1 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
2 // -*- mode: C++ -*-
3 //
4 // Copyright (C) 2013-2022 Red Hat, Inc.
5
6 /// @file
7
8 #include <string>
9 #include <iostream>
10
11 #include "abg-internal.h"
12 // <headers defining libabigail's API go under here>
13 ABG_BEGIN_EXPORT_DECLARATIONS
14
15 #include "abg-libxml-utils.h"
16
17 ABG_END_EXPORT_DECLARATIONS
18 // </headers defining libabigail's API>
19
20 namespace abigail
21 {
22
23 namespace sptr_utils
24 {
25 /// Build and return a shared_ptr for a pointer to xmlTextReader
26 template<>
27 shared_ptr<xmlTextReader>
build_sptr(::xmlTextReader * p)28 build_sptr<xmlTextReader>(::xmlTextReader *p)
29 {
30 return shared_ptr<xmlTextReader>(p, abigail::xml::textReaderDeleter());
31 }
32
33 /// Build and return a shared_ptr for a pointer to xmlChar
34 template<>
35 shared_ptr<xmlChar>
build_sptr(xmlChar * p)36 build_sptr<xmlChar>(xmlChar *p)
37 {
38 return shared_ptr<xmlChar>(p, abigail::xml::charDeleter());
39 }
40
41 }//end namespace sptr_utils
42
43 namespace xml
44 {
45 using std::istream;
46
47 /// Instantiate an xmlTextReader that parses the content of an on-disk
48 /// file, wrap it into a smart pointer and return it.
49 ///
50 /// @param path the path to the file to be parsed by the returned
51 /// instance of xmlTextReader.
52 reader_sptr
new_reader_from_file(const std::string & path)53 new_reader_from_file(const std::string& path)
54 {
55 reader_sptr p =
56 build_sptr(xmlNewTextReaderFilename (path.c_str()));
57
58 return p;
59 }
60
61 /// Instanciate an xmlTextReader that parses the content of an
62 /// in-memory buffer, wrap it into a smart pointer and return it.
63 ///
64 /// @param buffer the in-memory buffer to be parsed by the returned
65 /// instance of xmlTextReader.
66 reader_sptr
new_reader_from_buffer(const std::string & buffer)67 new_reader_from_buffer(const std::string& buffer)
68 {
69 reader_sptr p =
70 build_sptr(xmlReaderForMemory(buffer.c_str(),
71 buffer.length(),
72 "", 0, 0));
73 return p;
74 }
75
76 /// This is an xmlInputReadCallback, meant to be passed to
77 /// xmlNewTextReaderForIO. It reads a number of bytes from an istream.
78 ///
79 /// @param context an std::istream* cast into a void*. This is the
80 /// istream that the xmlTextReader is too read data from.
81 ///
82 /// @param buffer the buffer where to copy the data read from the
83 /// input stream.
84 ///
85 /// @param len the number of byte to read from the input stream and to
86 /// copy into @p buffer.
87 ///
88 /// @return the number of bytes read or -1 in case of error.
89 static int
xml_istream_input_read(void * context,char * buffer,int len)90 xml_istream_input_read(void* context,
91 char* buffer,
92 int len)
93 {
94 istream* in = reinterpret_cast<istream*>(context);
95 in->read(buffer, len);
96 return in->gcount();
97 }
98
99 /// This is an xmlInputCloseCallback, meant to be passed to
100 /// xmlNewTextReaderForIO. It's supposed to close the input stream
101 /// that the xmlTextReader is reading from. This particular
102 /// implementation is noop; it does nothing.
103 ///
104 /// @return 0.
105 static int
xml_istream_input_close(void *)106 xml_istream_input_close(void*)
107 {return 0;}
108
109 /// Instanciate an xmlTextReader that parses a content coming from an
110 /// input stream.
111 ///
112 /// @param in the input stream to consider.
113 ///
114 /// @return reader_sptr a pointer to the newly instantiated xml
115 /// reader.
116 reader_sptr
new_reader_from_istream(std::istream * in)117 new_reader_from_istream(std::istream* in)
118 {
119 reader_sptr p =
120 build_sptr(xmlReaderForIO(&xml_istream_input_read,
121 &xml_istream_input_close,
122 in, "", 0, 0));
123 return p;
124 }
125
126 /// Convert a shared pointer to xmlChar into an std::string.
