementsByTagName('Title')[0]
        title_text = getElementData(title)
        content = log.getElementsByTagName('Content')[0]
        content_text = getElementData(content)
        logdate = log.getElementsByTagName('LogDate')[0]
        pubdate = getElementData(logdate)
        writer = log.getElementsByTagName('Writer')[0]
        creator = owner # BlogBus的writer根本没包含元素！
        category = getElementData(log.getElementsByTagName('Sort')[0])
        tagi = log.getElementsByTagName('Tags')[0]
        tags = getElementData(tagi).split(' ')
        new_tags = unique_tag(category, tags) # 新的wp标签里包含了原BlogBus里的分类与标签
        comments = log.getElementsByTagName('Comment')
 
        #-----
        item = dom.createElement('item')
 
        # handle title
        title_element = createElement(dom, 'title', title_text, 'cdata')
        item.appendChild(title_element)
 
        # handle type
        type_element = createElement(dom, 'wp:post_type', 'post', 'cdata')
        item.appendChild(type_element)
 
        # handle pubdate
        pubdate_element = createElement(dom, 'pubDate', convertPubDate(pubdate))
        item.appendChild(pubdate_element)
 
        # handle creator
        creator_element = createElement(dom, 'dc:creator', creator, 'cdata')
        item.appendChild(creator_element)
 
        # handle categories with domain
        category_element = createElement(dom, 'category', busname_text, 'cdata') # 把BlogBus标题设置为分类，因为我要合并多个旧blog
        category_element.setAttribute('domain','category')
        category_element.setAttribute('nicename', busname_text)
        item.appendChild(category_element)
 
        # handle tags
        for tag in new_tags:
            tag = tag.replace('&ntilde;', 'n')
            tag = tag.replace('summary_of_BLF', 'summary_of_BLF(from_rincondebetty)')
            tag = tag.replace('summary_of_EcoModa', 'summary_of_EcoModa(from_rincondebetty)')
            category_element = createElement(dom, 'category', tag, 'cdata')
            category_element.setAttribute('domain','post_tag')
            category_element.setAttribute('nicename', tag)
            item.appendChild(category_element)
 
        # handle content
        content_element = createElement(dom, "content:encoded", content_text, 'cdata')        
        item.appendChild(content_element)
 
        # handle post_date
        post_date_element = createElement(dom, "wp:post_date", pubdate)
        item.appendChild(post_date_element)
 
        # handle status
        status_element = createElement(dom, "wp:status", 'publish')
        item.appendChild(status_element)
 
        # handle comments
        if comments:
            commentElements = createComments(dom, comments)
            for commentElement in commentElements:
                item.appendChild(commentElement)
 
        if item_list != None:
            item_list.append(item)
        else:
            channel.appendChild(item)
 
    if item_list:
        item_list.reverse()
        for m in item_list:
            channel.appendChild(m)
 
    global filename # 输出设置
    output = filename + '_xbus2wp.xml'
    f = open(output ,'wb+')
    import codecs
    writer = codecs.lookup('utf-8')[3](f)
    dom.writexml(writer, '', ' ' * 4, '\n', encoding='utf-8')
    writer.close()
 
def unique_tag(category,tags): # 只保留唯一的标签
    category = category.replace(' ', '_')
    l = category.split() + tags
    new_l = []
    for item in l:
        if item not in new_l and item != '(from_rincondebetty)':
            new_l.append(item.replace(' ', '_')) # 替换空格为下划线
    return new_l
 
def getElementData(element): # 获取节点数据
    """"""
    data = ''
    for node in element.childNodes:
        if node.nodeType in (node.TEXT_NODE, node.CDATA_SECTION_NODE):
            data += node.data
    return data
 
