xmllint.c

libxml,在UNIX/LINUX下非常重要的一个库,为XML相关应用提供方便.目前上载的是最新版本,若要取得最新版本,请参考里面的readme.

	        xmlChar *result = NULL;
		int size;

		size = xmlC14NDocDumpMemory(doc, NULL, 0, NULL, 1, &result);
		if (size >= 0) {
		    write(1, result, size);
		    xmlFree(result);
		} else {
		    fprintf(stderr, "Failed to canonicalize\n");
		    progresult = XMLLINT_ERR_OUT;
		}
	    } else
#endif
#ifdef HAVE_SYS_MMAN_H
	    if (memory) {
		xmlChar *result;
		int len;

		if (encoding != NULL) {
		    if ( format ) {
		        xmlDocDumpFormatMemoryEnc(doc, &result, &len, encoding, 1);
		    } else { 
			xmlDocDumpMemoryEnc(doc, &result, &len, encoding);
		    }
		} else {
		    if (format)
			xmlDocDumpFormatMemory(doc, &result, &len, 1);
		    else
			xmlDocDumpMemory(doc, &result, &len);
		}
		if (result == NULL) {
		    fprintf(stderr, "Failed to save\n");
		    progresult = XMLLINT_ERR_OUT;
		} else {
		    write(1, result, len);
		    xmlFree(result);
		}
	    } else
#endif /* HAVE_SYS_MMAN_H */
	    if (compress) {
		xmlSaveFile(output ? output : "-", doc);
	    }
	    else if (encoding != NULL) {
	        if ( format ) {
		    ret = xmlSaveFormatFileEnc(output ? output : "-", doc,
		                               encoding, 1);
		}
		else {
		    ret = xmlSaveFileEnc(output ? output : "-", doc, encoding);
		}
		if (ret < 0) {
		    fprintf(stderr, "failed save to %s\n",
		            output ? output : "-");
		    progresult = XMLLINT_ERR_OUT;
		}
	    }
	    else if (format) {
		ret = xmlSaveFormatFile(output ? output : "-", doc, 1);
		if (ret < 0) {
		    fprintf(stderr, "failed save to %s\n",
		            output ? output : "-");
		    progresult = XMLLINT_ERR_OUT;
		}
	    }
	    else {
		FILE *out;
		if (output == NULL)
		    out = stdout;
		else {
		    out = fopen(output,"wb");
		}
		if (out != NULL) {
		    if (xmlDocDump(out, doc) < 0)
		        progresult = XMLLINT_ERR_OUT;

		    if (output != NULL)
			fclose(out);
		} else {
		    fprintf(stderr, "failed to open %s\n", output);
		    progresult = XMLLINT_ERR_OUT;
		}
	    }
	    if ((timing) && (!repeat)) {
		endTimer("Saving");
	    }
#ifdef LIBXML_DEBUG_ENABLED
	} else {
	    FILE *out;
	    if (output == NULL)
	        out = stdout;
	    else {
		out = fopen(output,"wb");
	    }
	    if (out != NULL) {
		xmlDebugDumpDocument(out, doc);

		if (output != NULL)
		    fclose(out);
	    } else {
		fprintf(stderr, "failed to open %s\n", output);
		progresult = XMLLINT_ERR_OUT;
	    }
	}
#endif
    }
#endif /* LIBXML_OUTPUT_ENABLED */

#ifdef LIBXML_VALID_ENABLED
    /*
     * A posteriori validation test
     */
    if ((dtdvalid != NULL) || (dtdvalidfpi != NULL)) {
	xmlDtdPtr dtd;

	if ((timing) && (!repeat)) {
	    startTimer();
	}
	if (dtdvalid != NULL)
	    dtd = xmlParseDTD(NULL, (const xmlChar *)dtdvalid); 
	else
	    dtd = xmlParseDTD((const xmlChar *)dtdvalidfpi, NULL); 
	if ((timing) && (!repeat)) {
	    endTimer("Parsing DTD");
	}
	if (dtd == NULL) {
	    if (dtdvalid != NULL)
		xmlGenericError(xmlGenericErrorContext,
			"Could not parse DTD %s\n", dtdvalid);
	    else
		xmlGenericError(xmlGenericErrorContext,
			"Could not parse DTD %s\n", dtdvalidfpi);
	    progresult = XMLLINT_ERR_DTD;
	} else {
	    xmlValidCtxtPtr cvp;

	    if ((cvp = xmlNewValidCtxt()) == NULL) {
		xmlGenericError(xmlGenericErrorContext,
			"Couldn't allocate validation context\n");
		exit(-1);
	    }
	    cvp->userData = (void *) stderr;
	    cvp->error    = (xmlValidityErrorFunc) fprintf;
	    cvp->warning  = (xmlValidityWarningFunc) fprintf;

