Show More
@@ -1,214 +1,217 | |||
|
1 | 1 | """Markdown filters |
|
2 | ||
|
2 | 3 | This file contains a collection of utility filters for dealing with |
|
3 | 4 | markdown within Jinja templates. |
|
4 | 5 | """ |
|
5 | #----------------------------------------------------------------------------- | |
|
6 | # Copyright (c) 2013, the IPython Development Team. | |
|
7 | # | |
|
6 | # Copyright (c) IPython Development Team. | |
|
8 | 7 | # Distributed under the terms of the Modified BSD License. |
|
9 | # | |
|
10 | # The full license is in the file COPYING.txt, distributed with this software. | |
|
11 | #----------------------------------------------------------------------------- | |
|
12 | 8 | |
|
13 | #----------------------------------------------------------------------------- | |
|
14 | # Imports | |
|
15 | #----------------------------------------------------------------------------- | |
|
16 | 9 | from __future__ import print_function |
|
17 | 10 | |
|
18 | 11 | # Stdlib imports |
|
19 | 12 | import os |
|
20 | 13 | import subprocess |
|
21 | 14 | from io import TextIOWrapper, BytesIO |
|
22 | 15 | import re |
|
23 | 16 | |
|
24 | 17 | import mistune |
|
25 | 18 | from pygments import highlight |
|
26 | 19 | from pygments.lexers import get_lexer_by_name |
|
27 | 20 | from pygments.formatters import HtmlFormatter |
|
28 | 21 | from pygments.util import ClassNotFound |
|
29 | 22 | |
|
30 | 23 | # IPython imports |
|
31 | 24 | from IPython.nbconvert.utils.pandoc import pandoc |
|
32 | 25 | from IPython.nbconvert.utils.exceptions import ConversionException |
|
33 | 26 | from IPython.utils.process import get_output_error_code |
|
34 | 27 | from IPython.utils.py3compat import cast_bytes |
|
35 | 28 | from IPython.utils.version import check_version |
|
36 | 29 | |
|
37 | #----------------------------------------------------------------------------- | |
|
38 | # Functions | |
|
39 | #----------------------------------------------------------------------------- | |
|
30 | ||
|
40 | 31 | marked = os.path.join(os.path.dirname(__file__), "marked.js") |
|
41 | 32 | _node = None |
|
42 | 33 | |
|
43 | 34 | __all__ = [ |
|
44 | 35 | 'markdown2html', |
|
45 | 36 | 'markdown2html_pandoc', |
|
46 | 37 | 'markdown2html_marked', |
|
47 | 38 | 'markdown2html_mistune', |
|
48 | 39 | 'markdown2latex', |
|
49 | 40 | 'markdown2rst', |
|
50 | 41 | ] |
|
51 | 42 | |
|
52 | 43 | class NodeJSMissing(ConversionException): |
|
53 | 44 | """Exception raised when node.js is missing.""" |
|
54 | 45 | pass |
|
55 | 46 | |
|
56 | 47 | def markdown2latex(source): |
|
57 | 48 | """Convert a markdown string to LaTeX via pandoc. |
|
58 | 49 | |
|
59 | 50 | This function will raise an error if pandoc is not installed. |
|
60 | 51 | Any error messages generated by pandoc are printed to stderr. |
|
61 | 52 | |
|
62 | 53 | Parameters |
|
63 | 54 | ---------- |
|
64 | 55 | source : string |
|
65 | 56 | Input string, assumed to be valid markdown. |
|
66 | 57 | |
|
67 | 58 | Returns |
|
68 | 59 | ------- |
|
69 | 60 | out : string |
|
70 | 61 | Output as returned by pandoc. |
|
71 | 62 | """ |
|
72 | 63 | return pandoc(source, 'markdown', 'latex') |
|
73 | 64 | |
|
74 | 65 | class MathBlockGrammar(mistune.BlockGrammar): |
|
75 | block_math = re.compile("^\$\$(.*?)\$\$") | |
|
76 |
|
|
|
66 | block_math = re.compile("^\$\$(.*?)\$\$", re.DOTALL) | |
|
67 | latex_environment = re.compile(r"^\\begin\{([a-z]*\*?)\}(.*?)\\end\{\1\}", | |
|
68 | re.DOTALL) | |
|
77 | 69 | |