def createComments(dom, comments):
    """"""
    l = []
    count = 0
    for comment in comments:
        count += 1 # 每篇文章的评论序号，没有序号，评论只能导入每篇最后一条
        email = comment.getElementsByTagName('Email')[0]
        homepage = comment.getElementsByTagName('HomePage')[0]
        name = comment.getElementsByTagName('NiceName')[0]
        content = comment.getElementsByTagName('CommentText')[0]
        date = comment.getElementsByTagName('CreateTime')[0]
        comment_element = createCommentElement(count, dom, email, homepage, name, content, date)
        l.append(comment_element)
    return l
 
def createCommentElement(count, dom, email, homepage, name, content, date):
    """"""
    comment_author = getElementData(name)
    comment_author_email = getElementData(email)
    comment_author_url = getElementData(homepage)
    comment_date = getElementData(date)
    comment_content = getElementData(content)
 
    comment_id_element = createElement(dom, 'wp:comment_id', str(count))
    comment_author_element = createElement(dom, 'wp:comment_author', comment_author)
    comment_author_email_element = createElement(dom, 'wp:comment_author_email', comment_author_email)
    comment_author_url_element = createElement(dom, 'wp:comment_author_url', comment_author_url)
    comment_date_element = createElement(dom, 'wp:comment_date', comment_date)
    comment_date_gmt_element = createElement(dom, 'wp:comment_date_gmt', comment_date)
    comment_content_element = createElement(dom, 'wp:comment_content', comment_content, 'cdata')
    comment_approved_element = createElement(dom, 'wp:comment_approved', '1')
 
    # make the comment element
    comment_element = dom.createElement('wp:comment')
    comment_element.appendChild(comment_id_element)
    comment_element.appendChild(comment_author_element)
 
    # validate email and url
    validEmail = validateEmail(comment_author_email)
    if (validEmail):
        comment_element.appendChild(comment_author_email_element)
 
    validUrl = validateUrl(comment_author_url)
    if (validUrl):
        comment_element.appendChild(comment_author_url_element)    
 
    comment_element.appendChild(comment_date_element)
    comment_element.appendChild(comment_date_gmt_element)
    comment_element.appendChild(comment_content_element)
    comment_element.appendChild(comment_approved_element)
 
    return comment_element
 
def createElement(dom, elementName, elementValue, type='text'): #建立节点标签和节点
    """"""
    global owner
    tag = dom.createElement(elementName)
    if elementValue.find(']]>') > -1:
        type = 'text'
    if type == 'text':
        text = dom.createTextNode(elementValue)
    elif type == 'cdata':
        elementValue = elementValue.replace('&amp;', '&')
        elementValue = elementValue.replace('&lt;', '<')
        elementValue = elementValue.replace('&gt;', '>')
        elementValue = elementValue.replace('&apos;', '\'')
        elementValue = elementValue.replace('&quot;', '"')
 
