1+ from bs4 import BeautifulSoup , Tag
2+ from docutils import nodes
3+ import os .path
4+
5+ known_start_tags = ['p' ,'img' ,'table' ]
6+ hidden_tag = 'hidden'
7+
8+ def html2Docutils (app , doctree ):
9+ #find all raw nodes
10+ if not app .env .config .sphinx_md_processRaw :
11+ return
12+ filepath = doctree ['source' ]
13+ htmlCounter = 0
14+ for node in doctree .traverse (nodes .raw ):
15+ soup = BeautifulSoup (node .astext (),features = "html.parser" )
16+ if soup .find ():
17+ if soup .find ().name in known_start_tags :
18+ #send it to converter
19+ div = nodes .container ()
20+ div ['id' ]= 'html-content-' + str (htmlCounter )
21+ htmlCounter += 1
22+ convertHTML (soup , div , app , filepath )
23+ parent = node .parent
24+ parent .replace (node ,div .children )
25+ #replace raw node with output of converter
26+ #child = nodes.Text("poof")
27+ #node[0]=child
28+ elif soup .find ().name == hidden_tag :
29+ hidden_comment = nodes .comment ()
30+ comment_text = nodes .Text ("hidden" )
31+ hidden_comment .append (comment_text )
32+ parent = node .parent
33+ parent .replace (node ,hidden_comment )
34+
35+ def convertHTML (soup , parent , app , filepath ):
36+ if hasattr (soup ,"children" ):
37+ for child in soup .children :
38+ node = None
39+ if hasattr (child ,"name" ) and child .name is not None :
40+ if child .name == "table" :
41+ if filepath not in app .env .config .sphinx_md_tableIDs :
42+ app .env .config .sphinx_md_tableIDs ['filepath' ]= 0
43+ else :
44+ app .evn .config .sphinx_md_tableIDs ['filepath' ] += 1
45+ tNode = nodes .table ()
46+ tNode ['ids' ].append ("id" + str (app .env .config .sphinx_md_tableIDs ['filepath' ]))
47+ titleNode = nodes .title ()
48+ node = nodes .tgroup ()
49+ ncols = getNumCols (child )
50+ node ['cols' ]= ncols
51+ for x in range (ncols ):
52+ colspecNode = nodes .colspec ()
53+ colspecNode ["colwidth" ]= 1
54+ node += colspecNode
55+ tNode += titleNode
56+ tNode += node
57+ parent += tNode
58+ elif child .name == "p" :
59+ node = nodes .paragraph ()
60+ parent += node
61+ elif child .name == "img" :
62+ node = nodes .image ()
63+ imgPath = ""
64+ if "alt" in child .attrs :
65+ node ["alt" ]= child .attrs ['alt' ]
66+ if "src" in child .attrs :
67+ if "https" in child .attrs ['src' ]:
68+ node ["uri" ]= child .attrs ['src' ]
69+ else :
70+ basepath = app .env .srcdir + "/"
71+ docfilename = os .path .splitext (os .path .relpath (filepath ,basepath ))[0 ]
72+ relpath = os .path .dirname (os .path .relpath (filepath ,basepath ))
73+ imgPath = os .path .join (relpath ,child .attrs ['src' ])
74+ node ["uri" ]= imgPath
75+ if os .path .isfile (imgPath ):
76+ if imgPath not in app .env .images :
77+ imageFileName = os .path .basename (imgPath )
78+ imageTuple = ({docfilename },imageFileName )
79+ app .env .images [imgPath ]= imageTuple
80+ if "width" in child .attrs :
81+ suffix = ''
82+ if child .attrs ['width' ].isnumeric ():
83+ suffix = 'px'
84+ node ["width" ]= child .attrs ['width' ] + suffix
85+ if "height" in child .attrs :
86+ node ["height" ]= child .attrs ['height' ]
87+ node ["candidates" ]= "{'*': '" + imgPath + "'}"
88+ parent += node
89+ elif child .name == "thead" :
90+ node = nodes .thead ()
91+ parent += node
92+ elif child .name == "tbody" :
93+ node = nodes .tbody ()
94+ parent += node
95+ elif child .name == "tr" :
96+ node = nodes .row ()
97+ parent += node
98+ elif child .name == "th" or child .name == "td" :
99+ eNode = nodes .entry ()
100+ node = nodes .paragraph ()
101+ eNode += node
102+ parent += eNode
103+ elif child .name == "sup" :
104+ node = nodes .superscript ()
105+ parent += node
106+ elif child .name == "a" :
107+ node = nodes .reference ()
108+ node ["refuri" ] = child .attrs ['href' ]
109+ parent += node
110+ elif child .name == "code" :
111+ node = nodes .literal ()
112+ parent += node
113+ else :
114+ if isinstance (parent ,nodes .Node ):
115+ #if isinstance(parent, nodes.entry) or isinstance(parent, nodes.paragraph) or isinstance(parent, nodes.image) or isinstance(parent, nodes.superscript) or isinstance(parent, nodes.reference) or isinstance(parent, nodes.literal):
116+ node = nodes .Text (child )
117+ parent += node
118+ if node :
119+ convertHTML (child ,node ,app ,filepath )
120+
121+ def removeHTMLAttributes (soup ,tagName ):
122+ tags = soup .find_all (tagName )
123+ for tag in tags :
124+ attList = []
125+ for attr in tag .attrs :
126+ attList .append (attr )
127+ for att in attList :
128+ del tag [att ]
129+ return soup
130+
131+ def replaceTag (soup ,oldTag ,newTag ,delAttrs = True ):
132+ tags = soup .find_all (oldTag )
133+ for tag in tags :
134+ tag .name = newTag
135+ if delAttrs :
136+ attList = []
137+ for attr in tag .attrs :
138+ attList .append (attr )
139+ for att in attList :
140+ del tag [att ]
141+ return soup
142+
143+ def fixImages (soup ):
144+ imgTags = soup .find_all ('img' )
145+ for imgTag in imgTags :
146+ altTag = soup .new_tag ("alt" )
147+ imgTag .name = "image"
148+ imgTag ['href' ]= imgTag ['src' ]
149+ del imgTag ['src' ]
150+ altTag .string = imgTag ['alt' ]
151+ del imgTag ['alt' ]
152+ imgTag .append (altTag )
153+ return soup
154+
155+
156+ def addTGroup (soup ):
157+ numCols = getNumCols (soup )
158+ tags = soup .find_all ('table' )
159+ for tableTag in tags :
160+ tableTag .name = 'tgroup'
161+ tableTag ['cols' ]= numCols
162+ del tableTag ['class' ]
163+ wrap (tableTag ,soup .new_tag ("table" ))
164+ return soup
165+
166+ def wrap (to_wrap , wrap_in ):
167+ contents = to_wrap .replace_with (wrap_in )
168+ wrap_in .append (contents )
169+
170+ def getNumCols (soup ):
171+ rows = soup .find_all ('th' )
172+ return len (rows )
0 commit comments