python3生成标签云

Posted 2020-08-07 TTyb

标签云是现在大数据里面最喜欢使用的一种展现方式，其中在python3下也能实现标签云的效果，贴图如下：

-------------------进入正文---------------------

首先要安装以下几个库：

1 #!/usr/bin/python3.4
2 # -*- coding: utf-8 -*-
3 
4 # http://www.lfd.uci.edu/~gohlke/pythonlibs/#cx_freeze
5 # 万能仓库下载pygame
6 # pip3下载simplejson

还有最重要的库：

pip3 install pytagcloud

或者去官网下载：

https://pypi.python.org/pypi/pytagcloud/

安装完毕，利用官网的例子来做：

1 from pytagcloud import create_tag_image, make_tags
2 from pytagcloud.lang.counter import get_tag_counts
3 
4 YOUR_TEXT = "A tag cloud is a visual representation for text data, typically\\
5 used to depict keyword metadata on websites, or to visualize free form text."
6 
7 tags = make_tags(get_tag_counts(YOUR_TEXT), maxsize=120)
8 
9 create_tag_image(tags, \'cloud_large.png\', size=(900, 600), fontname=\'Lobster\')

果断报错：

Traceback (most recent call last):
  File "D:/code/pythonwork/Text.py", line 96, in <module>
    tags = make_tags(get_tag_counts(YOUR_TEXT), maxsize=120)
  File "C:\\Python34\\lib\\site-packages\\pytagcloud\\lang\\counter.py", line 25, in get_tag_counts
    return sorted(counted.iteritems(), key=itemgetter(1), reverse=True)
AttributeError: \'dict\' object has no attribute \'iteritems\'

看了发现问题出在库中的：

# counter.py
return sorted(counted.iteritems(), key=itemgetter(1), reverse=True)

原来是python3.4不支持写法：

在Python2.x中，items( )用于 返回一个字典的拷贝列表【Returns a copy of the list of all items (key/value pairs) in D】，占额外的内存。

iteritems() 用于返回本身字典列表操作后的迭代【Returns an iterator on all items(key/value pairs) in D】，不占用额外的内存。

Python 3.x 里面，iteritems() 和 viewitems() 这两个方法都已经废除了，而 items() 得到的结果是和 2.x 里面 viewitems() 一致的。在3.x 里 用 items()替换iteritems() ，可以用于 for 来循环遍历。

但是当我换成：

# counter.py
return sorted(counted.items(), key=itemgetter(1), reverse=True)

发现运行并没有错误，但是没有生成标签云啊，一遍一遍打印出来，终于找到问题了：

from pytagcloud import create_tag_image

这个是为了生成一个元组的东西：

    # counts =[(\'cloud\', 3),
    # (\'words\', 2),
    # (\'code\', 1),
    # (\'word\', 1),
    # (\'appear\', 1)]

但是python3里面的items()是达不到这个效果的，所以我就自己写吧。

读取txt文件，将每一行都按照空格划分成一个个数组的元素：

1  arr = []
2  file = open(\'../tagcloud/tag_file.txt\', \'r\')
3  data = file.read().split(\'\\r\\n\')
4  for content in data:
5     contents = validatecontent(content).split()
6     for word in contents:
7         arr.append(word)