        # 大量替换与我的旧blog有各种编码的西班牙语字符有关
        elementValue = elementValue.replace('&copy;', '') # 版权标志
        elementValue = elementValue.replace('&nbsp;', '') # 空格
        elementValue = elementValue.replace('&ldquo;', '“') # 左双引号
        elementValue = elementValue.replace('&rdquo;', '”') # 右双引号
        elementValue = elementValue.replace('&lsquo;', '‘') # 左单引号
        elementValue = elementValue.replace('&rsquo;', '’') # 右单引号
        elementValue = elementValue.replace('&acute;', '´') # 单引号
        elementValue = elementValue.replace('&hellip;', '...') # 省略号
        elementValue = elementValue.replace('&mdash;', '—') # 破折号
        elementValue = elementValue.replace('&middot;', '·') # 分隔号
        elementValue = elementValue.replace('&deg;', '°') # 单位度
        elementValue = elementValue.replace('&iexcl;', '¡') # 西班牙语反叹号
        elementValue = elementValue.replace('&iquest;', '¿') # 西班牙语反问号
        elementValue = elementValue.replace('&ntilde;', 'ñ') # 西班牙语n
        elementValue = elementValue.replace('&Ntilde;', 'Ñ') # 西班牙语N
        elementValue = elementValue.replace('&aacute;', 'á') # 西班牙语a
        elementValue = elementValue.replace('&eacute;', 'é') # 西班牙语e
        elementValue = elementValue.replace('&iacute;', 'í') # 西班牙语i
        elementValue = elementValue.replace('&oacute;', 'ó') # 西班牙语o
        elementValue = elementValue.replace('&uacute;', 'ú') # 西班牙语u
        elementValue = elementValue.replace('&Aacute;', 'Á') # 西班牙语A
        elementValue = elementValue.replace('&Eacute;', 'É') # 西班牙语E
        elementValue = elementValue.replace('&Iacute;', 'Í') # 西班牙语I
        elementValue = elementValue.replace('&Oacute;', 'Ó') # 西班牙语O
        elementValue = elementValue.replace('&Uacute;', 'Ú') # 西班牙语U
        elementValue = elementValue.replace('&Atilde;', 'Ã') # 西班牙语A~
        elementValue = elementValue.replace('&ordf;', 'ª') # 西班牙语上标a
        elementValue = elementValue.replace('&ordm;', 'º') # 西班牙语上标o
 
        elementValue = elementValue.replace('<!--msnavigation-->', '')
        elementValue = elementValue.replace('博主', owner)
        elementValue = elementValue.replace('<i>', '')
        elementValue = elementValue.replace('</i>', '')
        elementValue = elementValue.replace('<br /><br />', '<br />')
 
        elementValue = re.sub(r"(?:<\?xml.*?>)", "", elementValue)
        elementValue = re.sub(r"(?:<[TDSFHI].*?>)", "", elementValue)
        elementValue = re.sub(r"(?:<\/[TDSFHI].*?>)", "", elementValue)
        elementValue = re.sub(r"(?:<P.*?>)", "<p>", elementValue)
        elementValue = re.sub(r"(?:<(table|tbody|tr|td|div|span|img|script|font|hr|object|param).*?>)", "", elementValue)
        elementValue = re.sub(r"(?:<\/(table|tbody|tr|td|div|span|img|script|font|object).*?>)", "", elementValue)
        elementValue = re.sub(r"\n", "", elementValue) # 把替换造成的空行删除
 
        text = dom.createCDATASection(elementValue)
    tag.appendChild(text)
    return tag
 
def convertPubDate(date, timediff='+0000'):
    """
    convert 2003-08-22 16:01:56
    to Thu, 23 Aug 2007 05:47:54 +0000
    """
    year, mon, day = int(date[:4]), int(date[5:7]), int(date[8:10])
    time = date[11:]
    aday = datetime.datetime(year, mon, day)
    d = {'1':'Mon', '2':'Tus', '3':'Wen', '4':'Thur', '5':'Fri', '6':'Sat', '7':'Sun'}
    m = {'1':'Jan', '2':'Feb', '3':'Mar', '4':'Apr', '5':'May', '6':'Jun',
         '7':'Jul', '8':'Aug', '9':'Sep', '10':'Oct', '11':'Nov', '12':'Dec'}
    weekday = d[str(aday.isoweekday())]
    month = m[str(mon)]
    pubdate = "%s, %d %s %s %s %s" % (weekday, day, month, year, time, timediff)
    return pubdate
 
def validateEmail(email):
    '''
    '''
    pattern = r'^[0-9a-z][_.0-9a-z-]{0,31}@([0-9a-z][0-9a-z-]{0,30}[0-9a-z]\.){1,4}[a-z]{2,4}$'
    p = re.compile(pattern)
    m = p.match(email)
    if m:
        return True
    else:
        return False
 
def validateUrl(url):
    '''
    '''
    pattern = r'^[a-zA-z]+://(\w+(-\w+)*)(\.(\w+(-\w+)*))*(\?\S*)?$'
    p = re.compile(pattern)
    m = p.match(url)
    if m:
        return True
    else:
        return False
 
