TocTreeprocessor.add_permalink() - Code Metrics - Inspection of "pythonx/markdown_parser.py" - MikeCoder/markdown-preview.vim - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( 32cfa8...ec62d3 )

by Dongxin

created 2017-08-15 03:07 UTC

TocTreeprocessor.add_permalink() A

↳ Parent: TocTreeprocessor

Complexity

Conditions

Size

Total Lines

Duplication

Lines	0
Ratio	0 %

Importance

Changes	1
Bugs	0	Features	0

Metric	Value
cc	2
c	1
b	0
f	0
dl	0
loc	9
rs	9.6666

"""
Table of Contents Extension for Python-Markdown
===============================================

See <https://pythonhosted.org/Markdown/extensions/toc.html>
for documentation.

Oringinal code Copyright 2008 [Jack Miller](http://codezen.org)

All changes Copyright 2008-2014 The Python Markdown Project

License: [BSD](http://www.opensource.org/licenses/bsd-license.php)

"""

from __future__ import absolute_import
from __future__ import unicode_literals
from . import Extension
from ..treeprocessors import Treeprocessor
from ..util import etree, parseBoolValue, AMP_SUBSTITUTE, HTML_PLACEHOLDER_RE, string_type
import re
import unicodedata


def slugify(value, separator):
    """ Slugify a string, to make it URL friendly. """
    value = unicodedata.normalize('NFKD', value).encode('ascii', 'ignore')
    value = re.sub('[^\w\s-]', '', value.decode('ascii')).strip().lower()
    return re.sub('[%s\s]+' % separator, separator, value)


IDCOUNT_RE = re.compile(r'^(.*)_([0-9]+)$')


def unique(id, ids):
    """ Ensure id is unique in set of ids. Append '_1', '_2'... if not """
    while id in ids or not id:
        m = IDCOUNT_RE.match(id)
        if m:
            id = '%s_%d' % (m.group(1), int(m.group(2))+1)
        else:
            id = '%s_%d' % (id, 1)
    ids.add(id)
    return id


def stashedHTML2text(text, md):
    """ Extract raw HTML from stash, reduce to plain text and swap with placeholder. """
    def _html_sub(m):
        """ Substitute raw html with plain text. """
        try:
            raw, safe = md.htmlStash.rawHtmlBlocks[int(m.group(1))]
        except (IndexError, TypeError):  # pragma: no cover
            return m.group(0)
        if md.safeMode and not safe:  # pragma: no cover
            return ''
        # Strip out tags and entities - leaveing text
        return re.sub(r'(<[^>]+>)|(&[\#a-zA-Z0-9]+;)', '', raw)

    return HTML_PLACEHOLDER_RE.sub(_html_sub, text)


def nest_toc_tokens(toc_list):
    """Given an unsorted list with errors and skips, return a nested one.
    [{'level': 1}, {'level': 2}]
    =>
    [{'level': 1, 'children': [{'level': 2, 'children': []}]}]

    A wrong list is also converted:
    [{'level': 2}, {'level': 1}]
    =>
    [{'level': 2, 'children': []}, {'level': 1, 'children': []}]
    """

    ordered_list = []
    if len(toc_list):
        # Initialize everything by processing the first entry
        last = toc_list.pop(0)
        last['children'] = []
        levels = [last['level']]
        ordered_list.append(last)
        parents = []

        # Walk the rest nesting the entries properly
        while toc_list:
            t = toc_list.pop(0)
            current_level = t['level']
            t['children'] = []

            # Reduce depth if current level < last item's level
            if current_level < levels[-1]:
                # Pop last level since we know we are less than it
                levels.pop()

                # Pop parents and levels we are less than or equal to
                to_pop = 0
                for p in reversed(parents):
                    if current_level <= p['level']:
                        to_pop += 1
                    else:  # pragma: no cover
                        break
                if to_pop:
                    levels = levels[:-to_pop]
                    parents = parents[:-to_pop]

                # Note current level as last
                levels.append(current_level)

            # Level is the same, so append to
            # the current parent (if available)
            if current_level == levels[-1]:
                (parents[-1]['children'] if parents
                 else ordered_list).append(t)