|
78 | 70 | class MathBlockLexer(mistune.BlockLexer): |
|
79 |
default_features = ['block_math', ' |
|
|
71 | default_features = ['block_math', 'latex_environment'] + mistune.BlockLexer.default_features | |
|
80 | 72 | |
|
81 | 73 | def __init__(self, rules=None, **kwargs): |
|
82 | 74 | if rules is None: |
|
83 | 75 | rules = MathBlockGrammar() |
|
84 | 76 | super(MathBlockLexer, self).__init__(rules, **kwargs) |
|
85 | 77 | |
|
86 | 78 | def parse_block_math(self, m): |
|
87 | 79 | """Parse a $$math$$ block""" |
|
88 | 80 | self.tokens.append({ |
|
89 | 81 | 'type': 'block_math', |
|
90 | 82 | 'text': m.group(1) |
|
91 | 83 | }) |
|
92 | 84 | |
|
93 | parse_block_math2 = parse_block_math | |
|
85 | def parse_latex_environment(self, m): | |
|
86 | self.tokens.append({ | |
|
87 | 'type': 'latex_environment', | |
|
88 | 'name': m.group(1), | |
|
89 | 'text': m.group(2) | |
|
90 | }) | |
|
94 | 91 | |
|
95 | 92 | class MathInlineGrammar(mistune.InlineGrammar): |
|
96 | 93 | math = re.compile("^\$(.+?)\$") |
|
97 | 94 | |
|
98 | 95 | class MathInlineLexer(mistune.InlineLexer): |
|
99 | 96 | default_features = ['math'] + mistune.InlineLexer.default_features |
|
100 | 97 | |
|
101 | 98 | def __init__(self, renderer, rules=None, **kwargs): |
|
102 | 99 | if rules is None: |
|
103 | 100 | rules = MathInlineGrammar() |
|
104 | 101 | super(MathInlineLexer, self).__init__(renderer, rules, **kwargs) |
|
105 | 102 | |
|
106 | 103 | def output_math(self, m): |
|
107 | self.renderer.inline_math(m.group(1)) | |
|
104 | return self.renderer.inline_math(m.group(1)) | |
|
108 | 105 | |
|
109 | 106 | class MarkdownWithMath(mistune.Markdown): |
|
110 | 107 | def __init__(self, renderer, **kwargs): |
|
111 | 108 | if 'inline' not in kwargs: |
|
112 | 109 | kwargs['inline'] = MathInlineLexer(renderer, **kwargs) |
|
113 | 110 | if 'block' not in kwargs: |
|
114 | 111 | kwargs['block'] = MathBlockLexer(**kwargs) |
|
115 | 112 | super(MarkdownWithMath, self).__init__(renderer, **kwargs) |
|
116 | 113 | |
|
117 | 114 | def parse_block_math(self): |
|
118 | 115 | return self.renderer.block_math(self.token['text']) |
|
119 | 116 | |
|
117 | def parse_latex_environment(self): | |
|
118 | return self.renderer.latex_environment(self.token['name'], self.token['text']) | |
|
119 | ||
|
120 | 120 | class IPythonRenderer(mistune.Renderer): |
|
121 | 121 | def block_code(self, code, lang): |
|
122 | 122 | if lang: |
|
123 | 123 | try: |
|
124 | 124 | lexer = get_lexer_by_name(lang, stripall=True) |
|
125 | 125 | except ClassNotFound: |
|
126 | 126 | code = lang + '\n' + code |
|
127 | 127 | lang = None |
|
128 | 128 | |
|
129 | 129 | if not lang: |
|
130 | 130 | return '\n<pre><code>%s</code></pre>\n' % \ |
|
131 | 131 | mistune.escape(code) |
|
132 | 132 | |
|
133 | 133 | formatter = HtmlFormatter() |
|
134 | 134 | return highlight(code, lexer, formatter) |
|
135 | 135 | |
|
136 | 136 | # Pass math through unaltered - mathjax does the rendering in the browser |
|
137 | 137 | def block_math(self, text): |
|
138 | 138 | return '$$%s$$' % text |
|
139 | 139 | |
|
140 | def latex_environment(self, name, text): | |
|
141 | return r'\begin{%s}%s\end{%s}' % (name, text, name) | |
|
142 | ||
|
140 | 143 | def inline_math(self, text): |
|
141 | 144 | return '$%s$' % text |
|
142 | 145 | |
|
143 | 146 | def markdown2html_mistune(source): |
|
144 | 147 | """Convert a markdown string to HTML using mistune""" |
|
145 | 148 | return MarkdownWithMath(renderer=IPythonRenderer()).render(source) |
