Line data Source code
1 : /* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
2 : /*
3 : * This file is part of the LibreOffice project.
4 : *
5 : * This Source Code Form is subject to the terms of the Mozilla Public
6 : * License, v. 2.0. If a copy of the MPL was not distributed with this
7 : * file, You can obtain one at http://mozilla.org/MPL/2.0/.
8 : */
9 :
10 : #include <BasCodeTagger.hxx>
11 :
12 0 : LibXmlTreeWalker::LibXmlTreeWalker( xmlDocPtr doc )
13 : {
14 0 : if ( doc == NULL )
15 0 : throw BasicCodeTagger::NULL_DOCUMENT;
16 0 : m_pCurrentNode = xmlDocGetRootElement( doc );
17 0 : if ( m_pCurrentNode == NULL )
18 0 : throw BasicCodeTagger::EMPTY_DOCUMENT;
19 0 : else if ( m_pCurrentNode->xmlChildrenNode != NULL )
20 0 : m_Queue.push_back( m_pCurrentNode->xmlChildrenNode );
21 0 : nextNode();
22 0 : }
23 :
24 0 : void LibXmlTreeWalker::nextNode()
25 : {
26 :
27 : //next node
28 0 : if ( m_pCurrentNode->next == NULL )
29 : {
30 0 : m_pCurrentNode = m_Queue.front();
31 0 : m_Queue.pop_front();
32 : }
33 : else
34 0 : m_pCurrentNode = m_pCurrentNode->next;
35 : //queue chiledren if they exist
36 0 : if ( m_pCurrentNode->xmlChildrenNode != NULL )
37 0 : m_Queue.push_back( m_pCurrentNode->xmlChildrenNode );
38 0 : }
39 :
40 0 : void LibXmlTreeWalker::ignoreCurrNodesChildren()
41 : {
42 0 : if ( m_pCurrentNode->xmlChildrenNode != NULL )
43 0 : m_Queue.pop_back();
44 0 : }
45 :
46 0 : bool LibXmlTreeWalker::end()
47 : {
48 0 : return m_pCurrentNode->next == NULL && m_Queue.empty();
49 : }
50 :
51 :
52 :
53 :
54 0 : BasicCodeTagger::BasicCodeTagger( xmlDocPtr rootDoc ):
55 0 : m_Highlighter(HIGHLIGHT_BASIC)
56 : {
57 0 : if ( rootDoc == NULL )
58 0 : throw NULL_DOCUMENT;
59 0 : m_pDocument = rootDoc;
60 0 : m_pXmlTreeWalker = NULL;
61 0 : m_bTaggingCompleted = false;
62 :
63 0 : }
64 :
65 0 : BasicCodeTagger::~BasicCodeTagger()
66 : {
67 0 : if ( m_pXmlTreeWalker != NULL )
68 0 : delete m_pXmlTreeWalker;
69 0 : }
70 : //!Gathers all the <bascode> tag nodes from xml tree.
71 : /*!
72 : * Assumes m_pDocument is valid. Handles m_pXmlTreeWalker and m_BasicCodeContainerTags members.
73 : */
74 0 : void BasicCodeTagger::getBasicCodeContainerNodes()
75 : {
76 : xmlNodePtr currentNode;
77 :
78 0 : m_BasicCodeContainerTags.clear();
79 :
80 0 : if ( m_pXmlTreeWalker != NULL )
81 0 : delete m_pXmlTreeWalker;
82 0 : m_pXmlTreeWalker = new LibXmlTreeWalker( m_pDocument );
83 :
84 0 : currentNode = m_pXmlTreeWalker->currentNode();
85 0 : if ( !( xmlStrcmp( currentNode->name, (const xmlChar*) "bascode" ) ) )
86 : { //Found <bascode>
87 0 : m_BasicCodeContainerTags.push_back( currentNode ); //it goes to the end of the list
88 : }
89 0 : while ( !m_pXmlTreeWalker->end() )
90 : {
91 0 : m_pXmlTreeWalker->nextNode();
92 0 : if ( !( xmlStrcmp( m_pXmlTreeWalker->currentNode()->name, (const xmlChar*) "bascode" ) ) )
93 : { //Found <bascode>
94 0 : m_BasicCodeContainerTags.push_back( m_pXmlTreeWalker->currentNode() ); //it goes to the end of the list
95 0 : m_pXmlTreeWalker->ignoreCurrNodesChildren();
96 : }
97 : }
98 0 : }
99 :
100 : //! Extracts Basic Codes containted in <bascode> tags.
101 : /*!
102 : * For each <bascode> this method iterates trough it's <paragraph> tags and "inserts" <item> tags according
103 : * to the Basic code syntax found in that paragraph.