	    if ((timing) && (!repeat)) {
		startTimer();
	    }
	    if (!xmlValidateDtd(cvp, doc, dtd)) {
		if (dtdvalid != NULL)
		    xmlGenericError(xmlGenericErrorContext,
			    "Document %s does not validate against %s\n",
			    filename, dtdvalid);
		else
		    xmlGenericError(xmlGenericErrorContext,
			    "Document %s does not validate against %s\n",
			    filename, dtdvalidfpi);
		progresult = XMLLINT_ERR_VALID;
	    }
	    if ((timing) && (!repeat)) {
		endTimer("Validating against DTD");
	    }
	    xmlFreeValidCtxt(cvp);
	    xmlFreeDtd(dtd);
	}
    } else if (postvalid) {
	xmlValidCtxtPtr cvp;

	if ((cvp = xmlNewValidCtxt()) == NULL) {
	    xmlGenericError(xmlGenericErrorContext,
		    "Couldn't allocate validation context\n");
	    exit(-1);
	}

	if ((timing) && (!repeat)) {
	    startTimer();
	}
	cvp->userData = (void *) stderr;
	cvp->error    = (xmlValidityErrorFunc) fprintf;
	cvp->warning  = (xmlValidityWarningFunc) fprintf;
	if (!xmlValidateDocument(cvp, doc)) {
	    xmlGenericError(xmlGenericErrorContext,
		    "Document %s does not validate\n", filename);
	    progresult = XMLLINT_ERR_VALID;
	}
	if ((timing) && (!repeat)) {
	    endTimer("Validating");
	}
	xmlFreeValidCtxt(cvp);
    }
#endif /* LIBXML_VALID_ENABLED */
#ifdef LIBXML_SCHEMAS_ENABLED
    if (relaxngschemas != NULL) {
	xmlRelaxNGValidCtxtPtr ctxt;
	int ret;

	if ((timing) && (!repeat)) {
	    startTimer();
	}

	ctxt = xmlRelaxNGNewValidCtxt(relaxngschemas);
	xmlRelaxNGSetValidErrors(ctxt,
		(xmlRelaxNGValidityErrorFunc) fprintf,
		(xmlRelaxNGValidityWarningFunc) fprintf,
		stderr);
	ret = xmlRelaxNGValidateDoc(ctxt, doc);
	if (ret == 0) {
	    fprintf(stderr, "%s validates\n", filename);
	} else if (ret > 0) {
	    fprintf(stderr, "%s fails to validate\n", filename);
	    progresult = XMLLINT_ERR_VALID;
	} else {
	    fprintf(stderr, "%s validation generated an internal error\n",
		   filename);
	    progresult = XMLLINT_ERR_VALID;
	}
	xmlRelaxNGFreeValidCtxt(ctxt);
	if ((timing) && (!repeat)) {
	    endTimer("Validating");
	}
    } else if (wxschemas != NULL) {
	xmlSchemaValidCtxtPtr ctxt;
	int ret;

	if ((timing) && (!repeat)) {
	    startTimer();
	}

	ctxt = xmlSchemaNewValidCtxt(wxschemas);
	xmlSchemaSetValidErrors(ctxt,
		(xmlSchemaValidityErrorFunc) fprintf,
		(xmlSchemaValidityWarningFunc) fprintf,
		stderr);
	ret = xmlSchemaValidateDoc(ctxt, doc);
	if (ret == 0) {
	    fprintf(stderr, "%s validates\n", filename);
	} else if (ret > 0) {
	    fprintf(stderr, "%s fails to validate\n", filename);
	    progresult = XMLLINT_ERR_VALID;
	} else {
	    fprintf(stderr, "%s validation generated an internal error\n",
		   filename);
	    progresult = XMLLINT_ERR_VALID;
	}
	xmlSchemaFreeValidCtxt(ctxt);
	if ((timing) && (!repeat)) {
	    endTimer("Validating");
	}
    }
#endif

#ifdef LIBXML_DEBUG_ENABLED
    if ((debugent) && (!html))
	xmlDebugDumpEntities(stderr, doc);
#endif

    /*
     * free it.
     */
    if ((timing) && (!repeat)) {
	startTimer();
    }
    xmlFreeDoc(doc);
    if ((timing) && (!repeat)) {
	endTimer("Freeing");
    }
}

