xmllintコマンドでのXMLスキーマを使ったバリデーションのコードリーディング

  } else if ((!strcmp(argv[i], "-schema")) ||
	   (!strcmp(argv[i], "--schema"))) {
      i++;
      schema = argv[i];
      noent++;

static char * schema = NULL;

	xmlSchemaParserCtxtPtr ctxt;

	if (timing) {
	    startTimer();
	}
	ctxt = xmlSchemaNewParserCtxt(schema);
	xmlSchemaSetParserErrors(ctxt, xmlGenericError, xmlGenericError, NULL);
	wxschemas = xmlSchemaParse(ctxt);
	if (wxschemas == NULL) {
	    xmlGenericError(xmlGenericErrorContext,
		    "WXS schema %s failed to compile\n", schema);
	    progresult = XMLLINT_ERR_SCHEMACOMP;
	    schema = NULL;
	}
	xmlSchemaFreeParserCtxt(ctxt);
	if (timing) {
	    endTimer("Compiling the schemas");
	}

static xmlSchemaPtr wxschemas = NULL;

	int ret;
	xmlSchemaValidCtxtPtr vctxt;

	vctxt = xmlSchemaNewValidCtxt(wxschemas);
	xmlSchemaSetValidErrors(vctxt, xmlGenericError, xmlGenericError, NULL);
	xmlSchemaValidateSetFilename(vctxt, filename);

	ret = xmlSchemaValidateStream(vctxt, buf, 0, handler,
				      (void *)user_data);
	if (repeat == 0) {
	    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(vctxt);

/**
 * xmlParserInputBufferCreateStatic:
 * @mem:  the memory input
 * @size:  the length of the memory block
 * @enc:  the charset encoding if known
 *
 * Create a buffered parser input for the progressive parsing for the input
 * from an immutable memory area. This will not copy the memory area to
 * the buffer, but the memory is expected to be available until the end of
 * the parsing, this is useful for example when using mmap'ed file.
 *
 * Returns the new parser input or NULL
 */
xmlParserInputBufferPtr
xmlParserInputBufferCreateStatic(const char *mem, int size,
                                 xmlCharEncoding enc) {
    xmlParserInputBufferPtr ret;

    if (size < 0) return(NULL);
    if (mem == NULL) return(NULL);

    ret = (xmlParserInputBufferPtr) xmlMalloc(sizeof(xmlParserInputBuffer));
    if (ret == NULL) {
	xmlIOErrMemory("creating input buffer");
	return(NULL);
    }
    memset(ret, 0, (size_t) sizeof(xmlParserInputBuffer));
    ret->buffer = xmlBufCreateStatic((void *)mem, (size_t) size);
    if (ret->buffer == NULL) {
        xmlFree(ret);
	return(NULL);
    }
    ret->encoder = xmlGetCharEncodingHandler(enc);
    if (ret->encoder != NULL)
        ret->raw = xmlBufCreateSize(2 * xmlDefaultBufferSize);
    else
        ret->raw = NULL;
    ret->compressed = -1;
    ret->context = (void *) mem;
    ret->readcallback = NULL;
    ret->closecallback = NULL;

    return(ret);
}

/**
 * xmlFreeParserInputBuffer:
 * @in:  a buffered parser input
 *
 * Free up the memory used by a buffered parser input
 */
void
xmlFreeParserInputBuffer(xmlParserInputBufferPtr in) {
    if (in == NULL) return;

    if (in->raw) {
        xmlBufFree(in->raw);
	in->raw = NULL;
    }
    if (in->encoder != NULL) {
        xmlCharEncCloseFunc(in->encoder);
    }
    if (in->closecallback != NULL) {
	in->closecallback(in->context);
    }
    if (in->buffer != NULL) {
        xmlBufFree(in->buffer);
	in->buffer = NULL;
    }

    xmlFree(in);
}

/**
 * xmlSchemaNewParserCtxt:
 * @URL:  the location of the schema
 *
 * Create an XML Schemas parse context for that file/resource expected
 * to contain an XML Schemas file.
 *
 * Returns the parser context or NULL in case of error
 */
xmlSchemaParserCtxtPtr
xmlSchemaNewParserCtxt(const char *URL)
{
    xmlSchemaParserCtxtPtr ret;

    if (URL == NULL)
	return (NULL);

    ret = xmlSchemaParserCtxtCreate();
    if (ret == NULL)
	return(NULL);
    ret->dict = xmlDictCreate();
    ret->URL = xmlDictLookup(ret->dict, (const xmlChar *) URL, -1);
    return (ret);
}

/**
 * xmlDictLookup:
 * @dict: the dictionary
 * @name: the name of the userdata
 * @len: the length of the name, if -1 it is recomputed
 *
 * Add the @name to the dictionary @dict if not present.
 *
 * Returns the internal copy of the name or NULL in case of internal error
 */
const xmlChar *
xmlDictLookup(xmlDictPtr dict, const xmlChar *name, int len) {
    unsigned long key, okey, nbi = 0;
    xmlDictEntryPtr entry;
    xmlDictEntryPtr insert;
    const xmlChar *ret;
    unsigned int l;

    if ((dict == NULL) || (name == NULL))
	return(NULL);

    if (len < 0)
	l = strlen((const char *) name);
    else
	l = len;

    if (((dict->limit > 0) && (l >= dict->limit)) ||
	(l > INT_MAX / 2))
	return(NULL);

