Hi,
I'd like to show you a patch which adds the functionality to store and
retrieve the columns of xml nodes.
As I said in a previous email[1] me, and probably other people are in
need of this to highlight a certain text area in an xml document.
For example, if I have an xml document
<document><node1/></document>
and an xpath expression
//node1
I'd like to be able to highlight the corresponding part of the
document. Without this patch, it was impossible, because the parser did
not save the column information.
I applied the following changes in this patch:
- added a new member to _xmlNode called column. I know that this breaks
ABI compability, but I think this is an important feature and needs to
be added soon.
- increased the type size of _xmlNode->line from (unsigned short) to
(unsigned int). This is what the idea behind "XML_PARSE_BIG_LINES" is.
( can you tell me why it has been an unsigned short in the first
place ? )
- added a function xmlGetColumnNo() which is the counterpart to
xmlGetLineNo()
- wrote a unittest for xmlGetColumnNo()
I hope for any feedback and a soon integration.
Thanks,
Daniel Nagy
[1] https://mail.gnome.org/archives/xml/2012-February/msg00045.html
diff --git a/HTMLparser.c b/HTMLparser.c
index 09a9a4b..bb521a3 100644
--- a/HTMLparser.c
+++ b/HTMLparser.c
@@ -4645,6 +4645,7 @@ htmlParseDocument(htmlParserCtxtPtr ctxt) {
}
ctxt->html = 1;
ctxt->linenumbers = 1;
+ ctxt->columnnumbers = 1;
GROW;
/*
* SAX: beginning of the document processing.
@@ -4852,6 +4853,7 @@ htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
ctxt->wellFormed = 1;
ctxt->replaceEntities = 0;
ctxt->linenumbers = xmlLineNumbersDefaultValue;
+ ctxt->columnnumbers = xmlColumnNumbersDefaultValue;
ctxt->html = 1;
ctxt->vctxt.finishDtd = XML_CTXT_FINISH_DTD_0;
ctxt->vctxt.userData = ctxt;
diff --git a/SAX2.c b/SAX2.c
index 94db7da..d735384 100644
--- a/SAX2.c
+++ b/SAX2.c
@@ -1626,10 +1626,19 @@ xmlSAX2StartElement(void *ctx, const xmlChar *fullname, const xmlChar **atts)
ctxt->nodemem = -1;
if (ctxt->linenumbers) {
if (ctxt->input != NULL) {
- if (ctxt->input->line < 65535)
- ret->line = (short) ctxt->input->line;
+ if (ctxt->input->line < UINT_MAX)
+ ret->line = (unsigned int) ctxt->input->line;
+ else
+ ret->line = UINT_MAX;
+ }
+ }
+
+ if (ctxt->columnnumbers) {
+ if (ctxt->input != NULL) {
+ if (ctxt->input->col < UINT_MAX)
+ ret->column = (unsigned int) ctxt->input->col;
else
- ret->line = 65535;
+ ret->column = UINT_MAX;
}
}
@@ -1884,15 +1893,27 @@ skip:
if (ctxt->linenumbers) {
if (ctxt->input != NULL) {
- if (ctxt->input->line < 65535)
+ if (ctxt->input->line < UINT_MAX)
ret->line = (short) ctxt->input->line;
else {
- ret->line = 65535;
+ ret->line = UINT_MAX;
if (ctxt->options & XML_PARSE_BIG_LINES)
ret->psvi = (void *) ctxt->input->line;