            # Current level is > last item's level,
            # So make last item a parent and append current as child
            else:
                last['children'].append(t)
                parents.append(last)
                levels.append(current_level)
            last = t

    return ordered_list


class TocTreeprocessor(Treeprocessor):
    def __init__(self, md, config):
        super(TocTreeprocessor, self).__init__(md)

        self.marker = config["marker"]
        self.title = config["title"]
        self.base_level = int(config["baselevel"]) - 1
        self.slugify = config["slugify"]
        self.sep = config["separator"]
        self.use_anchors = parseBoolValue(config["anchorlink"])
        self.use_permalinks = parseBoolValue(config["permalink"], False)
        if self.use_permalinks is None:
            self.use_permalinks = config["permalink"]

        self.header_rgx = re.compile("[Hh][123456]")

    def iterparent(self, root):
        ''' Iterator wrapper to get parent and child all at once. '''
        for parent in root.iter():
            for child in parent:
                yield parent, child

    def replace_marker(self, root, elem):
        ''' Replace marker with elem. '''
        for (p, c) in self.iterparent(root):
            text = ''.join(c.itertext()).strip()
            if not text:
                continue

            # To keep the output from screwing up the
            # validation by putting a <div> inside of a <p>
            # we actually replace the <p> in its entirety.
            # We do not allow the marker inside a header as that
            # would causes an enless loop of placing a new TOC
            # inside previously generated TOC.
            if c.text and c.text.strip() == self.marker and \
               not self.header_rgx.match(c.tag) and c.tag not in ['pre', 'code']:
                for i in range(len(p)):
                    if p[i] == c:
                        p[i] = elem
                        break

    def set_level(self, elem):
        ''' Adjust header level according to base level. '''
        level = int(elem.tag[-1]) + self.base_level
        if level > 6:
            level = 6
        elem.tag = 'h%d' % level

    def add_anchor(self, c, elem_id):  # @ReservedAssignment
        anchor = etree.Element("a")
        anchor.text = c.text
        anchor.attrib["href"] = "#" + elem_id
        anchor.attrib["class"] = "toclink"
        c.text = ""
        for elem in c:
            anchor.append(elem)
        while c:
            c.remove(c[0])
        c.append(anchor)

    def add_permalink(self, c, elem_id):
        permalink = etree.Element("a")
        permalink.text = ("%spara;" % AMP_SUBSTITUTE
                          if self.use_permalinks is True
                          else self.use_permalinks)
        permalink.attrib["href"] = "#" + elem_id
        permalink.attrib["class"] = "headerlink"
        permalink.attrib["title"] = "Permanent link"
        c.append(permalink)

    def build_toc_div(self, toc_list):
        """ Return a string div given a toc list. """
        div = etree.Element("div")
        div.attrib["class"] = "toc"

        # Add title to the div
        if self.title:
            header = etree.SubElement(div, "span")
            header.attrib["class"] = "toctitle"
            header.text = self.title

        def build_etree_ul(toc_list, parent):
            ul = etree.SubElement(parent, "ul")
            for item in toc_list:
                # List item link, to be inserted into the toc div
                li = etree.SubElement(ul, "li")
                link = etree.SubElement(li, "a")
                link.text = item.get('name', '')
                link.attrib["href"] = '#' + item.get('id', '')
                if item['children']:
                    build_etree_ul(item['children'], li)
            return ul

        build_etree_ul(toc_list, div)
        prettify = self.markdown.treeprocessors.get('prettify')
        if prettify:
            prettify.run(div)
        return div

    def run(self, doc):
        # Get a list of id attributes
        used_ids = set()
        for el in doc.iter():
            if "id" in el.attrib:
                used_ids.add(el.attrib["id"])

        toc_tokens = []
        for el in doc.iter():
            if isinstance(el.tag, string_type) and self.header_rgx.match(el.tag):
                self.set_level(el)
                text = ''.join(el.itertext()).strip()

                # Do not override pre-existing ids
                if "id" not in el.attrib:
                    innertext = stashedHTML2text(text, self.markdown)
                    el.attrib["id"] = unique(self.slugify(innertext, self.sep), used_ids)

                toc_tokens.append({
                    'level': int(el.tag[-1]),
                    'id': el.attrib["id"],
                    'name': text
                })

                if self.use_anchors:
                    self.add_anchor(el, el.attrib["id"])
                if self.use_permalinks:
                    self.add_permalink(el, el.attrib["id"])

        div = self.build_toc_div(nest_toc_tokens(toc_tokens))
        if self.marker:
            self.replace_marker(doc, div)