def main(argv=None):    
    global filename
    global owner
 
    if argv is None:
        argv = sys.argv
    # parse command line options
 
    args = sys.argv[1:]
    order='asc'
    if (len(args) == 2):
        print ('Converting...'),
        sys.stdout.flush()
        start = time()
        filename = args[0].replace('.xml', '')
        owner = args[1] # BlogBus没把博主名字输出，只能手动
        convert(args[0], args[1], order)
        end = time()
        print ('Done. Elapse %g seconds.' % (end - start))
 
if __name__ == "__main__":
    sys.exit(main())
</p>

import string
import random
from collections import defaultdict
def set_book(fin1,num):
    d = defaultdict(list) # 默认键值为列表
    l = []
    header = ()
    for line in fin1:
        line = line.replace('-', ' ')
        for word in line.rstrip().split(): # 空格换行为分割，单词存入列表
            l.append(word)
    for i in range(len(l)-num): # 以列表序号逐一推进方式建立字典
        header = (l[i-1],) # 元组header为前缀，做键
        for j in range(i,i+num-1):
            header += (l[j],)
            j += 1
        if l[i+num-1] not in d[header]:
            d[header].append(l[i+num-1]) # 列表后缀做键值
    return d
def next(start, book):
    return random.choice(book[start])
fin1 = open('emma.txt', encoding='utf-8')
prefix_num = 3 # 前缀个数
suffix_num = 100 # 后缀个数
book = set_book(fin1,prefix_num)
start = random.choice(list(book.keys())) # 随机前缀开头
final =  start
for i in range(suffix_num): # 截取最后几个单词为前缀找后缀
    final += (next(final[len(final)-prefix_num:], book),) 
for word in final:
    print(word, end=' ')
# reigns alone. A very proper compliment! and then follows the application, 
# which I think, my dear, you said you had a great deal happier if she had no 
# intellectual superiority to make atonement to herself, or frighten those 
# who might hate her into outward respect. She had never seen her look so well, 
# so lovely, so engaging. There was consciousness, animation, and warmth; 
# there was every appearance of its being all in proof of how much he was 
# in love with, how to be able to return! I shall try what I can do. 
# Harriet's features are very delicate, which makes a likeness

import string
import random
from time import time
def set_book(fin1):
    useless = string.punctuation + string.whitespace + '“' + '”' # 标点符号、换行符全部咔嚓掉
    d = {}
    i = 1
    for line in fin1:
        line = line.replace('-', ' ') # 有-的单词全部一分为二，这样真的好吗？
        for word in line.split():
            word = word.strip(useless)
            word = word.lower()
            if word not in d:
                d[word] = i # 录入字典的时候键值就是序号
                i += 1
            # d[word] = d.get(word, 0) + 1 # 反正我不算词频，这个没必要了
    return d
fin1 = open('emma.txt', encoding='utf-8')
start = time()
book1 = set_book(fin1)
book2 = dict(zip(book1.values(), book1.keys())) # 键和键值互换，序号成了唯一索引号
print('100 random words in book')
for i in range(100):
    if i > 1 and i%8 == 0:
        print()
    print(random.choice(book2), end=' ') # 索引号找词，想多快有多快
print()
end = time()
print(end - start)
# 100 random words in book
# solicit laughing preserve inebriety elton's unimpeded effusions unselfish
# intimate connect native judges charities travel informs colours
# enigmas bragge case greensward cox's particularly unexampled promise
# prone greensward dignity maps fourth christmas creature maximum
# graver mildest pleasant corrected increased named partridge marks
# following kept gloom conjecturing parlour inheriting say consulting
# magnified abundant produces sons malt add unenforceability beautifully
# richly striking confuse greatness asleep steps humility upon
# already paper delight liberties confide appendages undecided male
# prophecies esteem unadorned likelihood shopping deeply unbiased horrors
# man's dumplings business chapter shakespeare sees counsels attentive
# silenced ventured singular double mean waltzes requisite checks
# unattended qualified blessed surmises
# 0.12100672721862793