[\'BAISC\', \'Python\', \'BASICA\', \'GVBASIC\', \'GWBASIC\', \'Python\', \'ETBASIC\', \'QBASIC\', \'Quick\', \'Basic\', \'Turbo\', \'Basic\', \'True\', \'Python\', \'java\', \'Basic\', \'Visual\', \'Basic\', \'Visual\', \'Basic\', \'Net\', \'Power\', \'Basic\', \'Python\', \'java\', \'SQL\', \'VB\', \'Small\', \'Basic\', \'Free\', \'Basic\', \'DarkBASIC\', \'VBScript\', \'Visual\', \'Basic\', \'For\', \'ApplicationsVBA\', \'REALbasic\', \'C\', \'C\', \'Turbo\', \'C\', \'Python\', \'java\', \'SQL\', \'VB\', \'php\', \'html\', \'Borland\', \'C\', \'C\', \'Builder\', \'CCLI\', \'Python\', \'java\', \'ObjectiveC\', \'C#\', \'Microsoft\', \'Visual\', \'C\', \'Pascal\', \'Delphi\', \'Turbo\', \'Python\', \'java\', \'SQL\', \'VB\', \'PHP\', \'HTML\', \'Pascal\', \'Object\', \'Pascal\', \'Free\', \'Pascal\', \'Lazarus\', \'FORTRAN\', \'MATLAB\', \'Scilab\', \'GNU\', \'Octave\', \'R\', \'SPlus\', \'Mathematica\', \'Maple\', \'Python\', \'java\', \'SQL\', \'VB\', \'PHP\', \'HTML\', \'Julia\', \'xBaseClipper\', \'Visual\', \'FoxPro\', \'SQLPLSQL\', \'TSQL\', \'SQLPSM\', \'LINQ\', \'Xquer\', \'Lua\', \'Python\', \'java\', \'SQL\', \'VB\', \'Perl\', \'PHP\', \'Python\', \'Ruby\', \'ASP\', \'JSP\', \'TclTk\', \'VBScript\', \'AppleScript\', \'AAuto\', \'ActionScript\', \'DMDScript\', \'ECMAScript\', \'javascript\', \'JScript\', \'TypeScript\', \'sh\', \'bash\', \'Python\', \'java\', \'SQL\', \'VB\', \'PHP\', \'HTML\', \'sed\', \'awk\', \'PowerShell\', \'csh\', \'tcsh\', \'ksh\', \'zsh\', \'XMLSVG\', \'XML\', \'Schema\', \'Python\', \'java\', \'XSLT\', \'XHTML\', \'MathML\', \'XAML\', \'SSML\', \'SGML\', \'HTML\', \'Python\', \'java\', \'SQL\', \'VB\', \'Curl\', \'SVG\', \'XML\', \'Schema\', \'XSLT\', \'XHTML\', \'MathML\', \'XAML\', \'SSML\', \'Java\', \'Jython\', \'JRuby\', \'JScheme\', \'Groovy\', \'Kawa\', \'Scala\', \'Clojure\', \'ALGOL\', \'APLJ\', \'Ada\', \'Falcon\', \'Forth\', \'Io\', \'MUMPS\', \'PLI\', \'PostScript\', \'REXX\', \'SAC\', \'Self\', \'Simula\', \'Swift\', \'IronPython\', \'IronRuby\', \'COBOL\', \'Python\', \'java\', \'SQL\', \'VB\', \'PHP\', \'HTML\']

其中validatecontent是起初非法字符的函数：

1 # 去除内容中的非法字符 (Windows)
2 def validatecontent(content):
3     # \'/\\:*?"<>|\'
4     rstr = r"[\\/\\\\\\:\\*\\?\\"\\<\\>\\|\\.\\*\\+\\-\\(\\)\\"\\\'\\（\\）\\！\\？\\“\\”\\,\\。\\；\\：\\{\\}\\{\\}\\=\\%\\*\\~\\·]"
5     new_content = re.sub(rstr, "", content)
6     return new_content

 

对每一个元素都来个计数：

from collections import Counter
counts = Counter(arr).items()
print(counts)