        # serialize and attach to markdown instance.
        toc = self.markdown.serializer(div)
        for pp in self.markdown.postprocessors.values():
            toc = pp.run(toc)
        self.markdown.toc = toc


class TocExtension(Extension):

    TreeProcessorClass = TocTreeprocessor

    def __init__(self, *args, **kwargs):
        self.config = {
            "marker": ['[TOC]',
                       'Text to find and replace with Table of Contents - '
                       'Set to an empty string to disable. Defaults to "[TOC]"'],
            "title": ["",
                      "Title to insert into TOC <div> - "
                      "Defaults to an empty string"],
            "anchorlink": [False,
                           "True if header should be a self link - "
                           "Defaults to False"],
            "permalink": [0,
                          "True or link text if a Sphinx-style permalink should "
                          "be added - Defaults to False"],
            "baselevel": ['1', 'Base level for headers.'],
            "slugify": [slugify,
                        "Function to generate anchors based on header text - "
                        "Defaults to the headerid ext's slugify function."],
            'separator': ['-', 'Word separator. Defaults to "-".']
        }

        super(TocExtension, self).__init__(*args, **kwargs)

    def extendMarkdown(self, md, md_globals):
        md.registerExtension(self)
        self.md = md
        self.reset()
        tocext = self.TreeProcessorClass(md, self.getConfigs())
        # Headerid ext is set to '>prettify'. With this set to '_end',
        # it should always come after headerid ext (and honor ids assinged
        # by the header id extension) if both are used. Same goes for
        # attr_list extension. This must come last because we don't want
        # to redefine ids after toc is created. But we do want toc prettified.
        md.treeprocessors.add("toc", tocext, "_end")

    def reset(self):
        self.md.toc = ''


def makeExtension(*args, **kwargs):
    return TocExtension(*args, **kwargs)