/************************************************************************
 * 									*
 * 			Usage and Main					*
 * 									*
 ************************************************************************/

static void showVersion(const char *name) {
    fprintf(stderr, "%s: using libxml version %s\n", name, xmlParserVersion);
    fprintf(stderr, "   compiled with: ");
#ifdef LIBXML_VALID_ENABLED
    fprintf(stderr, "DTDValid ");
#endif
#ifdef LIBXML_FTP_ENABLED
    fprintf(stderr, "FTP ");
#endif
#ifdef LIBXML_HTTP_ENABLED
    fprintf(stderr, "HTTP ");
#endif
#ifdef LIBXML_HTML_ENABLED
    fprintf(stderr, "HTML ");
#endif
#ifdef LIBXML_C14N_ENABLED
    fprintf(stderr, "C14N ");
#endif
#ifdef LIBXML_CATALOG_ENABLED
    fprintf(stderr, "Catalog ");
#endif
#ifdef LIBXML_XPATH_ENABLED
    fprintf(stderr, "XPath ");
#endif
#ifdef LIBXML_XPTR_ENABLED
    fprintf(stderr, "XPointer ");
#endif
#ifdef LIBXML_XINCLUDE_ENABLED
    fprintf(stderr, "XInclude ");
#endif
#ifdef LIBXML_ICONV_ENABLED
    fprintf(stderr, "Iconv ");
#endif
#ifdef DEBUG_MEMORY_LOCATION
    fprintf(stderr, "MemDebug ");
#endif
#ifdef LIBXML_UNICODE_ENABLED
    fprintf(stderr, "Unicode ");
#endif
#ifdef LIBXML_REGEXP_ENABLED
    fprintf(stderr, "Regexps ");
#endif
#ifdef LIBXML_AUTOMATA_ENABLED
    fprintf(stderr, "Automata ");
#endif
#ifdef LIBXML_SCHEMAS_ENABLED
    fprintf(stderr, "Schemas ");
#endif
#ifdef LIBXML_MODULES_ENABLED
    fprintf(stderr, "Modules ");
#endif
    fprintf(stderr, "\n");
}

static void usage(const char *name) {
    printf("Usage : %s [options] XMLfiles ...\n", name);
#ifdef LIBXML_OUTPUT_ENABLED
    printf("\tParse the XML files and output the result of the parsing\n");
#else
    printf("\tParse the XML files\n");
#endif /* LIBXML_OUTPUT_ENABLED */
    printf("\t--version : display the version of the XML library used\n");
#ifdef LIBXML_DEBUG_ENABLED
    printf("\t--debug : dump a debug tree of the in-memory document\n");
    printf("\t--shell : run a navigating shell\n");
    printf("\t--debugent : debug the entities defined in the document\n");
#else
#ifdef LIBXML_READER_ENABLED
    printf("\t--debug : dump the nodes content when using --stream\n");
#endif /* LIBXML_READER_ENABLED */
#endif
#ifdef LIBXML_TREE_ENABLED
    printf("\t--copy : used to test the internal copy implementation\n");
#endif /* LIBXML_TREE_ENABLED */
    printf("\t--recover : output what was parsable on broken XML documents\n");
    printf("\t--noent : substitute entity references by their value\n");
    printf("\t--noout : don't output the result tree\n");
    printf("\t--path 'paths': provide a set of paths for resources\n");
    printf("\t--load-trace : print trace of all external entites loaded\n");
    printf("\t--nonet : refuse to fetch DTDs or entities over network\n");
    printf("\t--htmlout : output results as HTML\n");
    printf("\t--nowrap : do not put HTML doc wrapper\n");
#ifdef LIBXML_VALID_ENABLED
    printf("\t--valid : validate the document in addition to std well-formed check\n");
    printf("\t--postvalid : do a posteriori validation, i.e after parsing\n");
    printf("\t--dtdvalid URL : do a posteriori validation against a given DTD\n");
    printf("\t--dtdvalidfpi FPI : same but name the DTD with a Public Identifier\n");
#endif /* LIBXML_VALID_ENABLED */
    printf("\t--timing : print some timings\n");
    printf("\t--output file or -o file: save to a given file\n");
    printf("\t--repeat : repeat 100 times, for timing or profiling\n");
    printf("\t--insert : ad-hoc test for valid insertions\n");
#ifdef LIBXML_OUTPUT_ENABLED
#ifdef HAVE_ZLIB_H
    printf("\t--compress : turn on gzip compression of output\n");
#endif
#endif /* LIBXML_OUTPUT_ENABLED */
#ifdef LIBXML_HTML_ENABLED
    printf("\t--html : use the HTML parser\n");
    printf("\t--xmlout : force to use the XML serializer when using --html\n");
#endif
#ifdef LIBXML_PUSH_ENABLED
    printf("\t--push : use the push mode of the parser\n");
#endif /* LIBXML_PUSH_ENABLED */
#ifdef HAVE_SYS_MMAN_H
    printf("\t--memory : parse from memory\n");
#endif
    printf("\t--maxmem nbbytes : limits memory allocation to nbbytes bytes\n");
    printf("\t--nowarning : do not emit warnings from parser/validator\n");
    printf("\t--noblanks : drop (ignorable?) blanks spaces\n");
    printf("\t--nocdata : replace cdata section with text nodes\n");
#ifdef LIBXML_OUTPUT_ENABLED
    printf("\t--format : reformat/reindent the input\n");
    printf("\t--encode encoding : output in the given encoding\n");
    printf("\t--dropdtd : remove the DOCTYPE of the input docs\n");
#endif /* LIBXML_OUTPUT_ENABLED */
    printf("\t--c14n: save in W3C canonical format (with comments)\n");
#ifdef LIBXML_C14N_ENABLED
#endif /* LIBXML_C14N_ENABLED */
    printf("\t--nsclean : remove redundant namespace declarations\n");
    printf("\t--testIO : test user I/O support\n");
#ifdef LIBXML_CATALOG_ENABLED
    printf("\t--catalogs : use SGML catalogs from $SGML_CATALOG_FILES\n");
    printf("\t             otherwise XML Catalogs starting from \n");
    printf("\t         %s are activated by default\n", XML_XML_DEFAULT_CATALOG);
    printf("\t--nocatalogs: deactivate all catalogs\n");
#endif
    printf("\t--auto : generate a small doc on the fly\n");
#ifdef LIBXML_XINCLUDE_ENABLED
    printf("\t--xinclude : do XInclude processing\n");
    printf("\t--noxincludenode : same but do not generate XInclude nodes\n");
#endif
    printf("\t--loaddtd : fetch external DTD\n");
    printf("\t--dtdattr : loaddtd + populate the tree with inherited attributes \n");
#ifdef LIBXML_READER_ENABLED
    printf("\t--stream : use the streaming interface to process very large files\n");
    printf("\t--walker : create a reader and walk though the resulting doc\n");
#endif /* LIBXML_READER_ENABLED */
#ifdef LIBXML_PATTERN_ENABLED
    printf("\t--pattern pattern_value : test the pattern support\n");
#endif
    printf("\t--chkregister : verify the node registration code\n");
#ifdef LIBXML_SCHEMAS_ENABLED
    printf("\t--relaxng schema : do RelaxNG validation against the schema\n");
    printf("\t--schema schema : do validation against the WXS schema\n");
#endif
    printf("\nLibxml project home page: http://xmlsoft.org/\n");
    printf("To report bugs or get some help check: http://xmlsoft.org/bugs.html\n");
}