import string
fin = open('words.txt')
mydict = {}
for line in fin:
    word = line.strip()
    mydict[word] = ''
file = open('emma.txt', encoding = 'utf-8')
essay = file.read().lower()
essay = essay.replace('-', ' ')
pun = {}
str_all = '“' + '”' + string.punctuation
for x in str_all: # 建立各种标点符号字符的字典
    pun[x] = ''
useless = essay.maketrans(pun) # maketrans必须被替换和替换等长，字典完美解决这个问题
l = essay.translate(useless).split() # 那些含-的单词会死得很惨，但仍然算是个单词
print('this book has', len(l), 'words')
book = {}
for item in l: # 读取文件为字符串，字符串转为单词列表，列表转为计数的字典，单词为键，次数为键值
    book[item] = book.get(item, 0) + 1
list_words1 = sorted(list(zip(book.values(), book.keys())), reverse = True) # 字典转为列表，键与键值换位
print('this book has', len(list_words1), 'different words')
print('times', 'word', sep='\t')
count = 1
word_len = 0 # 限制最小词长
for times, word in list_words1: # 打印大于某长度用得最多的20个词（不限制，3个字母及以下最最简单的会刷屏）
    if len(word) > word_len:
        print(times, word, sep='\t')
        count += 1
    if count > 20:
        break
count = 0
for word in book:
    if word not in mydict:
        # print(word, end=' ')
        count += 1
print(count, 'words in book not in dict') # 结果惨不忍睹，合计590个
# this book has 164065 words
# this book has 7479 different words
# times   word
# 5379    the
# 5322    to
# 4965    and
# 4412    of
# 3191    i
# 3187    a
# 2544    it
# 2483    her
# 2401    was
# 2365    she
# 2246    in
# 2172    not
# 2069    you
# 1995    be
# 1815    that
# 1813    he
# 1626    had
# 1448    as
# 1446    but
# 1373    for
# 590 words in book not in dict
# -----------------------------解法二----------------------------- 其实就是切单词方法有差异
import string
def set_book(fin1):
    useless = string.punctuation + string.whitespace + '“' + '”'
    d = {}
    for line in fin1:
        line = line.replace('-', ' ')
        for word in line.split():
            word = word.strip(useless)
            word = word.lower()
            d[word] = d.get(word, 0) + 1
    return d
def set_dict(fin2):
    d = {}
    for line in fin2:
        word = line.strip()
        d[word] = d.get(word, 0) + 1
    return d
fin1 = open('emma.txt', encoding='utf-8')
fin2 = open('words.txt')
book = set_book(fin1)
mydict = set_dict(fin2)
l = sorted(list(zip(book.values(), book.keys())), reverse=True)
count = 0
for key in book:
    count = count + book[key]
print('this book has', count, 'words')
print('this book has', len(book), 'different words')
num = 20
print(num, 'most common words in this book')
print('times', 'word', sep='\t')
for times, word in l:
    print(times, word, sep='\t')
    num -= 1
    if num < 1:
        break
count = 0
for word in book:
    if word not in mydict:
        # print(word, end=' ')
        count += 1
# print()
print(count, 'words in book not in dict')
# this book has 164120 words
# this book has 7531 different words
# 20 most common words in this book
# times   word
# 5379    the
# 5322    to
# 4965    and
# 4412    of
# 3191    i
# 3187    a
# 2544    it
# 2483    her
# 2401    was
# 2364    she
# 2246    in
# 2172    not
# 2069    you
# 1995    be
# 1815    that
# 1813    he
# 1626    had
# 1448    as
# 1446    but
# 1373    for
# 683 words in book not in dict