|
146 | 149 | |
|
147 | 150 | def markdown2html_pandoc(source): |
|
148 | 151 | """Convert a markdown string to HTML via pandoc""" |
|
149 | 152 | return pandoc(source, 'markdown', 'html', extra_args=['--mathjax']) |
|
150 | 153 | |
|
151 | 154 | def _find_nodejs(): |
|
152 | 155 | global _node |
|
153 | 156 | if _node is None: |
|
154 | 157 | # prefer md2html via marked if node.js >= 0.9.12 is available |
|
155 | 158 | # node is called nodejs on debian, so try that first |
|
156 | 159 | _node = 'nodejs' |
|
157 | 160 | if not _verify_node(_node): |
|
158 | 161 | _node = 'node' |
|
159 | 162 | return _node |
|
160 | 163 | |
|
161 | 164 | def markdown2html_marked(source, encoding='utf-8'): |
|
162 | 165 | """Convert a markdown string to HTML via marked""" |
|
163 | 166 | command = [_find_nodejs(), marked] |
|
164 | 167 | try: |
|
165 | 168 | p = subprocess.Popen(command, |
|
166 | 169 | stdin=subprocess.PIPE, stdout=subprocess.PIPE |
|
167 | 170 | ) |
|
168 | 171 | except OSError as e: |
|
169 | 172 | raise NodeJSMissing( |
|
170 | 173 | "The command '%s' returned an error: %s.\n" % (" ".join(command), e) + |
|
171 | 174 | "Please check that Node.js is installed." |
|
172 | 175 | ) |
|
173 | 176 | out, _ = p.communicate(cast_bytes(source, encoding)) |
|
174 | 177 | out = TextIOWrapper(BytesIO(out), encoding, 'replace').read() |
|
175 | 178 | return out.rstrip('\n') |
|
176 | 179 | |
|
177 | 180 | # The mistune renderer is the default, because it's simple to depend on it |
|
178 | 181 | markdown2html = markdown2html_mistune |
|
179 | 182 | |
|
180 | 183 | def markdown2rst(source): |
|
181 | 184 | """Convert a markdown string to ReST via pandoc. |
|
182 | 185 | |
|
183 | 186 | This function will raise an error if pandoc is not installed. |
|
184 | 187 | Any error messages generated by pandoc are printed to stderr. |
|
185 | 188 | |
|
186 | 189 | Parameters |
|
187 | 190 | ---------- |
|
188 | 191 | source : string |
|
189 | 192 | Input string, assumed to be valid markdown. |
|
190 | 193 | |
|
191 | 194 | Returns |
|
192 | 195 | ------- |
|
193 | 196 | out : string |
|
194 | 197 | Output as returned by pandoc. |
|
195 | 198 | """ |
|
196 | 199 | return pandoc(source, 'markdown', 'rst') |
|
197 | 200 | |
|
198 | 201 | def _verify_node(cmd): |
|
199 | 202 | """Verify that the node command exists and is at least the minimum supported |
|
200 | 203 | version of node. |
|
201 | 204 | |
|
202 | 205 | Parameters |
|
203 | 206 | ---------- |
|
204 | 207 | cmd : string |
|
205 | 208 | Node command to verify (i.e 'node').""" |
|
206 | 209 | try: |
|
207 | 210 | out, err, return_code = get_output_error_code([cmd, '--version']) |
|
208 | 211 | except OSError: |
|
209 | 212 | # Command not found |
|
210 | 213 | return False |
|
211 | 214 | if return_code: |
|
212 | 215 | # Command error |
|
213 | 216 | return False |
|
214 | 217 | return check_version(out.lstrip('v'), '0.9.12') |
@@ -1,91 +1,88 | |||
|
1 | """Tests for conversions from markdown to other formats""" | |
|
1 | 2 | |
|
2 | """ | |
|
3 | Module with tests for Markdown | |
|
4 | """ | |
|
5 | ||
|
6 | #----------------------------------------------------------------------------- | |
|
7 | # Copyright (c) 2013, the IPython Development Team. | |
|
8 | # | |
|
3 | # Copyright (c) IPython Development Team. | |
|
9 | 4 | # Distributed under the terms of the Modified BSD License. |
|
10 | # | |
|
11 | # The full license is in the file COPYING.txt, distributed with this software. | |
