Class: RXerces::XML::Document

Inherits:
Object
  • Object
show all
Defined in:
ext/rxerces/rxerces.cpp

Class Method Summary collapse

Instance Method Summary collapse

Class Method Details

.parse(str) ⇒ Object

RXerces::XML::Document.parse(string)



176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
 
# File 'ext/rxerces/rxerces.cpp', line 176

static VALUE document_parse(VALUE klass, VALUE str) {
    if (!xerces_initialized) {
        try {
            XMLPlatformUtils::Initialize();
            xerces_initialized = true;
        } catch (const XMLException& e) {
            rb_raise(rb_eRuntimeError, "Xerces initialization failed");
        }
    }

    Check_Type(str, T_STRING);
    const char* xml_str = StringValueCStr(str);

    XercesDOMParser* parser = new XercesDOMParser();
    parser->setValidationScheme(XercesDOMParser::Val_Never);
    parser->setDoNamespaces(false);
    parser->setDoSchema(false);

    try {
        MemBufInputSource input((const XMLByte*)xml_str, strlen(xml_str), "memory");
        parser->parse(input);

        DOMDocument* doc = parser->getDocument();

        DocumentWrapper* wrapper = ALLOC(DocumentWrapper);
        wrapper->doc = doc;
        wrapper->parser = parser;

        VALUE rb_doc = TypedData_Wrap_Struct(rb_cDocument, &document_type, wrapper);
        return rb_doc;
    } catch (const XMLException& e) {
        CharStr message(e.getMessage());
        delete parser;
        rb_raise(rb_eRuntimeError, "XML parsing error: %s", message.localForm());
    } catch (const DOMException& e) {
        CharStr message(e.getMessage());
        delete parser;
        rb_raise(rb_eRuntimeError, "DOM error: %s", message.localForm());
    } catch (...) {
        delete parser;
        rb_raise(rb_eRuntimeError, "Unknown XML parsing error");
    }

    return Qnil;
}

Instance Method Details

#root ⇒ Object

document.root



223
224
225
226
227
228
229
230
231
232
233
 
# File 'ext/rxerces/rxerces.cpp', line 223

static VALUE document_root(VALUE self) {
    DocumentWrapper* wrapper;
    TypedData_Get_Struct(self, DocumentWrapper, &document_type, wrapper);

    if (!wrapper->doc) {
        return Qnil;
    }

    DOMElement* root = wrapper->doc->getDocumentElement();
    return wrap_node(root, self);
}

#to_s ⇒ Object

document.to_s / document.to_xml



236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
 
# File 'ext/rxerces/rxerces.cpp', line 236

static VALUE document_to_s(VALUE self) {
    DocumentWrapper* wrapper;
    TypedData_Get_Struct(self, DocumentWrapper, &document_type, wrapper);

    if (!wrapper->doc) {
        return rb_str_new_cstr("");
    }

    try {
        DOMImplementation* impl = DOMImplementationRegistry::getDOMImplementation(XStr("LS").unicodeForm());
        DOMLSSerializer* serializer = ((DOMImplementationLS*)impl)->createLSSerializer();

        XMLCh* xml_str = serializer->writeToString(wrapper->doc);
        CharStr utf8_str(xml_str);

        VALUE result = rb_str_new_cstr(utf8_str.localForm());

        XMLString::release(&xml_str);
        serializer->release();

        return result;
    } catch (...) {
        rb_raise(rb_eRuntimeError, "Failed to serialize document");
    }

    return Qnil;
}

#to_xml ⇒ Object

document.to_s / document.to_xml



236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
 
# File 'ext/rxerces/rxerces.cpp', line 236

static VALUE document_to_s(VALUE self) {
    DocumentWrapper* wrapper;
    TypedData_Get_Struct(self, DocumentWrapper, &document_type, wrapper);

    if (!wrapper->doc) {
        return rb_str_new_cstr("");
    }

    try {
        DOMImplementation* impl = DOMImplementationRegistry::getDOMImplementation(XStr("LS").unicodeForm());
        DOMLSSerializer* serializer = ((DOMImplementationLS*)impl)->createLSSerializer();

        XMLCh* xml_str = serializer->writeToString(wrapper->doc);
        CharStr utf8_str(xml_str);

        VALUE result = rb_str_new_cstr(utf8_str.localForm());

        XMLString::release(&xml_str);
        serializer->release();

        return result;
    } catch (...) {
        rb_raise(rb_eRuntimeError, "Failed to serialize document");
    }

    return Qnil;
}

#xpath(path) ⇒ Object

document.xpath(path)



265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
 
# File 'ext/rxerces/rxerces.cpp', line 265

static VALUE document_xpath(VALUE self, VALUE path) {
    DocumentWrapper* doc_wrapper;
    TypedData_Get_Struct(self, DocumentWrapper, &document_type, doc_wrapper);

    if (!doc_wrapper->doc) {
        NodeSetWrapper* wrapper = ALLOC(NodeSetWrapper);
        wrapper->nodes_array = rb_ary_new();
        return TypedData_Wrap_Struct(rb_cNodeSet, &nodeset_type, wrapper);
    }

    Check_Type(path, T_STRING);
    const char* xpath_str = StringValueCStr(path);

    try {
        DOMElement* root = doc_wrapper->doc->getDocumentElement();
        if (!root) {
            NodeSetWrapper* wrapper = ALLOC(NodeSetWrapper);
            wrapper->nodes_array = rb_ary_new();
            return TypedData_Wrap_Struct(rb_cNodeSet, &nodeset_type, wrapper);
        }

        DOMXPathNSResolver* resolver = doc_wrapper->doc->createNSResolver(root);
        DOMXPathExpression* expression = doc_wrapper->doc->createExpression(
            XStr(xpath_str).unicodeForm(), resolver);

        DOMXPathResult* result = expression->evaluate(
            doc_wrapper->doc->getDocumentElement(),
            DOMXPathResult::ORDERED_NODE_SNAPSHOT_TYPE,
            NULL);

        VALUE nodes_array = rb_ary_new();
        XMLSize_t length = result->getSnapshotLength();

        for (XMLSize_t i = 0; i < length; i++) {
            result->snapshotItem(i);
            DOMNode* node = result->getNodeValue();
            if (node) {
                rb_ary_push(nodes_array, wrap_node(node, self));
            }
        }

        expression->release();
        resolver->release();
        result->release();

        NodeSetWrapper* wrapper = ALLOC(NodeSetWrapper);
        wrapper->nodes_array = nodes_array;
        return TypedData_Wrap_Struct(rb_cNodeSet, &nodeset_type, wrapper);

    } catch (const DOMXPathException& e) {
        CharStr message(e.getMessage());
        rb_raise(rb_eRuntimeError, "XPath error: %s", message.localForm());
    } catch (const DOMException& e) {
        CharStr message(e.getMessage());
        rb_raise(rb_eRuntimeError, "DOM error: %s", message.localForm());
    } catch (...) {
        rb_raise(rb_eRuntimeError, "Unknown XPath error");
    }

    NodeSetWrapper* wrapper = ALLOC(NodeSetWrapper);
    wrapper->nodes_array = rb_ary_new();
    return TypedData_Wrap_Struct(rb_cNodeSet, &nodeset_type, wrapper);
}