    /*
     * Check for duplicate and insertion location.
     */
    okey = xmlDictComputeKey(dict, name, l);
    key = okey % dict->size;
    if (dict->dict[key].valid == 0) {
	insert = NULL;
    } else {
	for (insert = &(dict->dict[key]); insert->next != NULL;
	     insert = insert->next) {
#ifdef __GNUC__
	    if ((insert->okey == okey) && (insert->len == l)) {
		if (!memcmp(insert->name, name, l))
		    return(insert->name);
	    }
#else
	    if ((insert->okey == okey) && (insert->len == l) &&
		(!xmlStrncmp(insert->name, name, l)))
		return(insert->name);
#endif
	    nbi++;
	}
#ifdef __GNUC__
	if ((insert->okey == okey) && (insert->len == l)) {
	    if (!memcmp(insert->name, name, l))
		return(insert->name);
	}
#else
	if ((insert->okey == okey) && (insert->len == l) &&
	    (!xmlStrncmp(insert->name, name, l)))
	    return(insert->name);
#endif
    }

    if (dict->subdict) {
	unsigned long skey;

	/* we cannot always reuse the same okey for the subdict */
	if (((dict->size == MIN_DICT_SIZE) &&
	     (dict->subdict->size != MIN_DICT_SIZE)) ||
	    ((dict->size != MIN_DICT_SIZE) &&
	     (dict->subdict->size == MIN_DICT_SIZE)))
	    skey = xmlDictComputeKey(dict->subdict, name, l);
	else
	    skey = okey;

	key = skey % dict->subdict->size;
	if (dict->subdict->dict[key].valid != 0) {
	    xmlDictEntryPtr tmp;

	    for (tmp = &(dict->subdict->dict[key]); tmp->next != NULL;
		 tmp = tmp->next) {
#ifdef __GNUC__
		if ((tmp->okey == skey) && (tmp->len == l)) {
		    if (!memcmp(tmp->name, name, l))
			return(tmp->name);
		}
#else
		if ((tmp->okey == skey) && (tmp->len == l) &&
		    (!xmlStrncmp(tmp->name, name, l)))
		    return(tmp->name);
#endif
		nbi++;
	    }
#ifdef __GNUC__
	    if ((tmp->okey == skey) && (tmp->len == l)) {
		if (!memcmp(tmp->name, name, l))
		    return(tmp->name);
	    }
#else
	    if ((tmp->okey == skey) && (tmp->len == l) &&
		(!xmlStrncmp(tmp->name, name, l)))
		return(tmp->name);
#endif
	}
	key = okey % dict->size;
    }

    ret = xmlDictAddString(dict, name, l);
    if (ret == NULL)
	return(NULL);
    if (insert == NULL) {
	entry = &(dict->dict[key]);
    } else {
	entry = xmlMalloc(sizeof(xmlDictEntry));
	if (entry == NULL)
	     return(NULL);
    }
    entry->name = ret;
    entry->len = l;
    entry->next = NULL;
    entry->valid = 1;
    entry->okey = okey;


    if (insert != NULL)
	insert->next = entry;

    dict->nbElems++;

    if ((nbi > MAX_HASH_LEN) &&
	(dict->size <= ((MAX_DICT_HASH / 2) / MAX_HASH_LEN))) {
	if (xmlDictGrow(dict, MAX_HASH_LEN * 2 * dict->size) != 0)
	    return(NULL);
    }
    /* Note that entry may have been freed at this point by xmlDictGrow */

    return(ret);
}

#define MAX_HASH_LEN 3
#define MIN_DICT_SIZE 128
#define MAX_DICT_HASH 8 * 2048
#define WITH_BIG_KEY

#ifdef WITH_BIG_KEY
#define xmlDictComputeKey(dict, name, len)                              \
    (((dict)->size == MIN_DICT_SIZE) ?                                  \
     xmlDictComputeFastKey(name, len, (dict)->seed) :                   \
     xmlDictComputeBigKey(name, len, (dict)->seed))

#define xmlDictComputeQKey(dict, prefix, plen, name, len)               \
    (((prefix) == NULL) ?                                               \
      (xmlDictComputeKey(dict, name, len)) :                             \
      (((dict)->size == MIN_DICT_SIZE) ?                                \
       xmlDictComputeFastQKey(prefix, plen, name, len, (dict)->seed) :	\
       xmlDictComputeBigQKey(prefix, plen, name, len, (dict)->seed)))

#else /* !WITH_BIG_KEY */
#define xmlDictComputeKey(dict, name, len)                              \
        xmlDictComputeFastKey(name, len, (dict)->seed)
#define xmlDictComputeQKey(dict, prefix, plen, name, len)               \
        xmlDictComputeFastQKey(prefix, plen, name, len, (dict)->seed)
#endif /* WITH_BIG_KEY */