|
12 | #----------------------------------------------------------------------------- | |
|
13 | ||
|
14 | #----------------------------------------------------------------------------- | |
|
15 | # Imports | |
|
16 | #----------------------------------------------------------------------------- | |
|
17 | 5 | |
|
18 | 6 | from copy import copy |
|
19 | 7 | |
|
20 | 8 | from IPython.utils.py3compat import string_types |
|
21 | 9 | from IPython.testing import decorators as dec |
|
22 | 10 | |
|
23 | 11 | from ...tests.base import TestsBase |
|
24 | 12 | from ..markdown import markdown2latex, markdown2html, markdown2rst |
|
25 | 13 | |
|
26 | 14 | |
|
27 | #----------------------------------------------------------------------------- | |
|
28 | # Class | |
|
29 | #----------------------------------------------------------------------------- | |
|
30 | ||
|
31 | 15 | class TestMarkdown(TestsBase): |
|
32 | 16 | |
|
33 | 17 | tests = [ |
|
34 | 18 | '*test', |
|
35 | 19 | '**test', |
|
36 | 20 | '*test*', |
|
37 | 21 | '_test_', |
|
38 | 22 | '__test__', |
|
39 | 23 | '__*test*__', |
|
40 | 24 | '**test**', |
|
41 | 25 | '#test', |
|
42 | 26 | '##test', |
|
43 | 27 | 'test\n----', |
|
44 | 28 | 'test [link](https://google.com/)'] |
|
45 | 29 | |
|
46 | 30 | tokens = [ |
|
47 | 31 | '*test', |
|
48 | 32 | '**test', |
|
49 | 33 | 'test', |
|
50 | 34 | 'test', |
|
51 | 35 | 'test', |
|
52 | 36 | 'test', |
|
53 | 37 | 'test', |
|
54 | 38 | 'test', |
|
55 | 39 | 'test', |
|
56 | 40 | 'test', |
|
57 | 41 | ('test', 'https://google.com/')] |
|
58 | 42 | |
|
59 | 43 | |
|
60 | 44 | @dec.onlyif_cmds_exist('pandoc') |
|
61 | 45 | def test_markdown2latex(self): |
|
62 | 46 | """markdown2latex test""" |
|
63 | 47 | for index, test in enumerate(self.tests): |
|
64 | 48 | self._try_markdown(markdown2latex, test, self.tokens[index]) |
|
65 | 49 | |
|
66 | 50 | def test_markdown2html(self): |
|
67 | 51 | """markdown2html test""" |
|
68 | 52 | for index, test in enumerate(self.tests): |
|
69 | 53 | self._try_markdown(markdown2html, test, self.tokens[index]) |
|
70 | 54 | |
|
55 | def test_markdown2html_math(self): | |
|
56 | # Mathematical expressions should be passed through unaltered | |
|
57 | cases = [("\\begin{equation*}\n" | |
|
58 | "\\left( \\sum_{k=1}^n a_k b_k \\right)^2 \\leq \\left( \\sum_{k=1}^n a_k^2 \\right) \\left( \\sum_{k=1}^n b_k^2 \\right)\n" | |
|
59 | "\\end{equation*}"), | |
|
60 | ("$$\n" | |
|
61 | "a = 1 *3* 5\n" | |
|
62 | "$$"), | |
|
63 | "$ a = 1 *3* 5 $", | |
|
64 | ] | |
|
65 | for case in cases: | |
|
66 | self.assertIn(case, markdown2html(case)) | |
|
67 | ||
|
71 | 68 | |
|
72 | 69 | @dec.onlyif_cmds_exist('pandoc') |
|
73 | 70 | def test_markdown2rst(self): |
|
74 | 71 | """markdown2rst test""" |
|
75 | 72 | |
|
76 | 73 | #Modify token array for rst, escape asterik |
|
77 | 74 | tokens = copy(self.tokens) |
|
78 | 75 | tokens[0] = r'\*test' |
|
79 | 76 | tokens[1] = r'\*\*test' |
|
80 | 77 | |
|
81 | 78 | for index, test in enumerate(self.tests): |
|
82 | 79 | self._try_markdown(markdown2rst, test, tokens[index]) |
|
83 | 80 | |
|
84 | 81 | |
|
85 | 82 | def _try_markdown(self, method, test, tokens): |
|
86 | 83 | results = method(test) |
|
87 | 84 | if isinstance(tokens, string_types): |
|
88 | 85 | assert tokens in results |
|
89 | 86 | else: |
|
90 | 87 | for token in tokens: |
|
91 | 88 | assert token in results |
General Comments 0
You need to be logged in to leave comments.
Login now