static void registerNode(xmlNodePtr node)
{
    node->_private = malloc(sizeof(long));
    *(long*)node->_private = (long) 0x81726354;
    nbregister++;
}

static void deregisterNode(xmlNodePtr node)
{
    assert(node->_private != NULL);
    assert(*(long*)node->_private == (long) 0x81726354);
    free(node->_private);
    nbregister--;
}

int
main(int argc, char **argv) {
    int i, acount;
    int files = 0;
    int version = 0;
    const char* indent;
    
    if (argc <= 1) {
	usage(argv[0]);
	return(1);
    }
    LIBXML_TEST_VERSION
    for (i = 1; i < argc ; i++) {
	if (!strcmp(argv[i], "-"))
	    break;

	if (argv[i][0] != '-')
	    continue;
	if ((!strcmp(argv[i], "-debug")) || (!strcmp(argv[i], "--debug")))
	    debug++;
	else
#ifdef LIBXML_DEBUG_ENABLED
	if ((!strcmp(argv[i], "-shell")) ||
	         (!strcmp(argv[i], "--shell"))) {
	    shell++;
            noout = 1;
        } else 
#endif
#ifdef LIBXML_TREE_ENABLED
	if ((!strcmp(argv[i], "-copy")) || (!strcmp(argv[i], "--copy")))
	    copy++;
	else
#endif /* LIBXML_TREE_ENABLED */
	if ((!strcmp(argv[i], "-recover")) ||
	         (!strcmp(argv[i], "--recover"))) {
	    recovery++;
	    options |= XML_PARSE_RECOVER;
	} else if ((!strcmp(argv[i], "-noent")) ||
	         (!strcmp(argv[i], "--noent"))) {
	    noent++;
	    options |= XML_PARSE_NOENT;
	} else if ((!strcmp(argv[i], "-nsclean")) ||
	         (!strcmp(argv[i], "--nsclean"))) {
	    options |= XML_PARSE_NSCLEAN;
	} else if ((!strcmp(argv[i], "-nocdata")) ||
	         (!strcmp(argv[i], "--nocdata"))) {
	    options |= XML_PARSE_NOCDATA;
	} else if ((!strcmp(argv[i], "-nodict")) ||
	         (!strcmp(argv[i], "--nodict"))) {
	    options |= XML_PARSE_NODICT;
	} else if ((!strcmp(argv[i], "-version")) ||
	         (!strcmp(argv[i], "--version"))) {