/*
 * xmlDictAddString:
 * @dict: the dictionary
 * @name: the name of the userdata
 * @len: the length of the name
 *
 * Add the string to the array[s]
 *
 * Returns the pointer of the local string, or NULL in case of error.
 */
static const xmlChar *
xmlDictAddString(xmlDictPtr dict, const xmlChar *name, unsigned int namelen) {
    xmlDictStringsPtr pool;
    const xmlChar *ret;
    size_t size = 0; /* + sizeof(_xmlDictStrings) == 1024 */
    size_t limit = 0;

#ifdef DICT_DEBUG_PATTERNS
    fprintf(stderr, "-");
#endif
    pool = dict->strings;
    while (pool != NULL) {
	if ((size_t)(pool->end - pool->free) > namelen)
	    goto found_pool;
	if (pool->size > size) size = pool->size;
        limit += pool->size;
	pool = pool->next;
    }
    /*
     * Not found, need to allocate
     */
    if (pool == NULL) {
        if ((dict->limit > 0) && (limit > dict->limit)) {
            return(NULL);
        }

        if (size == 0) size = 1000;
	else size *= 4; /* exponential growth */
        if (size < 4 * namelen)
	    size = 4 * namelen; /* just in case ! */
	pool = (xmlDictStringsPtr) xmlMalloc(sizeof(xmlDictStrings) + size);
	if (pool == NULL)
	    return(NULL);
	pool->size = size;
	pool->nbStrings = 0;
	pool->free = &pool->array[0];
	pool->end = &pool->array[size];
	pool->next = dict->strings;
	dict->strings = pool;
#ifdef DICT_DEBUG_PATTERNS
        fprintf(stderr, "+");
#endif
    }
found_pool:
    ret = pool->free;
    memcpy(pool->free, name, namelen);
    pool->free += namelen;
    *(pool->free++) = 0;
    pool->nbStrings++;
    return(ret);
}

/**
 * xmlSchemaParse:
 * @ctxt:  a schema validation context
 *
 * parse a schema definition resource and build an internal
 * XML Schema structure which can be used to validate instances.
 *
 * Returns the internal XML Schema structure built from the resource or
 *         NULL in case of error
 */
xmlSchemaPtr
xmlSchemaParse(xmlSchemaParserCtxtPtr ctxt)
{
    xmlSchemaPtr mainSchema = NULL;
    xmlSchemaBucketPtr bucket = NULL;
    int res;

    /*
    * This one is used if the schema to be parsed was specified via
    * the API; i.e. not automatically by the validated instance document.
    */

    xmlSchemaInitTypes();

    if (ctxt == NULL)
        return (NULL);

    /* TODO: Init the context. Is this all we need?*/
    ctxt->nberrors = 0;
    ctxt->err = 0;
    ctxt->counter = 0;

    /* Create the *main* schema. */
    mainSchema = xmlSchemaNewSchema(ctxt);
    if (mainSchema == NULL)
	goto exit_failure;
    /*
    * Create the schema constructor.
    */
    if (ctxt->constructor == NULL) {
	ctxt->constructor = xmlSchemaConstructionCtxtCreate(ctxt->dict);
	if (ctxt->constructor == NULL)
	    return(NULL);
	/* Take ownership of the constructor to be able to free it. */
	ctxt->ownsConstructor = 1;
    }
    ctxt->constructor->mainSchema = mainSchema;
    /*
    * Locate and add the schema document.
    */
    res = xmlSchemaAddSchemaDoc(ctxt, XML_SCHEMA_SCHEMA_MAIN,
	ctxt->URL, ctxt->doc, ctxt->buffer, ctxt->size, NULL,
	NULL, NULL, &bucket);
    if (res == -1)
	goto exit_failure;
    if (res != 0)
	goto exit;

    if (bucket == NULL) {
	/* TODO: Error code, actually we failed to *locate* the schema. */
	if (ctxt->URL)
	    xmlSchemaCustomErr(ACTXT_CAST ctxt, XML_SCHEMAP_FAILED_LOAD,
		NULL, NULL,
		"Failed to locate the main schema resource at '%s'",
		ctxt->URL, NULL);
	else
	    xmlSchemaCustomErr(ACTXT_CAST ctxt, XML_SCHEMAP_FAILED_LOAD,
		NULL, NULL,
		"Failed to locate the main schema resource",
		    NULL, NULL);
	goto exit;
    }
    /* Then do the parsing for good. */
    if (xmlSchemaParseNewDocWithContext(ctxt, mainSchema, bucket) == -1)
	goto exit_failure;
    if (ctxt->nberrors != 0)
	goto exit;

    mainSchema->doc = bucket->doc;
    mainSchema->preserve = ctxt->preserve;

    ctxt->schema = mainSchema;

    if (xmlSchemaFixupComponents(ctxt, WXS_CONSTRUCTOR(ctxt)->mainBucket) == -1)
	goto exit_failure;