104 : */
105 0 : void BasicCodeTagger::tagBasCodeParagraphs()
106 : {
107 : //helper variables
108 : xmlNodePtr currBascodeNode;
109 : xmlNodePtr currParagraph;
110 0 : while ( !m_BasicCodeContainerTags.empty() )
111 : {
112 0 : currBascodeNode = m_BasicCodeContainerTags.front();
113 0 : currParagraph = currBascodeNode->xmlChildrenNode; //first <paragraph>
114 0 : while ( currParagraph != NULL )
115 : {
116 0 : tagParagraph( currParagraph );
117 0 : currParagraph=currParagraph->next;
118 : }
119 0 : m_BasicCodeContainerTags.pop_front(); //next element
120 : }
121 0 : }
122 :
123 : //! Used by tagBasCodeParagraphs(). It does the work on the current paragraph containing Basic code.
124 0 : void BasicCodeTagger::tagParagraph( xmlNodePtr paragraph )
125 : {
126 : //1. get paragraph text
127 : xmlChar* codeSnippet;
128 0 : codeSnippet = xmlNodeListGetString( m_pDocument, paragraph->xmlChildrenNode, 1 );
129 0 : if ( codeSnippet == NULL )
130 : {
131 0 : return; //no text, nothing more to do here
132 : }
133 : //2. delete every child from paragraph (except attributes)
134 0 : xmlNodePtr curNode = paragraph->xmlChildrenNode;
135 : xmlNodePtr sibling;
136 0 : while ( curNode != NULL )
137 : {
138 0 : sibling = curNode->next;
139 0 : xmlUnlinkNode( curNode );
140 0 : xmlFreeNode( curNode );
141 0 : curNode = sibling;
142 : }
143 :
144 : //3. create new paragraph content
145 : OUString strLine( reinterpret_cast<const sal_Char*>(codeSnippet),
146 0 : strlen(reinterpret_cast<const char*>(codeSnippet)),
147 0 : RTL_TEXTENCODING_UTF8 );
148 0 : std::vector<HighlightPortion> portions;
149 0 : m_Highlighter.getHighlightPortions( strLine, portions );
150 0 : for (std::vector<HighlightPortion>::iterator i(portions.begin());
151 0 : i != portions.end(); ++i)
152 : {
153 0 : OString sToken(OUStringToOString(strLine.copy(i->nBegin, i->nEnd-i->nBegin), RTL_TEXTENCODING_UTF8));
154 0 : xmlNodePtr text = xmlNewText((const xmlChar*)sToken.getStr());
155 0 : if ( i->tokenType != TT_WHITESPACE )
156 : {
157 0 : xmlChar* typeStr = getTypeString( i->tokenType );
158 0 : curNode = xmlNewTextChild( paragraph, 0, (xmlChar*)"item", 0 );
159 0 : xmlNewProp( curNode, (xmlChar*)"type", typeStr );
160 0 : xmlAddChild( curNode, text );
161 0 : xmlFree( typeStr );
162 : }
163 : else
164 0 : xmlAddChild( paragraph, text );
165 0 : }
166 0 : xmlFree( codeSnippet );
167 : }
168 :
169 : //! Manages tagging process.
170 : /*!
171 : * This is the "main" function of BasicCodeTagger.
172 : */
173 0 : void BasicCodeTagger::tagBasicCodes()
174 : {
175 0 : if ( m_bTaggingCompleted )
176 0 : return;
177 : //gather <bascode> nodes
178 : try
179 : {
180 0 : getBasicCodeContainerNodes();
181 : }
182 0 : catch (TaggerException &ex)
183 : {
184 0 : std::cout << "BasCodeTagger error occurred. Error code:" << ex << std::endl;
185 : }
186 :
187 : //tag basic code paragraphs in <bascode> tag
188 0 : tagBasCodeParagraphs();
189 0 : m_bTaggingCompleted = true;
190 : }
191 :
192 : //! Converts SyntaxHighlighter's TokenTypes enum to a type string for <item type=... >
193 0 : xmlChar* BasicCodeTagger::getTypeString( TokenTypes tokenType )
194 : {
195 : const char* str;
196 0 : switch ( tokenType )
197 : {
198 : case TT_UNKNOWN :
199 0 : str = "unknown";
200 0 : break;
201 : case TT_IDENTIFIER :
202 0 : str = "identifier";
203 0 : break;
204 : case TT_WHITESPACE :
205 0 : str = "whitespace";
206 0 : break;
207 : case TT_NUMBER :
208 0 : str = "number";
209 0 : break;
210 : case TT_STRING :
211 0 : str = "string";
212 0 : break;
213 : case TT_EOL :
214 0 : str = "eol";
215 0 : break;
216 : case TT_COMMENT :
217 0 : str = "comment";
218 0 : break;
219 : case TT_ERROR :
220 0 : str = "error";
221 0 : break;
222 : case TT_OPERATOR :
223 0 : str = "operator";
224 0 : break;
225 : case TT_KEYWORDS :
226 0 : str = "keyword";
227 0 : break;
228 : case TT_PARAMETER :
229 0 : str = "parameter";
230 0 : break;
231 : default :
232 0 : str = "unknown";
233 0 : break;
234 : }
235 0 : return xmlCharStrdup( str );
236 426 : }
237 :
238 : /* vim:set shiftwidth=4 softtabstop=4 expandtab: */
|
