added the --stream flag to use the TextReader API small performance tweak
* xmllint.c: added the --stream flag to use the TextReader API
* xmlreader.c: small performance tweak
Daniel
diff --git a/ChangeLog b/ChangeLog
index 81404ed..1fda442 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,8 @@
+Fri Jan 3 17:18:32 CET 2003 Daniel Veillard <daniel@veillard.com>
+
+ * xmllint.c: added the --stream flag to use the TextReader API
+ * xmlreader.c: small performance tweak
+
Fri Jan 3 13:50:55 CET 2003 Daniel Veillard <daniel@veillard.com>
* xmlreader.c python/tests/reader2py: okay the DTD validation
diff --git a/xmllint.c b/xmllint.c
index 5c16660..41dac2f 100644
--- a/xmllint.c
+++ b/xmllint.c
@@ -81,6 +81,7 @@
#include <libxml/DOCBparser.h>
#endif
#include <libxml/globals.h>
+#include <libxml/xmlreader.h>
#ifdef LIBXML_DEBUG_ENABLED
static int debug = 0;
@@ -124,6 +125,7 @@
static int catalogs = 0;
static int nocatalogs = 0;
#endif
+static int stream = 0;
static const char *output = NULL;
@@ -563,7 +565,75 @@
/************************************************************************
* *
- * Test processing *
+ * Stream Test processing *
+ * *
+ ************************************************************************/
+static int count = 0;
+static int elem, attrs;
+
+static void processNode(xmlTextReaderPtr reader) {
+ if (debug) {
+ xmlChar *name, *value;
+
+ name = xmlTextReaderName(reader);
+ if (name == NULL)
+ name = xmlStrdup(BAD_CAST "--");
+ value = xmlTextReaderValue(reader);
+
+ printf("%d %d %d %s",
+ xmlTextReaderDepth(reader),
+ xmlTextReaderNodeType(reader),
+ xmlTextReaderIsEmptyElement(reader),
+ name);
+ xmlFree(name);
+ if (value == NULL)
+ printf("\n");
+ else {
+ printf(" %s\n", value);
+ xmlFree(value);
+ }
+ }
+}
+
+static void streamFile(char *filename) {
+ xmlTextReaderPtr reader;
+ int ret;
+
+ if (count) {
+ elem = 0;
+ attrs = 0;
+ }
+
+ reader = xmlNewTextReaderFilename(filename);
+ if (reader != NULL) {
+ if (valid)
+ xmlTextReaderSetParserProp(reader, XML_PARSER_VALIDATE, 1);
+
+ /*
+ * Process all nodes in sequence
+ */
+ ret = xmlTextReaderRead(reader);
+ while (ret == 1) {
+ if (debug)
+ processNode(reader);
+ ret = xmlTextReaderRead(reader);
+ }
+
+ /*
+ * Done, cleanup and status
+ */
+ xmlFreeTextReader(reader);
+ if (ret != 0) {
+ printf("%s : failed to parse\n", filename);
+ }
+ } else {
+ fprintf(stderr, "Unable to open %s\n", filename);
+ }
+}
+
+/************************************************************************
+ * *
+ * Tree Test processing *
* *
************************************************************************/
static void parseAndPrintFile(char *filename) {
@@ -1106,12 +1176,13 @@
printf("\t--loaddtd : fetch external DTD\n");
printf("\t--dtdattr : loaddtd + populate the tree with inherited attributes \n");
printf("\t--dropdtd : remove the DOCTYPE of the input docs\n");
+ printf("\t--stream : use the streaming interface to process very large files\n");
printf("\nLibxml project home page: http://xmlsoft.org/\n");
printf("To report bugs or get some help check: http://xmlsoft.org/bugs.html\n");
}
int
main(int argc, char **argv) {
- int i, count;
+ int i, acount;
int files = 0;
int version = 0;
@@ -1277,6 +1348,10 @@
noblanks++;
format++;
xmlKeepBlanksDefault(0);
+ }
+ else if ((!strcmp(argv[i], "-stream")) ||
+ (!strcmp(argv[i], "--stream"))) {
+ stream++;
} else {
fprintf(stderr, "Unknown option %s\n", argv[i]);
usage(argv[0]);
@@ -1338,10 +1413,17 @@
/* Remember file names. "-" means stdin. <sven@zen.org> */
if ((argv[i][0] != '-') || (strcmp(argv[i], "-") == 0)) {
if (repeat) {
- for (count = 0;count < 100 * repeat;count++)
+ for (acount = 0;acount < 100 * repeat;acount++)
+ if (stream != 0)
+ streamFile(argv[i]);
+ else
+ parseAndPrintFile(argv[i]);
+ } else {
+ if (stream != 0)
+ streamFile(argv[i]);
+ else
parseAndPrintFile(argv[i]);
- } else
- parseAndPrintFile(argv[i]);
+ }
files ++;
if ((timing) && (repeat)) {
endTimer("100 iterations");
diff --git a/xmlreader.c b/xmlreader.c
index 5bf9355..159447a 100644
--- a/xmlreader.c
+++ b/xmlreader.c
@@ -10,6 +10,14 @@
* daniel@veillard.com
*/
+/*
+ * TODOs:
+ * - provide an API to expand part of the tree
+ * - provide an API to preserve part of the tree
+ * - Streaming XInclude support
+ * - setting(s) for NoBlanks
+ * - performances and tuning ...
+ */
#define IN_LIBXML
#include "libxml.h"
@@ -357,8 +365,7 @@
} else
break;
}
- if ((inbuf->content[cur] == '>') || (inbuf->content[cur] == '&') ||
- (inbuf->content[cur] == ';')) {
+ if (inbuf->content[cur] == '>') {
cur = cur + 1;
val = xmlParseChunk(reader->ctxt,
(const char *) &inbuf->content[reader->cur],