}
}
}
+
+ if (ctxt->columnnumbers) {
+ if (ctxt->input != NULL) {
+ if (ctxt->input->col < UINT_MAX)
+ ret->column = (short) ctxt->input->col;
+ else {
+ ret->column = UINT_MAX;
+ //if (ctxt->options & XML_PARSE_BIG_LINES)
+ // ret->psvi = (void *) ctxt->input->column;
+ }
+ }
+ }
if ((__xmlRegisterCallbacks) && (xmlRegisterNodeDefaultValue))
xmlRegisterNodeDefaultValue(ret);
@@ -2261,10 +2282,19 @@ xmlSAX2StartElementNs(void *ctx,
}
if (ctxt->linenumbers) {
if (ctxt->input != NULL) {
- if (ctxt->input->line < 65535)
- ret->line = (short) ctxt->input->line;
+ if (ctxt->input->line < UINT_MAX)
+ ret->line = (unsigned int) ctxt->input->line;
else
- ret->line = 65535;
+ ret->line = UINT_MAX;
+ }
+ }
+
+ if (ctxt->columnnumbers) {
+ if (ctxt->input != NULL) {
+ if (ctxt->input->col < UINT_MAX)
+ ret->column = (unsigned int) ctxt->input->col;
+ else
+ ret->column = UINT_MAX;
}
}
@@ -2653,12 +2683,22 @@ xmlSAX2ProcessingInstruction(void *ctx, const xmlChar *target,
if (ctxt->linenumbers) {
if (ctxt->input != NULL) {
- if (ctxt->input->line < 65535)
+ if (ctxt->input->line < UINT_MAX)
ret->line = (short) ctxt->input->line;
else
- ret->line = 65535;
+ ret->line = UINT_MAX;
+ }
+ }
+
+ if (ctxt->columnnumbers) {
+ if (ctxt->input != NULL) {
+ if (ctxt->input->col < UINT_MAX)
+ ret->column = (short) ctxt->input->col;
+ else
+ ret->column = UINT_MAX;
}
}
+
if (ctxt->inSubset == 1) {
xmlAddChild((xmlNodePtr) ctxt->myDoc->intSubset, ret);
return;
@@ -2713,10 +2753,19 @@ xmlSAX2Comment(void *ctx, const xmlChar *value)
if (ret == NULL) return;
if (ctxt->linenumbers) {
if (ctxt->input != NULL) {
- if (ctxt->input->line < 65535)
+ if (ctxt->input->line < UINT_MAX)
ret->line = (short) ctxt->input->line;
else
- ret->line = 65535;
+ ret->line = UINT_MAX;
+ }
+ }
+
+ if (ctxt->columnnumbers) {
+ if (ctxt->input != NULL) {
+ if (ctxt->input->col < UINT_MAX)
+ ret->column = (short) ctxt->input->col;
+ else
+ ret->column = UINT_MAX;
}
}
diff --git a/globals.c b/globals.c
index 69002f0..0584025 100644
--- a/globals.c
+++ b/globals.c
@@ -155,6 +155,7 @@ xmlStrdupFunc xmlMemStrdup = (xmlStrdupFunc) xmlStrdup;
#undef xmlTreeIndentString
#undef xmlKeepBlanksDefaultValue
#undef xmlLineNumbersDefaultValue
+#undef xmlColumnNumbersDefaultValue
#undef xmlLoadExtDtdDefaultValue
#undef xmlParserDebugEntities
#undef xmlParserVersion
@@ -253,6 +254,16 @@ static int xmlPedanticParserDefaultValueThrDef = 0;
int xmlLineNumbersDefaultValue = 0;
static int xmlLineNumbersDefaultValueThrDef = 0;
/**
+ * xmlColumnNumbersDefaultValue:
+ *
+ * Global setting, indicate that the parser should store the column number
+ * in the content field of elements in the DOM tree.
+ * Disabled by default since this may not be safe for old classes of
+ * applicaton.