127 ///
128 /// If the xmlChar is NULL, set "" to the string.
129 ///
130 /// @param ssptr the shared point to xmlChar to convert.
131 ///
132 /// @param s the output string.
133 ///
134 /// @return true if the shared pointer to xmlChar contained a non NULL
135 /// string, false otherwise.
136 bool
xml_char_sptr_to_string(xml_char_sptr ssptr,std::string & s)137 xml_char_sptr_to_string(xml_char_sptr ssptr, std::string& s)
138 {
139 bool non_nil = false;
140 if (CHAR_STR(ssptr))
141 {
142 s = CHAR_STR(ssptr);
143 non_nil = true;
144 }
145 else
146 {
147 s = "";
148 non_nil = false;
149 }
150
151 return non_nil;
152 }
153
154 /// Return the depth of an xml element node.
155 ///
156 /// Note that the node must be attached to an XML document.
157 ///
158 /// @param n the xml to consider.
159 ///
160 /// @return a positive or zero number for an XML node properly
161 /// attached to an xml document, -1 otherwise. Note that the function
162 /// returns -1 if passed an xml document as well.
163 int
get_xml_node_depth(xmlNodePtr n)164 get_xml_node_depth(xmlNodePtr n)
165 {
166 if (n->type == XML_DOCUMENT_NODE || n->parent == NULL)
167 return -1;
168
169 if (n->parent->type == XML_DOCUMENT_NODE)
170 return 0;
171
172 return 1 + get_xml_node_depth(n->parent);
173 }
174
175 /// Escape the 5 characters representing the predefined XML entities.
176 ///
177 /// The resulting entities and their matching characters are:
178 ///
179 /// < for the character '<', > for the character '>', ' for
180 /// the character ''', " for the character '"', and & for the
181 /// character '&'.
182 ///
183 //// @param str the input string to read to search for the characters
184 //// to escape.
185 ////
186 //// @param escaped the output string where to write the resulting
187 //// string that contains the pre-defined characters escaped as
188 //// predefined entitites.
189 void
escape_xml_string(const std::string & str,std::string & escaped)190 escape_xml_string(const std::string& str,
191 std::string& escaped)
192 {
193 for (std::string::const_iterator i = str.begin(); i != str.end(); ++i)
194 switch (*i)
195 {
196 case '<':
197 escaped += "<";
198 break;
199 case '>':
200 escaped += ">";
201 break;
202 case '&':
203 escaped += "&";
204 break;
205 case '\'':
206 escaped += "'";
207 break;
208 case '"':
209 escaped += """;
210 break;
211 default:
212 escaped += *i;
213 }
214 }
215
216 /// Escape the 5 characters representing the predefined XML entities.
217 ///
218 /// The resulting entities and their matching characters are:
219 ///
220 /// < for the character '<', > for the character '>', ' for
221 /// the character ''', " for the character '"', and & for the
222 /// character '&'.
223 ///
224 //// @param str the input string to read to search for the characters
225 //// to escape.
226 ////
227 //// @return the resulting string that contains the pre-defined
228 //// characters escaped as predefined entitites.
229 std::string
escape_xml_string(const std::string & str)230 escape_xml_string(const std::string& str)
231 {
232 std::string result;
233 escape_xml_string(str, result);
234 return result;
235 }
236
237 /// Escape the '-' character, to avoid having a '--' in a comment.
238 ///
239 /// The resulting entity for '-' is '-'.
240 ///
241 //// @param str the input string to read to search for the characters
242 //// to escape.
243 ////
244 //// @param escaped the output string where to write the resulting
245 //// string that contains the pre-defined characters escaped as
246 //// predefined entitites.
247 void
escape_xml_comment(const std::string & str,std::string & escaped)248 escape_xml_comment(const std::string& str,
249 std::string& escaped)
250 {
251 for (std::string::const_iterator i = str.begin(); i != str.end(); ++i)
252 switch (*i)
253 {
254 case '-':
255 escaped += "-";
256 break;
257 default:
258 escaped += *i;
259 }
260 }
261
262 /// Escape the '-' character, to avoid having a '--' in a comment.
263 ///
264 /// The resulting entity for '-' is '-'.
265 ///
266 //// @param str the input string to read to search for the characters
267 //// to escape.