    /*
    * TODO: This is not nice, since we cannot distinguish from the
    * result if there was an internal error or not.
    */
exit:
    if (ctxt->nberrors != 0) {
	if (mainSchema) {
	    xmlSchemaFree(mainSchema);
	    mainSchema = NULL;
	}
	if (ctxt->constructor) {
	    xmlSchemaConstructionCtxtFree(ctxt->constructor);
	    ctxt->constructor = NULL;
	    ctxt->ownsConstructor = 0;
	}
    }
    ctxt->schema = NULL;
    return(mainSchema);
exit_failure:
    /*
    * Quite verbose, but should catch internal errors, which were
    * not communicated.
    */
    if (mainSchema) {
        xmlSchemaFree(mainSchema);
	mainSchema = NULL;
    }
    if (ctxt->constructor) {
	xmlSchemaConstructionCtxtFree(ctxt->constructor);
	ctxt->constructor = NULL;
	ctxt->ownsConstructor = 0;
    }
    PERROR_INT2("xmlSchemaParse",
	"An internal error occurred");
    ctxt->schema = NULL;
    return(NULL);
}

/**
 * xmlSchemaAddSchemaDoc:
 * @pctxt:  a schema validation context
 * @schema:  the schema being built
 * @node:  a subtree containing XML Schema information
 *
 * Parse an included (and to-be-redefined) XML schema document.
 *
 * Returns 0 on success, a positive error code on errors and
 *         -1 in case of an internal or API error.
 */

static int
xmlSchemaAddSchemaDoc(xmlSchemaParserCtxtPtr pctxt,
		int type, /* import or include or redefine */
		const xmlChar *schemaLocation,
		xmlDocPtr schemaDoc,
		const char *schemaBuffer,
		int schemaBufferLen,
		xmlNodePtr invokingNode,
		const xmlChar *sourceTargetNamespace,
		const xmlChar *importNamespace,
		xmlSchemaBucketPtr *bucket)
{
    const xmlChar *targetNamespace = NULL;
    xmlSchemaSchemaRelationPtr relation = NULL;
    xmlDocPtr doc = NULL;
    int res = 0, err = 0, located = 0, preserveDoc = 0;
    xmlSchemaBucketPtr bkt = NULL;

    if (bucket != NULL)
	*bucket = NULL;

    switch (type) {
	case XML_SCHEMA_SCHEMA_IMPORT:
	case XML_SCHEMA_SCHEMA_MAIN:
	    err = XML_SCHEMAP_SRC_IMPORT;
	    break;
	case XML_SCHEMA_SCHEMA_INCLUDE:
	    err = XML_SCHEMAP_SRC_INCLUDE;
	    break;
	case XML_SCHEMA_SCHEMA_REDEFINE:
	    err = XML_SCHEMAP_SRC_REDEFINE;
	    break;
    }


    /* Special handling for the main schema:
    * skip the location and relation logic and just parse the doc.
    * We need just a bucket to be returned in this case.
    */
    if ((type == XML_SCHEMA_SCHEMA_MAIN) || (! WXS_HAS_BUCKETS(pctxt)))
	goto doc_load;

    /* Note that we expect the location to be an absolute URI. */
    if (schemaLocation != NULL) {
	bkt = xmlSchemaGetSchemaBucket(pctxt, schemaLocation);
	if ((bkt != NULL) &&
	    (pctxt->constructor->bucket == bkt)) {
	    /* Report self-imports/inclusions/redefinitions. */

	    xmlSchemaCustomErr(ACTXT_CAST pctxt, err,
		invokingNode, NULL,
		"The schema must not import/include/redefine itself",
		NULL, NULL);
	    goto exit;
	}
    }
    /*
    * Create a relation for the graph of schemas.
    */
    relation = xmlSchemaSchemaRelationCreate();
    if (relation == NULL)
	return(-1);
    xmlSchemaSchemaRelationAddChild(pctxt->constructor->bucket,
	relation);
    relation->type = type;

    /*
    * Save the namespace import information.
    */
    if (WXS_IS_BUCKET_IMPMAIN(type)) {
	relation->importNamespace = importNamespace;
	if (schemaLocation == NULL) {
	    /*
	    * No location; this is just an import of the namespace.
	    * Note that we don't assign a bucket to the relation
	    * in this case.
	    */
	    goto exit;
	}
	targetNamespace = importNamespace;
    }