效果出来了：

dict_items([(\'For\', 1), (\'SQL\', 8), (\'JRuby\', 1), (\'Builder\', 1), (\'HTML\', 6), (\'LINQ\', 1), (\'BAISC\', 1), (\'BASICA\', 1), (\'PHP\', 6), (\'Octave\', 1), (\'csh\', 1), (\'PostScript\', 1), (\'awk\', 1), (\'Ruby\', 1), (\'AppleScript\', 1), (\'Object\', 1), (\'java\', 11), (\'TclTk\', 1), (\'Xquer\', 1), (\'ksh\', 1), (\'zsh\', 1), (\'ETBASIC\', 1), (\'AAuto\', 1), (\'Borland\', 1), (\'SVG\', 1), (\'Jython\', 1), (\'Simula\', 1), (\'IronPython\', 1), (\'Python\', 14), (\'Microsoft\', 1), (\'ActionScript\', 1), (\'XHTML\', 2), (\'REXX\', 1), (\'COBOL\', 1), (\'Scilab\', 1), (\'Ada\', 1), (\'Basic\', 9), (\'GVBASIC\', 1), (\'ECMAScript\', 1), (\'TypeScript\', 1), (\'Falcon\', 1), (\'Clojure\', 1), (\'ASP\', 1), (\'ALGOL\', 1), (\'XMLSVG\', 1), (\'GWBASIC\', 1), (\'VBScript\', 2), (\'CCLI\', 1), (\'Lazarus\', 1), (\'Julia\', 1), (\'JSP\', 1), (\'PowerShell\', 1), (\'IronRuby\', 1), (\'Power\', 1), (\'FORTRAN\', 1), (\'Self\', 1), (\'Perl\', 1), (\'Small\', 1), (\'FoxPro\', 1), (\'REALbasic\', 1), (\'GNU\', 1), (\'Mathematica\', 1), (\'True\', 1), (\'Visual\', 5), (\'JScheme\', 1), (\'Maple\', 1), (\'Quick\', 1), (\'Turbo\', 3), (\'SAC\', 1), (\'JScript\', 1), (\'APLJ\', 1), (\'sh\', 1), (\'Kawa\', 1), (\'Pascal\', 4), (\'TSQL\', 1), (\'SPlus\', 1), (\'C\', 6), (\'xBaseClipper\', 1), (\'tcsh\', 1), (\'SQLPSM\', 1), (\'ApplicationsVBA\', 1), (\'SSML\', 2), (\'R\', 1), (\'Groovy\', 1), (\'XSLT\', 2), (\'MUMPS\', 1), (\'bash\', 1), (\'DarkBASIC\', 1), (\'SGML\', 1), (\'XAML\', 2), (\'VB\', 8), (\'Curl\', 1), (\'Schema\', 2), (\'MATLAB\', 1), (\'MathML\', 2), (\'Lua\', 1), (\'Net\', 1), (\'ObjectiveC\', 1), (\'JavaScript\', 1), (\'Java\', 1), (\'Io\', 1), (\'Free\', 2), (\'Delphi\', 1), (\'sed\', 1), (\'XML\', 2), (\'Forth\', 1), (\'C#\', 1), (\'SQLPLSQL\', 1), (\'QBASIC\', 1), (\'DMDScript\', 1), (\'Swift\', 1), (\'Scala\', 1), (\'PLI\', 1)])

最后直接代入进去就行了：

1 tags = make_tags(counts, maxsize=120)
2 
3 create_tag_image(tags, \'cloud_large.png\', size=(900, 600), fontname=\'Lobster\')

具体的修正需要自己慢慢去琢磨了，比如文字大小、图片大小、背景颜色等等。

到这里标签云是算完成了的，但是却是不支持中文，原因是没有合适的ttf字体文件，准备一个 ttf 中文字体，如MicrosoftYaHei.ttf ，将其移动到

# C:\\Python34\\Lib\\site-packages\\pytagcloud\\fonts

接着就是更改fonts.json文件，按照样式添加类似于css的东西：

{
        "name": "MicrosoftYaHei",
        "ttf": "MicrosoftYaHei.ttf",
        "web": "none"
    }

注意前后的逗号就行。最后将这里的代码改一下：

create_tag_image(tags, \'cloud_large.png\', size=(900, 600), fontname=\'MicrosoftYaHei\')

运行，搞定！中文效果图：

我的在github里面，可以去下载看看。