1			"""
2			Table of Contents Extension for Python-Markdown
3			===============================================
4
5			See <https://pythonhosted.org/Markdown/extensions/toc.html>
6			for documentation.
7
8			Oringinal code Copyright 2008 [Jack Miller](http://codezen.org)
9
10			All changes Copyright 2008-2014 The Python Markdown Project
11
12			License: [BSD](http://www.opensource.org/licenses/bsd-license.php)
13
14			"""
15
16			from __future__ import absolute_import
17			from __future__ import unicode_literals
18			from . import Extension
19			from ..treeprocessors import Treeprocessor
20			from ..util import etree, parseBoolValue, AMP_SUBSTITUTE, HTML_PLACEHOLDER_RE, string_type
21			import re
22			import unicodedata
23
24
25			def slugify(value, separator):
26			""" Slugify a string, to make it URL friendly. """
27			value = unicodedata.normalize('NFKD', value).encode('ascii', 'ignore')
28			value = re.sub('[^\w\s-]', '', value.decode('ascii')).strip().lower()
29			return re.sub('[%s\s]+' % separator, separator, value)
30
31
32			IDCOUNT_RE = re.compile(r'^(.*)_([0-9]+)$')
33
34
35			def unique(id, ids):
36			""" Ensure id is unique in set of ids. Append '_1', '_2'... if not """
37			while id in ids or not id:
38			m = IDCOUNT_RE.match(id)
39			if m:
40			id = '%s_%d' % (m.group(1), int(m.group(2))+1)
41			else:
42			id = '%s_%d' % (id, 1)
43			ids.add(id)
44			return id
45
46
47			def stashedHTML2text(text, md):
48			""" Extract raw HTML from stash, reduce to plain text and swap with placeholder. """
49			def _html_sub(m):
50			""" Substitute raw html with plain text. """
51			try:
52			raw, safe = md.htmlStash.rawHtmlBlocks[int(m.group(1))]
53			except (IndexError, TypeError): # pragma: no cover
54			return m.group(0)
55			if md.safeMode and not safe: # pragma: no cover
56			return ''
57			# Strip out tags and entities - leaveing text
58			return re.sub(r'(<[^>]+>)\|(&[\#a-zA-Z0-9]+;)', '', raw)
59
60			return HTML_PLACEHOLDER_RE.sub(_html_sub, text)
61
62
63			def nest_toc_tokens(toc_list):
64			"""Given an unsorted list with errors and skips, return a nested one.
65			[{'level': 1}, {'level': 2}]
66			=>
67			[{'level': 1, 'children': [{'level': 2, 'children': []}]}]
68
69			A wrong list is also converted:
70			[{'level': 2}, {'level': 1}]
71			=>
72			[{'level': 2, 'children': []}, {'level': 1, 'children': []}]
73			"""
74
75			ordered_list = []
76			if len(toc_list):
77			# Initialize everything by processing the first entry
78			last = toc_list.pop(0)
79			last['children'] = []
80			levels = [last['level']]
81			ordered_list.append(last)
82			parents = []
83
84			# Walk the rest nesting the entries properly
85			while toc_list:
86			t = toc_list.pop(0)
87			current_level = t['level']
88			t['children'] = []
89
90			# Reduce depth if current level < last item's level
91			if current_level < levels[-1]:
92			# Pop last level since we know we are less than it
93			levels.pop()
94
95			# Pop parents and levels we are less than or equal to
96			to_pop = 0
97			for p in reversed(parents):
98			if current_level <= p['level']:
99			to_pop += 1
100			else: # pragma: no cover
101			break
102			if to_pop:
103			levels = levels[:-to_pop]
104			parents = parents[:-to_pop]
105
106			# Note current level as last
107			levels.append(current_level)
108
109			# Level is the same, so append to
110			# the current parent (if available)
111			if current_level == levels[-1]:
112			(parents[-1]['children'] if parents
113			else ordered_list).append(t)
114
115			# Current level is > last item's level,
116			# So make last item a parent and append current as child
117			else:
118			last['children'].append(t)
119			parents.append(last)
120			levels.append(current_level)
121			last = t
122
123			return ordered_list
124
125
126			class TocTreeprocessor(Treeprocessor):
127			def __init__(self, md, config):
128			super(TocTreeprocessor, self).__init__(md)
129
130			self.marker = config["marker"]
131			self.title = config["title"]
132			self.base_level = int(config["baselevel"]) - 1
133			self.slugify = config["slugify"]
134			self.sep = config["separator"]
135			self.use_anchors = parseBoolValue(config["anchorlink"])
136			self.use_permalinks = parseBoolValue(config["permalink"], False)
137			if self.use_permalinks is None:
138			self.use_permalinks = config["permalink"]
139
140			self.header_rgx = re.compile("[Hh][123456]")
141
142			def iterparent(self, root):
143			''' Iterator wrapper to get parent and child all at once. '''
144			for parent in root.iter():
145			for child in parent:
146			yield parent, child
147
148			def replace_marker(self, root, elem):
149			''' Replace marker with elem. '''
150			for (p, c) in self.iterparent(root):
151			text = ''.join(c.itertext()).strip()
152			if not text:
153			continue
154
155			# To keep the output from screwing up the
156			# validation by putting a <div> inside of a <p>
157			# we actually replace the <p> in its entirety.
158			# We do not allow the marker inside a header as that
159			# would causes an enless loop of placing a new TOC
160			# inside previously generated TOC.