    /* Did we already fetch the doc? */
    if (bkt != NULL) {
	if ((WXS_IS_BUCKET_IMPMAIN(type)) && (! bkt->imported)) {
	    /*
	    * We included/redefined and then try to import a schema,
	    * but the new location provided for import was different.
	    */
	    if (schemaLocation == NULL)
		schemaLocation = BAD_CAST "in_memory_buffer";
	    if (!xmlStrEqual(schemaLocation,
		bkt->schemaLocation)) {
		xmlSchemaCustomErr(ACTXT_CAST pctxt, err,
		    invokingNode, NULL,
		    "The schema document '%s' cannot be imported, since "
		    "it was already included or redefined",
		    schemaLocation, NULL);
		goto exit;
	    }
	} else if ((! WXS_IS_BUCKET_IMPMAIN(type)) && (bkt->imported)) {
	    /*
	    * We imported and then try to include/redefine a schema,
	    * but the new location provided for the include/redefine
	    * was different.
	    */
	    if (schemaLocation == NULL)
		schemaLocation = BAD_CAST "in_memory_buffer";
	    if (!xmlStrEqual(schemaLocation,
		bkt->schemaLocation)) {
		xmlSchemaCustomErr(ACTXT_CAST pctxt, err,
		    invokingNode, NULL,
		    "The schema document '%s' cannot be included or "
		    "redefined, since it was already imported",
		    schemaLocation, NULL);
		goto exit;
	    }
	}
    }

    if (WXS_IS_BUCKET_IMPMAIN(type)) {
	/*
	* Given that the schemaLocation [attribute] is only a hint, it is open
	* to applications to ignore all but the first <import> for a given
	* namespace, regardless of the `actual value` of schemaLocation, but
	* such a strategy risks missing useful information when new
	* schemaLocations are offered.
	*
	* We will use the first <import> that comes with a location.
	* Further <import>s *with* a location, will result in an error.
	* TODO: Better would be to just report a warning here, but
	* we'll try it this way until someone complains.
	*
	* Schema Document Location Strategy:
	* 3 Based on the namespace name, identify an existing schema document,
	* either as a resource which is an XML document or a <schema> element
	* information item, in some local schema repository;
	* 5 Attempt to resolve the namespace name to locate such a resource.
	*
	* NOTE: (3) and (5) are not supported.
	*/
	if (bkt != NULL) {
	    relation->bucket = bkt;
	    goto exit;
	}
	bkt = xmlSchemaGetSchemaBucketByTNS(pctxt,
	    importNamespace, 1);

	if (bkt != NULL) {
	    relation->bucket = bkt;
	    if (bkt->schemaLocation == NULL) {
		/* First given location of the schema; load the doc. */
		bkt->schemaLocation = schemaLocation;
	    } else {
		if (!xmlStrEqual(schemaLocation,
		    bkt->schemaLocation)) {
		    /*
		    * Additional location given; just skip it.
		    * URGENT TODO: We should report a warning here.
		    * res = XML_SCHEMAP_SRC_IMPORT;
		    */
		    if (schemaLocation == NULL)
			schemaLocation = BAD_CAST "in_memory_buffer";

		    xmlSchemaCustomWarning(ACTXT_CAST pctxt,
			XML_SCHEMAP_WARN_SKIP_SCHEMA,
			invokingNode, NULL,
			"Skipping import of schema located at '%s' for the "
			"namespace '%s', since this namespace was already "
			"imported with the schema located at '%s'",
			schemaLocation, importNamespace, bkt->schemaLocation);
		}
		goto exit;
	    }
	}
	/*
	* No bucket + first location: load the doc and create a
	* bucket.
	*/
    } else {
	/* <include> and <redefine> */
	if (bkt != NULL) {

	    if ((bkt->origTargetNamespace == NULL) &&
		(bkt->targetNamespace != sourceTargetNamespace)) {
		xmlSchemaBucketPtr chamel;

		/*
		* Chameleon include/redefine: skip loading only if it was
		* already build for the targetNamespace of the including
		* schema.
		*/
		/*
		* URGENT TODO: If the schema is a chameleon-include then copy
		* the components into the including schema and modify the
		* targetNamespace of those components, do nothing otherwise.
		* NOTE: This is currently worked-around by compiling the
		* chameleon for every distinct including targetNamespace; thus
		* not performant at the moment.
		* TODO: Check when the namespace in wildcards for chameleons
		* needs to be converted: before we built wildcard intersections
		* or after.
		*   Answer: after!
		*/
		chamel = xmlSchemaGetChameleonSchemaBucket(pctxt,
		    schemaLocation, sourceTargetNamespace);
		if (chamel != NULL) {
		    /* A fitting chameleon was already parsed; NOP. */
		    relation->bucket = chamel;
		    goto exit;
		}
		/*
		* We need to parse the chameleon again for a different
		* targetNamespace.
		* CHAMELEON TODO: Optimize this by only parsing the
		* chameleon once, and then copying the components to
		* the new targetNamespace.
		*/
		bkt = NULL;
	    } else {
		relation->bucket = bkt;
		goto exit;
	    }
	}
    }
    if ((bkt != NULL) && (bkt->doc != NULL)) {
	PERROR_INT("xmlSchemaAddSchemaDoc",
	    "trying to load a schema doc, but a doc is already "
	    "assigned to the schema bucket");
	goto exit_failure;
    }