+ */
+int xmlColumnNumbersDefaultValue = 0;
+static int xmlColumnNumbersDefaultValueThrDef = 0;
+/**
* xmlKeepBlanksDefaultValue:
*
* Global setting, indicate that the parser should keep all blanks
@@ -980,6 +991,23 @@ int xmlThrDefLineNumbersDefaultValue(int v) {
return ret;
}
+#undef xmlColumnNumbersDefaultValue
+int *
+__xmlColumnNumbersDefaultValue(void) {
+ if (IS_MAIN_THREAD)
+ return (&xmlColumnNumbersDefaultValue);
+ else
+ return (&xmlGetGlobalState()->xmlColumnNumbersDefaultValue);
+}
+int xmlThrDefColumnNumbersDefaultValue(int v) {
+ int ret;
+ xmlMutexLock(xmlThrDefMutex);
+ ret = xmlColumnNumbersDefaultValueThrDef;
+ xmlColumnNumbersDefaultValueThrDef = v;
+ xmlMutexUnlock(xmlThrDefMutex);
+ return ret;
+}
+
#undef xmlLoadExtDtdDefaultValue
int *
__xmlLoadExtDtdDefaultValue(void) {
diff --git a/include/libxml/globals.h b/include/libxml/globals.h
index 9d688e0..d072924 100644
--- a/include/libxml/globals.h
+++ b/include/libxml/globals.h
@@ -82,6 +82,7 @@ XMLCALL xmlOutputBufferCreateFilenameDefault (xmlOutputBufferCreateFilenameFunc
#undef xmlTreeIndentString
#undef xmlKeepBlanksDefaultValue
#undef xmlLineNumbersDefaultValue
+#undef xmlColumnNumbersDefaultValue
#undef xmlLoadExtDtdDefaultValue
#undef xmlMalloc
#undef xmlMallocAtomic
@@ -143,6 +144,7 @@ struct _xmlGlobalState
int xmlGetWarningsDefaultValue;
int xmlKeepBlanksDefaultValue;
int xmlLineNumbersDefaultValue;
+ int xmlColumnNumbersDefaultValue;
int xmlLoadExtDtdDefaultValue;
int xmlParserDebugEntities;
int xmlPedanticParserDefaultValue;
@@ -410,6 +412,15 @@ XMLPUBVAR int xmlLineNumbersDefaultValue;
#endif
XMLPUBFUN int XMLCALL xmlThrDefLineNumbersDefaultValue(int v);
+XMLPUBFUN int * XMLCALL __xmlColumnNumbersDefaultValue(void);
+#ifdef LIBXML_THREAD_ENABLED
+#define xmlColumnNumbersDefaultValue \
+(*(__xmlColumnNumbersDefaultValue()))
+#else
+XMLPUBVAR int xmlColumnNumbersDefaultValue;
+#endif
+XMLPUBFUN int XMLCALL xmlThrDefColumnNumbersDefaultValue(int v);
+
XMLPUBFUN int * XMLCALL __xmlLoadExtDtdDefaultValue(void);
#ifdef LIBXML_THREAD_ENABLED
#define xmlLoadExtDtdDefaultValue \
diff --git a/include/libxml/parser.h b/include/libxml/parser.h
index 1f11fd9..987f16a 100644
--- a/include/libxml/parser.h
+++ b/include/libxml/parser.h
@@ -257,6 +257,7 @@ struct _xmlParserCtxt {
int loadsubset; /* should the external subset be loaded */
int linenumbers; /* set line number in element content */
+ int columnnumbers; /* set column number in element content */
void *catalogs; /* document's own catalog */
int recovery; /* run in recovery mode */
int progressive; /* is this a progressive parsing */
diff --git a/include/libxml/tree.h b/include/libxml/tree.h
index 68f92f9..7e18a01 100644
--- a/include/libxml/tree.h
+++ b/include/libxml/tree.h
@@ -502,7 +502,8 @@ struct _xmlNode {
struct _xmlAttr *properties;/* properties list */
xmlNs *nsDef; /* namespace definitions on this node */
void *psvi; /* for type/PSVI informations */
- unsigned short line; /* line number */
+ unsigned int line; /* line number */
+ unsigned int column; /* column index */
unsigned short extra; /* extra data for XPath/XSLT */
};
@@ -907,6 +908,8 @@ XMLPUBFUN xmlNodePtr XMLCALL
