OLD | NEW |
| (Empty) |
1 from __future__ import unicode_literals | |
2 from __future__ import absolute_import | |
3 from . import util | |
4 from . import odict | |
5 | |
6 class State(list): | |
7 """ Track the current and nested state of the parser. | |
8 | |
9 This utility class is used to track the state of the BlockParser and | |
10 support multiple levels if nesting. It's just a simple API wrapped around | |
11 a list. Each time a state is set, that state is appended to the end of the | |
12 list. Each time a state is reset, that state is removed from the end of | |
13 the list. | |
14 | |
15 Therefore, each time a state is set for a nested block, that state must be | |
16 reset when we back out of that level of nesting or the state could be | |
17 corrupted. | |
18 | |
19 While all the methods of a list object are available, only the three | |
20 defined below need be used. | |
21 | |
22 """ | |
23 | |
24 def set(self, state): | |
25 """ Set a new state. """ | |
26 self.append(state) | |
27 | |
28 def reset(self): | |
29 """ Step back one step in nested state. """ | |
30 self.pop() | |
31 | |
32 def isstate(self, state): | |
33 """ Test that top (current) level is of given state. """ | |
34 if len(self): | |
35 return self[-1] == state | |
36 else: | |
37 return False | |
38 | |
39 class BlockParser: | |
40 """ Parse Markdown blocks into an ElementTree object. | |
41 | |
42 A wrapper class that stitches the various BlockProcessors together, | |
43 looping through them and creating an ElementTree object. | |
44 """ | |
45 | |
46 def __init__(self, markdown): | |
47 self.blockprocessors = odict.OrderedDict() | |
48 self.state = State() | |
49 self.markdown = markdown | |
50 | |
51 def parseDocument(self, lines): | |
52 """ Parse a markdown document into an ElementTree. | |
53 | |
54 Given a list of lines, an ElementTree object (not just a parent Element) | |
55 is created and the root element is passed to the parser as the parent. | |
56 The ElementTree object is returned. | |
57 | |
58 This should only be called on an entire document, not pieces. | |
59 | |
60 """ | |
61 # Create a ElementTree from the lines | |
62 self.root = util.etree.Element(self.markdown.doc_tag) | |
63 self.parseChunk(self.root, '\n'.join(lines)) | |
64 return util.etree.ElementTree(self.root) | |
65 | |
66 def parseChunk(self, parent, text): | |
67 """ Parse a chunk of markdown text and attach to given etree node. | |
68 | |
69 While the ``text`` argument is generally assumed to contain multiple | |
70 blocks which will be split on blank lines, it could contain only one | |
71 block. Generally, this method would be called by extensions when | |
72 block parsing is required. | |
73 | |
74 The ``parent`` etree Element passed in is altered in place. | |
75 Nothing is returned. | |
76 | |
77 """ | |
78 self.parseBlocks(parent, text.split('\n\n')) | |
79 | |
80 def parseBlocks(self, parent, blocks): | |
81 """ Process blocks of markdown text and attach to given etree node. | |
82 | |
83 Given a list of ``blocks``, each blockprocessor is stepped through | |
84 until there are no blocks left. While an extension could potentially | |
85 call this method directly, it's generally expected to be used internally
. | |
86 | |
87 This is a public method as an extension may need to add/alter additional | |
88 BlockProcessors which call this method to recursively parse a nested | |
89 block. | |
90 | |
91 """ | |
92 while blocks: | |
93 for processor in self.blockprocessors.values(): | |
94 if processor.test(parent, blocks[0]): | |
95 if processor.run(parent, blocks) is not False: | |
96 # run returns True or None | |
97 break | |
98 | |
99 | |
OLD | NEW |