doc_load:
    /*
    * Load the document.
    */
    if (schemaDoc != NULL) {
	doc = schemaDoc;
	/* Don' free this one, since it was provided by the caller. */
	preserveDoc = 1;
	/* TODO: Does the context or the doc hold the location? */
	if (schemaDoc->URL != NULL)
	    schemaLocation = xmlDictLookup(pctxt->dict,
		schemaDoc->URL, -1);
        else
	    schemaLocation = BAD_CAST "in_memory_buffer";
    } else if ((schemaLocation != NULL) || (schemaBuffer != NULL)) {
	xmlParserCtxtPtr parserCtxt;

	parserCtxt = xmlNewParserCtxt();
	if (parserCtxt == NULL) {
	    xmlSchemaPErrMemory(NULL, "xmlSchemaGetDoc, "
		"allocating a parser context", NULL);
	    goto exit_failure;
	}
	if ((pctxt->dict != NULL) && (parserCtxt->dict != NULL)) {
	    /*
	    * TODO: Do we have to burden the schema parser dict with all
	    * the content of the schema doc?
	    */
	    xmlDictFree(parserCtxt->dict);
	    parserCtxt->dict = pctxt->dict;
	    xmlDictReference(parserCtxt->dict);
	}
	if (schemaLocation != NULL) {
	    /* Parse from file. */
	    doc = xmlCtxtReadFile(parserCtxt, (const char *) schemaLocation,
		NULL, SCHEMAS_PARSE_OPTIONS);
	} else if (schemaBuffer != NULL) {
	    /* Parse from memory buffer. */
	    doc = xmlCtxtReadMemory(parserCtxt, schemaBuffer, schemaBufferLen,
		NULL, NULL, SCHEMAS_PARSE_OPTIONS);
	    schemaLocation = BAD_CAST "in_memory_buffer";
	    if (doc != NULL)
		doc->URL = xmlStrdup(schemaLocation);
	}
	/*
	* For <import>:
	* 2.1 The referent is (a fragment of) a resource which is an
	* XML document (see clause 1.1), which in turn corresponds to
	* a <schema> element information item in a well-formed information
	* set, which in turn corresponds to a valid schema.
	* TODO: (2.1) fragments of XML documents are not supported.
	*
	* 2.2 The referent is a <schema> element information item in
	* a well-formed information set, which in turn corresponds
	* to a valid schema.
	* TODO: (2.2) is not supported.
	*/
	if (doc == NULL) {
	    xmlErrorPtr lerr;
	    lerr = xmlGetLastError();
	    /*
	    * Check if this a parser error, or if the document could
	    * just not be located.
	    * TODO: Try to find specific error codes to react only on
	    * localisation failures.
	    */
	    if ((lerr == NULL) || (lerr->domain != XML_FROM_IO)) {
		/*
		* We assume a parser error here.
		*/
		located = 1;
		/* TODO: Error code ?? */
		res = XML_SCHEMAP_SRC_IMPORT_2_1;
		xmlSchemaCustomErr(ACTXT_CAST pctxt, res,
		    invokingNode, NULL,
		    "Failed to parse the XML resource '%s'",
		    schemaLocation, NULL);
	    }
	}
	xmlFreeParserCtxt(parserCtxt);
	if ((doc == NULL) && located)
	    goto exit_error;
    } else {
	xmlSchemaPErr(pctxt, NULL,
	    XML_SCHEMAP_NOTHING_TO_PARSE,
	    "No information for parsing was provided with the "
	    "given schema parser context.\n",
	    NULL, NULL);
	goto exit_failure;
    }
    /*
    * Preprocess the document.
    */
    if (doc != NULL) {
	xmlNodePtr docElem = NULL;

	located = 1;
	docElem = xmlDocGetRootElement(doc);
	if (docElem == NULL) {
	    xmlSchemaCustomErr(ACTXT_CAST pctxt, XML_SCHEMAP_NOROOT,
		invokingNode, NULL,
		"The document '%s' has no document element",
		schemaLocation, NULL);
	    goto exit_error;
	}
	/*
	* Remove all the blank text nodes.
	*/
	xmlSchemaCleanupDoc(pctxt, docElem);
	/*
	* Check the schema's top level element.
	*/
	if (!IS_SCHEMA(docElem, "schema")) {
	    xmlSchemaCustomErr(ACTXT_CAST pctxt, XML_SCHEMAP_NOT_SCHEMA,
		invokingNode, NULL,
		"The XML document '%s' is not a schema document",
		schemaLocation, NULL);
	    goto exit_error;
	}
	/*
	* Note that we don't apply a type check for the
	* targetNamespace value here.
	*/
	targetNamespace = xmlSchemaGetProp(pctxt, docElem,
	    "targetNamespace");
    }