*/
XMLPUBFUN long XMLCALL
xmlGetLineNo (xmlNodePtr node);
+XMLPUBFUN long XMLCALL
+ xmlGetColumnNo (xmlNodePtr node);
#if defined(LIBXML_TREE_ENABLED) || defined(LIBXML_DEBUG_ENABLED)
XMLPUBFUN xmlChar * XMLCALL
xmlGetNodePath (xmlNodePtr node);
diff --git a/parser.c b/parser.c
index 9a57b01..0288d66 100644
--- a/parser.c
+++ b/parser.c
@@ -12880,6 +12880,7 @@ xmlParseCtxtExternalEntity(xmlParserCtxtPtr ctx, const xmlChar *URL,
ctxt->attsDefault = ctx->attsDefault;
ctxt->attsSpecial = ctx->attsSpecial;
ctxt->linenumbers = ctx->linenumbers;
+ ctxt->columnnumbers = ctx->columnnumbers;
xmlParseContent(ctxt);
@@ -13960,6 +13961,7 @@ xmlCreateURLParserCtxt(const char *filename, int options)
if (options)
xmlCtxtUseOptionsInternal(ctxt, options, NULL);
ctxt->linenumbers = 1;
+ ctxt->columnnumbers = 1;
inputStream = xmlLoadExternalEntity(filename, NULL, ctxt);
if (inputStream == NULL) {
@@ -15011,6 +15013,7 @@ xmlCtxtUseOptionsInternal(xmlParserCtxtPtr ctxt, int options, const char *encodi
options -= XML_PARSE_BIG_LINES;
}
ctxt->linenumbers = 1;
+ ctxt->columnnumbers = 1;
return (options);
}
diff --git a/parserInternals.c b/parserInternals.c
index b8d5bbc..b8fc1d0 100644
--- a/parserInternals.c
+++ b/parserInternals.c
@@ -1694,6 +1694,7 @@ xmlInitParserCtxt(xmlParserCtxtPtr ctxt)
ctxt->validate = xmlDoValidityCheckingDefaultValue;
ctxt->pedantic = xmlPedanticParserDefaultValue;
ctxt->linenumbers = xmlLineNumbersDefaultValue;
+ ctxt->columnnumbers = xmlColumnNumbersDefaultValue;
ctxt->keepBlanks = xmlKeepBlanksDefaultValue;
if (ctxt->keepBlanks == 0)
ctxt->sax->ignorableWhitespace = xmlSAX2IgnorableWhitespace;
diff --git a/testapi.c b/testapi.c
index deb8cca..e20a333 100644
--- a/testapi.c
+++ b/testapi.c
@@ -20580,6 +20580,37 @@ test_xmlGetLineNo(void) {
return(test_ret);
}
+static int
+test_xmlGetColumnNo(void) {
+ int test_ret = 0;
+
+ int mem_base;
+ long ret_val;
+ xmlNodePtr node; /* valid node */
+ int n_node;
+
+ for (n_node = 0;n_node < gen_nb_xmlNodePtr;n_node++) {
+ mem_base = xmlMemBlocks();
+ node = gen_xmlNodePtr(n_node, 0);
+
+ ret_val = xmlGetColumnNo(node);
+ desret_long(ret_val);
+ call_tests++;
+ des_xmlNodePtr(n_node, node, 0);
+ xmlResetLastError();
+ if (mem_base != xmlMemBlocks()) {
+ printf("Leak of %d blocks found in xmlGetColumnNo",
+ xmlMemBlocks() - mem_base);
+ test_ret++;
+ printf(" %d", n_node);
+ printf("\n");
+ }
+ }
+ function_tests++;
+
+ return(test_ret);
+}
+
static int
test_xmlGetNoNsProp(void) {
@@ -24245,6 +24276,7 @@ test_tree(void) {
test_ret += test_xmlGetIntSubset();
test_ret += test_xmlGetLastChild();
test_ret += test_xmlGetLineNo();
+ test_ret += test_xmlGetColumnNo();
test_ret += test_xmlGetNoNsProp();
test_ret += test_xmlGetNodePath();
test_ret += test_xmlGetNsList();
diff --git a/tree.c b/tree.c
index df6f608..7a76a5d 100644
--- a/tree.c
+++ b/tree.c
@@ -4192,6 +4192,7 @@ xmlStaticCopyNode(const xmlNodePtr node, xmlDocPtr doc, xmlNodePtr parent,
}else{
if (node->type == XML_ELEMENT_NODE)
ret->line = node->line;
+ ret->column = node->column;
}
if (parent != NULL) {
xmlNodePtr tmp;
@@ -4551,7 +4552,7 @@ xmlGetLineNoInternal(xmlNodePtr node, int depth)
