"""
Text widget specialized to display the HTML formatted Python documentation
as generated by Sphinx.

The renderer is fragile enough that you won't likely get very far if you
feed it anything but the IDLE help file.
"""

from tkinter import Tk, Text, Scrollbar, Toplevel
from html.parser import HTMLParser
import os


class SphinxHTMLViewer(Text):

    class MyHTMLParser(HTMLParser):
        def __init__(self, t):
            HTMLParser.__init__(self)
            self.t = t              # text widget we're rendering into
            self.tags = ''          # current text tags to apply
            self.show = False       # used so we exclude page navigation
            self.hdrlink = False    # used so we don't show header links
            self.level = 0          # indentation level
            self.pre = False        # displaying preformatted text
            self.hprefix = ''       # strip e.g. '25.5' from headings
            self.nested_dl = False  # if we're in a nested <dl>

        def indent(self, amt=1):
            self.level += amt
            self.tags = '' if self.level == 0 else 'l'+str(self.level)

        def handle_starttag(self, tag, attrs):
            class_ = ''
            for a, v in attrs:
                if a == 'class':
                    class_ = v
            s = ''
            if tag == 'div' and class_ == 'section':
                self.show = True    # start of main content
            elif tag == 'div' and class_ == 'sphinxsidebar':
                self.show = False   # end of main content
            elif tag == 'p' and class_ != 'first':
                s = '\n\n'
            elif tag == 'span' and class_ == 'pre':
                self.tags = 'pre'
            elif tag == 'span' and class_ == 'versionmodified':
                self.tags = 'em'
            elif tag == 'em':
                self.tags = 'em'
            elif tag in ['ul', 'ol']:
                self.indent()
            elif tag == 'dl':
                if self.level > 0:
                    self.nested_dl = True
            elif tag == 'li':
                s = '\n\n* '
            elif tag == 'dt':
                s = '\n\n' if not self.nested_dl else '\n'  # avoid extra line
                self.nested_dl = False
            elif tag == 'dd':
                self.indent()
                s = '\n'
            elif tag == 'pre':
                self.pre = True
                self.tags = 'pre'
                s = '\n\n'
            elif tag == 'a' and class_ == 'headerlink':
                self.hdrlink = True
            elif tag in ['h1', 'h2', 'h3']:
                if self.show:
                    self.t.insert('end', '\n\n\n')
                self.tags = tag
            if self.show:
                self.t.insert('end', s, self.tags)

        def handle_endtag(self, tag):
            if tag in ['h1', 'h2', 'h3', 'span', 'em']:
                self.indent(0)  # clear tag, reset indent
            elif tag == 'a':
                self.hdrlink = False
            elif tag == 'pre':
                self.pre = False
                self.tags = ''
            elif tag in ['ul', 'dd', 'ol']:
                self.indent(amt=-1)

        def handle_data(self, data):
            if self.show and not self.hdrlink:
                d = data if self.pre else data.replace('\n', ' ')
                if self.tags == 'h1':
                    self.hprefix = d[0:d.index(' ')]
                if self.tags in ['h1', 'h2', 'h3'] and self.hprefix != '':
                    if d[0:len(self.hprefix)] == self.hprefix:
                        d = d[len(self.hprefix):].strip()
                self.t.insert('end', d, self.tags)

    def __init__(self, parent, filename):
        Text.__init__(self, parent, wrap='word', font=('helvetica', 12),
                      highlightthickness=0, padx=10, spacing1='2px',
                      spacing2='2px')
        p = self.MyHTMLParser(self)
        with open(filename, encoding='utf-8') as f:
            contents = f.read()
        p.feed(contents)
        self.tag_configure('em', font=('helvetica', 12, 'italic'))
        self.tag_configure('h1', font=('helvetica', 24, 'bold'))
        self.tag_configure('h2', font=('helvetica', 20, 'bold'))
        self.tag_configure('h3', font=('helvetica', 15, 'bold'))
        self.tag_configure('pre', font=('courier', 12), lmargin1=25)
        self.tag_configure('l1', lmargin1=25, lmargin2=25)
        self.tag_configure('l2', lmargin1=50, lmargin2=50)
        self.tag_configure('l3', lmargin1=75, lmargin2=75)
        self.tag_configure('l4', lmargin1=100, lmargin2=100)
        self['state'] = 'disabled'


class SphinxHTMLViewerWindow(Toplevel):

    def __init__(self, parent, filename, title):
        Toplevel.__init__(self, parent)
        self.wm_title(title)        
        self.protocol("WM_DELETE_WINDOW", self.destroy)
        v = SphinxHTMLViewer(self, filename)
        s = Scrollbar(self, command=v.yview)
        v['yscrollcommand'] = s.set
        v.grid(column=0, row=0, sticky='nsew')
        s.grid(column=1, row=0, sticky='ns')
        self.grid_columnconfigure(0, weight=1)
        self.grid_rowconfigure(0, weight=1)


if __name__ == '__main__':
    root = Tk()
    w = SphinxHTMLViewerWindow(root,
             os.path.join(os.path.abspath(os.path.dirname(__file__)),
             'idle.html'), 'IDLE Help')
    root.mainloop()