268 ////
269 //// @return the resulting string that contains the pre-defined
270 //// characters escaped as predefined entitites.
271 std::string
escape_xml_comment(const std::string & str)272 escape_xml_comment(const std::string& str)
273 {
274 std::string result;
275 escape_xml_comment(str, result);
276 return result;
277 }
278
279 /// Read a string, detect the 5 predefined XML entities it may contain
280 /// and un-escape them, by writting their corresponding characters
281 /// back in. The pre-defined entities are:
282 ///
283 /// < for the character '<', > for the character '>', ' for
284 /// the character ''', " for the character '"', and & for the
285 /// character '&'.
286 ///
287 /// @param str the input XML string to consider.
288 ///
289 /// @param escaped where to write the resulting un-escaped string.
290 void
unescape_xml_string(const std::string & str,std::string & escaped)291 unescape_xml_string(const std::string& str,
292 std::string& escaped)
293 {
294 std::string::size_type i = 0;
295 while (i < str.size())
296 {
297 if (str[i] == '&')
298 {
299 if (str[i+1] == 'l'
300 && str[i+2] == 't'
301 && str[i+3] == ';')
302 {
303 escaped += '<';
304 i+= 4;
305 }
306 else if (str[i+1] == 'g'
307 && str[i+2] == 't'
308 && str[i+3] == ';')
309 {
310 escaped += '>';
311 i += 4;
312 }
313 else if (str[i+1] == 'a'
314 && str[i+2] == 'm'
315 && str[i+3] == 'p'
316 && str[i+4] == ';')
317 {
318 escaped += '&';
319 i += 5;
320 }
321 else if (str[i+1] == 'a'
322 && str[i+2] == 'p'
323 && str[i+3] == 'o'
324 && str[i+4] == 's'
325 && str[i+5] == ';')
326 {
327 escaped += '\'';
328 i += 6;
329 }
330 else if (str[i+1] == 'q'
331 && str[i+2] == 'u'
332 && str[i+3] == 'o'
333 && str[i+4] == 't'
334 && str[i+5] == ';')
335 {
336 escaped += '"';
337 i += 6;
338 }
339 else
340 {
341 escaped += str[i];
342 ++i;
343 }
344 }
345 else
346 {
347 escaped += str[i];
348 ++i;
349 }
350 }
351 }
352
353 /// Read a string, detect the 5 predefined XML entities it may contain
354 /// and un-escape them, by writting their corresponding characters
355 /// back in. The pre-defined entities are:
356 ///
357 /// < for the character '<', > for the character '>', ' for
358 /// the character ''', " for the character '"', and & for the
359 /// character '&'.
360 ///
361 /// @param str the input XML string to consider.
362 ///
363 /// @return escaped where to write the resulting un-escaped string.
364 std::string
unescape_xml_string(const std::string & str)365 unescape_xml_string(const std::string& str)
366 {
367 std::string result;
368 unescape_xml_string(str, result);
369 return result;
370 }
371
372 /// Read a string, detect the '#&45;' entity and un-escape it into
373 /// the '-' character.
374 ///
375 /// @param str the input XML string to consider.
376 ///
377 /// @param escaped where to write the resulting un-escaped string.
378 void
unescape_xml_comment(const std::string & str,std::string & escaped)379 unescape_xml_comment(const std::string& str,
380 std::string& escaped)
381 {
382 std::string::size_type i = 0;
383 while (i < str.size())
384 {
385 if (str[i] == '&'
386 && str[i + 1] == '#'
387 && str[i + 2] == '4'
388 && str[i + 3] == '5'
389 && str[i + 4] == ';')
390 {
391 escaped += '-';
392 i += 5;
393 }
394 else
395 {
396 escaped += str[i];
397 ++i;
398 }
399 }
400 }
401
402 /// Read a string, detect the '#&45;' entity and un-escape it into
403 /// the '-' character.
404 ///
405 /// @param str the input XML string to consider.
406 ///
407 /// @return escaped where to write the resulting un-escaped string.
408 std::string
unescape_xml_comment(const std::string & str)409 unescape_xml_comment(const std::string& str)
410 {
411 std::string result;
412 unescape_xml_comment(str, result);
413 return result;
414 }
415
416 }//end namespace xml
417 }//end namespace abigail
418