161			if c.text and c.text.strip() == self.marker and \
162			not self.header_rgx.match(c.tag) and c.tag not in ['pre', 'code']:
163			for i in range(len(p)):
164			if p[i] == c:
165			p[i] = elem
166			break
167
168			def set_level(self, elem):
169			''' Adjust header level according to base level. '''
170			level = int(elem.tag[-1]) + self.base_level
171			if level > 6:
172			level = 6
173			elem.tag = 'h%d' % level
174
175			def add_anchor(self, c, elem_id): # @ReservedAssignment
176			anchor = etree.Element("a")
177			anchor.text = c.text
178			anchor.attrib["href"] = "#" + elem_id
179			anchor.attrib["class"] = "toclink"
180			c.text = ""
181			for elem in c:
182			anchor.append(elem)
183			while c:
184			c.remove(c[0])
185			c.append(anchor)
186
187			def add_permalink(self, c, elem_id):
188			permalink = etree.Element("a")
189			permalink.text = ("%spara;" % AMP_SUBSTITUTE
190			if self.use_permalinks is True
191			else self.use_permalinks)
192			permalink.attrib["href"] = "#" + elem_id
193			permalink.attrib["class"] = "headerlink"
194			permalink.attrib["title"] = "Permanent link"
195			c.append(permalink)
196
197			def build_toc_div(self, toc_list):
198			""" Return a string div given a toc list. """
199			div = etree.Element("div")
200			div.attrib["class"] = "toc"
201
202			# Add title to the div
203			if self.title:
204			header = etree.SubElement(div, "span")
205			header.attrib["class"] = "toctitle"
206			header.text = self.title
207
208			def build_etree_ul(toc_list, parent):
209			ul = etree.SubElement(parent, "ul")
210			for item in toc_list:
211			# List item link, to be inserted into the toc div
212			li = etree.SubElement(ul, "li")
213			link = etree.SubElement(li, "a")
214			link.text = item.get('name', '')
215			link.attrib["href"] = '#' + item.get('id', '')
216			if item['children']:
217			build_etree_ul(item['children'], li)
218			return ul
219
220			build_etree_ul(toc_list, div)
221			prettify = self.markdown.treeprocessors.get('prettify')
222			if prettify:
223			prettify.run(div)
224			return div
225
226			def run(self, doc):
227			# Get a list of id attributes
228			used_ids = set()
229			for el in doc.iter():
230			if "id" in el.attrib:
231			used_ids.add(el.attrib["id"])
232
233			toc_tokens = []
234			for el in doc.iter():
235			if isinstance(el.tag, string_type) and self.header_rgx.match(el.tag):
236			self.set_level(el)
237			text = ''.join(el.itertext()).strip()
238
239			# Do not override pre-existing ids
240			if "id" not in el.attrib:
241			innertext = stashedHTML2text(text, self.markdown)
242			el.attrib["id"] = unique(self.slugify(innertext, self.sep), used_ids)
243
244			toc_tokens.append({
245			'level': int(el.tag[-1]),
246			'id': el.attrib["id"],
247			'name': text
248			})
249
250			if self.use_anchors:
251			self.add_anchor(el, el.attrib["id"])
252			if self.use_permalinks:
253			self.add_permalink(el, el.attrib["id"])
254
255			div = self.build_toc_div(nest_toc_tokens(toc_tokens))
256			if self.marker:
257			self.replace_marker(doc, div)
258
259			# serialize and attach to markdown instance.
260			toc = self.markdown.serializer(div)
261			for pp in self.markdown.postprocessors.values():
262			toc = pp.run(toc)
263			self.markdown.toc = toc
264
265
266			class TocExtension(Extension):
267
268			TreeProcessorClass = TocTreeprocessor
269
270			def __init__(self, args, *kwargs):
271			self.config = {
272			"marker": ['[TOC]',
273			'Text to find and replace with Table of Contents - '
274			'Set to an empty string to disable. Defaults to "[TOC]"'],
275			"title": ["",
276			"Title to insert into TOC <div> - "
277			"Defaults to an empty string"],
278			"anchorlink": [False,
279			"True if header should be a self link - "
280			"Defaults to False"],
281			"permalink": [0,
282			"True or link text if a Sphinx-style permalink should "
283			"be added - Defaults to False"],
284			"baselevel": ['1', 'Base level for headers.'],
285			"slugify": [slugify,
286			"Function to generate anchors based on header text - "
287			"Defaults to the headerid ext's slugify function."],
288			'separator': ['-', 'Word separator. Defaults to "-".']
289			}
290
291			super(TocExtension, self).__init__(args, *kwargs)
292
293			def extendMarkdown(self, md, md_globals):
294			md.registerExtension(self)
295			self.md = md
296			self.reset()
297			tocext = self.TreeProcessorClass(md, self.getConfigs())
298			# Headerid ext is set to '>prettify'. With this set to '_end',
299			# it should always come after headerid ext (and honor ids assinged
300			# by the header id extension) if both are used. Same goes for
301			# attr_list extension. This must come last because we don't want
302			# to redefine ids after toc is created. But we do want toc prettified.
303			md.treeprocessors.add("toc", tocext, "_end")
304
305			def reset(self):
306			self.md.toc = ''
307
308
309			def makeExtension(args, *kwargs):
310			return TocExtension(args, *kwargs)
311

MikeCoder / markdown-preview.vim

Push — master ( 32cfa8...ec62d3 )

TocTreeprocessor.add_permalink() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like