/* after_doc_loading: */
    if ((bkt == NULL) && located) {
	/* Only create a bucket if the schema was located. */
        bkt = xmlSchemaBucketCreate(pctxt, type,
	    targetNamespace);
	if (bkt == NULL)
	    goto exit_failure;
    }
    if (bkt != NULL) {
	bkt->schemaLocation = schemaLocation;
	bkt->located = located;
	if (doc != NULL) {
	    bkt->doc = doc;
	    bkt->targetNamespace = targetNamespace;
	    bkt->origTargetNamespace = targetNamespace;
	    if (preserveDoc)
		bkt->preserveDoc = 1;
	}
	if (WXS_IS_BUCKET_IMPMAIN(type))
	    bkt->imported++;
	    /*
	    * Add it to the graph of schemas.
	    */
	if (relation != NULL)
	    relation->bucket = bkt;
    }

exit:
    /*
    * Return the bucket explicitly; this is needed for the
    * main schema.
    */
    if (bucket != NULL)
	*bucket = bkt;
    return (0);

exit_error:
    if ((doc != NULL) && (! preserveDoc)) {
	xmlFreeDoc(doc);
	if (bkt != NULL)
	    bkt->doc = NULL;
    }
    return(pctxt->err);

exit_failure:
    if ((doc != NULL) && (! preserveDoc)) {
	xmlFreeDoc(doc);
	if (bkt != NULL)
	    bkt->doc = NULL;
    }
    return (-1);
}

	if (schemaLocation != NULL) {
	    /* Parse from file. */
	    doc = xmlCtxtReadFile(parserCtxt, (const char *) schemaLocation,
		NULL, SCHEMAS_PARSE_OPTIONS);
	} else if (schemaBuffer != NULL) {
	    /* Parse from memory buffer. */
	    doc = xmlCtxtReadMemory(parserCtxt, schemaBuffer, schemaBufferLen,
		NULL, NULL, SCHEMAS_PARSE_OPTIONS);
	    schemaLocation = BAD_CAST "in_memory_buffer";
	    if (doc != NULL)
		doc->URL = xmlStrdup(schemaLocation);
	}

typedef struct _xmlSchemaParserCtxt xmlSchemaParserCtxt;
typedef xmlSchemaParserCtxt *xmlSchemaParserCtxtPtr;

struct _xmlSchemaParserCtxt {
    int type;
    void *errCtxt;             /* user specific error context */
    xmlSchemaValidityErrorFunc error;   /* the callback in case of errors */
    xmlSchemaValidityWarningFunc warning;       /* the callback in case of warning */
    int err;
    int nberrors;
    xmlStructuredErrorFunc serror;

    xmlSchemaConstructionCtxtPtr constructor;
    int ownsConstructor; /* TODO: Move this to parser *flags*. */

    /* xmlSchemaPtr topschema;	*/
    /* xmlHashTablePtr namespaces;  */

    xmlSchemaPtr schema;        /* The main schema in use */
    int counter;

    const xmlChar *URL;
    xmlDocPtr doc;
    int preserve;		/* Whether the doc should be freed  */

    const char *buffer;
    int size;

    /*
     * Used to build complex element content models
     */
    xmlAutomataPtr am;
    xmlAutomataStatePtr start;
    xmlAutomataStatePtr end;
    xmlAutomataStatePtr state;

    xmlDictPtr dict;		/* dictionary for interned string names */
    xmlSchemaTypePtr ctxtType; /* The current context simple/complex type */
    int options;
    xmlSchemaValidCtxtPtr vctxt;
    int isS4S;
    int isRedefine;
    int xsiAssemble;
    int stop; /* If the parser should stop; i.e. a critical error. */
    const xmlChar *targetNamespace;
    xmlSchemaBucketPtr redefined; /* The schema to be redefined. */

    xmlSchemaRedefPtr redef; /* Used for redefinitions. */
    int redefCounter; /* Used for redefinitions. */
    xmlSchemaItemListPtr attrProhibs;
};

static int
xmlSchemaParseNewDocWithContext(xmlSchemaParserCtxtPtr pctxt,
		     xmlSchemaPtr schema,
		     xmlSchemaBucketPtr bucket)
{
    int oldFlags;
    xmlDocPtr oldDoc;
    xmlNodePtr node;
    int ret, oldErrs;
    xmlSchemaBucketPtr oldbucket = pctxt->constructor->bucket;

    /*
    * Save old values; reset the *main* schema.
    * URGENT TODO: This is not good; move the per-document information
    * to the parser. Get rid of passing the main schema to the
    * parsing functions.
    */
    oldFlags = schema->flags;
    oldDoc = schema->doc;
    if (schema->flags != 0)
	xmlSchemaClearSchemaDefaults(schema);
    schema->doc = bucket->doc;
    pctxt->schema = schema;
    /*
    * Keep the current target namespace on the parser *not* on the
    * main schema.
    */
    pctxt->targetNamespace = bucket->targetNamespace;
    WXS_CONSTRUCTOR(pctxt)->bucket = bucket;

    if ((bucket->targetNamespace != NULL) &&
	xmlStrEqual(bucket->targetNamespace, xmlSchemaNs)) {
	/*
	* We are parsing the schema for schemas!
	*/
	pctxt->isS4S = 1;
    }
    /* Mark it as parsed, even if parsing fails. */
    bucket->parsed++;
    /* Compile the schema doc. */
    node = xmlDocGetRootElement(bucket->doc);
    ret = xmlSchemaParseSchemaElement(pctxt, schema, node);
    if (ret != 0)
	goto exit;
    /* An empty schema; just get out. */
    if (node->children == NULL)
	goto exit;
    oldErrs = pctxt->nberrors;
    ret = xmlSchemaParseSchemaTopLevel(pctxt, schema, node->children);
    if (ret != 0)
	goto exit;
    /*
    * TODO: Not nice, but I'm not 100% sure we will get always an error
    * as a result of the above functions; so better rely on pctxt->err
    * as well.
    */
    if ((ret == 0) && (oldErrs != pctxt->nberrors)) {
	ret = pctxt->err;
	goto exit;
    }