(node->type == XML_TEXT_NODE) ||
(node->type == XML_COMMENT_NODE) ||
(node->type == XML_PI_NODE)) {
- if (node->line == 65535) {
+ if (node->line == UINT_MAX) {
if ((node->type == XML_TEXT_NODE) && (node->psvi != NULL))
result = (long) node->psvi;
else if ((node->type == XML_ELEMENT_NODE) &&
@@ -4562,7 +4563,7 @@ xmlGetLineNoInternal(xmlNodePtr node, int depth)
else if (node->prev != NULL)
result = xmlGetLineNoInternal(node->prev, depth + 1);
}
- if ((result == -1) || (result == 65535))
+ if ((result == -1) || (result == UINT_MAX))
result = (long) node->line;
} else if ((node->prev != NULL) &&
((node->prev->type == XML_ELEMENT_NODE) ||
@@ -4593,6 +4594,40 @@ xmlGetLineNo(xmlNodePtr node)
return(xmlGetLineNoInternal(node, 0));
}
+/**
+ * xmlGetColumnNo:
+ * @node: valid node
+ *
+ * Get column number of @node. This requires activation of this option
+ * before invoking the parser by calling xmlColumnNumbersDefault(1)
+ *
+ * Returns the col number if successful, -1 otherwise
+ */
+long
+xmlGetColumnNo(xmlNodePtr node)
+{
+ long result = -1;
+
+ if (!node)
+ return result;
+ if ((node->type == XML_ELEMENT_NODE) ||
+ (node->type == XML_TEXT_NODE) ||
+ (node->type == XML_COMMENT_NODE) ||
+ (node->type == XML_PI_NODE))
+ result = (long) node->column;
+ else if ((node->prev != NULL) &&
+ ((node->prev->type == XML_ELEMENT_NODE) ||
+ (node->prev->type == XML_TEXT_NODE) ||
+ (node->prev->type == XML_COMMENT_NODE) ||
+ (node->prev->type == XML_PI_NODE)))
+ result = xmlGetColumnNo(node->prev);
+ else if ((node->parent != NULL) &&
+ (node->parent->type == XML_ELEMENT_NODE))
+ result = xmlGetColumnNo(node->parent);
+
+ return result;
+}
+
#if defined(LIBXML_TREE_ENABLED) || defined(LIBXML_DEBUG_ENABLED)
/**
* xmlGetNodePath:
@@ -9143,6 +9178,7 @@ ns_end:
if (cur->type == XML_ELEMENT_NODE) {
cur->psvi = NULL;
cur->line = 0;
+ cur->column = 0;
cur->extra = 0;
/*
* Walk attributes.
diff --git a/xmlreader.c b/xmlreader.c
index c6ca46e..97771ad 100644
--- a/xmlreader.c
+++ b/xmlreader.c
@@ -2140,6 +2140,7 @@ xmlNewTextReader(xmlParserInputBufferPtr input, const char *URI) {
ret->ctxt->parseMode = XML_PARSE_READER;
ret->ctxt->_private = ret;
ret->ctxt->linenumbers = 1;
+ ret->ctxt->columnnumbers = 1;
ret->ctxt->dictNames = 1;
ret->allocs = XML_TEXTREADER_CTXT;
/*
@@ -5217,6 +5218,7 @@ xmlTextReaderSetup(xmlTextReaderPtr reader,
}
reader->ctxt->_private = reader;
reader->ctxt->linenumbers = 1;
+ reader->ctxt->columnnumbers = 1;
reader->ctxt->dictNames = 1;
/*
* use the parser dictionnary to allocate all elements and attributes names
diff --git a/xmlschemas.c b/xmlschemas.c
index c3a080f..0d32870 100644
--- a/xmlschemas.c
+++ b/xmlschemas.c
@@ -28785,6 +28785,7 @@ xmlSchemaValidateStream(xmlSchemaValidCtxtPtr ctxt,
xmlCtxtUseOptions(pctxt, options);
#endif
pctxt->linenumbers = 1;
+ pctxt->columnnumbers = 1;
xmlSchemaValidateSetLocator(ctxt, xmlSchemaValidateStreamLocator, pctxt);
inputStream = xmlNewIOInputStream(pctxt, input, enc);;
_______________________________________________
xml mailing list, project page http://xmlsoft.org/
[email protected]
https://mail.gnome.org/mailman/listinfo/xml