exit:
    WXS_CONSTRUCTOR(pctxt)->bucket = oldbucket;
    /* Restore schema values. */
    schema->doc = oldDoc;
    schema->flags = oldFlags;
    return(ret);
}

/**
 * xmlParserInputBufferCreateFilename:
 * @URI:  a C string containing the URI or filename
 * @enc:  the charset encoding if known
 *
 * Create a buffered parser input for the progressive parsing of a file
 * If filename is "-' then we use stdin as the input.
 * Automatic support for ZLIB/Compress compressed document is provided
 * by default if found at compile-time.
 * Do an encoding check if enc == XML_CHAR_ENCODING_NONE
 *
 * Returns the new parser input or NULL
 */
xmlParserInputBufferPtr
xmlParserInputBufferCreateFilename(const char *URI, xmlCharEncoding enc) {
    if ((xmlParserInputBufferCreateFilenameValue)) {
		return xmlParserInputBufferCreateFilenameValue(URI, enc);
	}
	return __xmlParserInputBufferCreateFilename(URI, enc);
}

xmlParserInputBufferPtr
__xmlParserInputBufferCreateFilename(const char *URI, xmlCharEncoding enc) {
    xmlParserInputBufferPtr ret;
    int i = 0;
    void *context = NULL;

    if (xmlInputCallbackInitialized == 0)
	xmlRegisterDefaultInputCallbacks();

    if (URI == NULL) return(NULL);

    /*
     * Try to find one of the input accept method accepting that scheme
     * Go in reverse to give precedence to user defined handlers.
     */
    if (context == NULL) {
	for (i = xmlInputCallbackNr - 1;i >= 0;i--) {
	    if ((xmlInputCallbackTable[i].matchcallback != NULL) &&
		(xmlInputCallbackTable[i].matchcallback(URI) != 0)) {
		context = xmlInputCallbackTable[i].opencallback(URI);
		if (context != NULL) {
		    break;
		}
	    }
	}
    }
    if (context == NULL) {
	return(NULL);
    }

    /*
     * Allocate the Input buffer front-end.
     */
    ret = xmlAllocParserInputBuffer(enc);
    if (ret != NULL) {
	ret->context = context;
	ret->readcallback = xmlInputCallbackTable[i].readcallback;
	ret->closecallback = xmlInputCallbackTable[i].closecallback;
#ifdef LIBXML_ZLIB_ENABLED
	if ((xmlInputCallbackTable[i].opencallback == xmlGzfileOpen) &&
		(strcmp(URI, "-") != 0)) {
#if defined(ZLIB_VERNUM) && ZLIB_VERNUM >= 0x1230
            ret->compressed = !gzdirect(context);
#else
	    if (((z_stream *)context)->avail_in > 4) {
	        char *cptr, buff4[4];
		cptr = (char *) ((z_stream *)context)->next_in;
		if (gzread(context, buff4, 4) == 4) {
		    if (strncmp(buff4, cptr, 4) == 0)
		        ret->compressed = 0;
		    else
		        ret->compressed = 1;
		    gzrewind(context);
		}
	    }
#endif
	}
#endif
#ifdef LIBXML_LZMA_ENABLED
	if ((xmlInputCallbackTable[i].opencallback == xmlXzfileOpen) &&
		(strcmp(URI, "-") != 0)) {
            ret->compressed = __libxml2_xzcompressed(context);
	}
#endif
    }
    else
      xmlInputCallbackTable[i].closecallback (context);

    return(ret);
}

/*
 * Input I/O callback sets
 */
typedef struct _xmlInputCallback {
    xmlInputMatchCallback matchcallback;
    xmlInputOpenCallback opencallback;
    xmlInputReadCallback readcallback;
    xmlInputCloseCallback closecallback;
} xmlInputCallback;

#define MAX_INPUT_CALLBACK 15

static xmlInputCallback xmlInputCallbackTable[MAX_INPUT_CALLBACK];
static int xmlInputCallbackNr = 0;
static int xmlInputCallbackInitialized = 0;

/*
 * Those are the functions and datatypes for the parser input
 * I/O structures.
 */

/**
 * xmlInputMatchCallback:
 * @filename: the filename or URI
 *
 * Callback used in the I/O Input API to detect if the current handler
 * can provide input functionality for this resource.
 *
 * Returns 1 if yes and 0 if another Input module should be used
 */
typedef int (XMLCALL *xmlInputMatchCallback) (char const *filename);
/**
 * xmlInputOpenCallback:
 * @filename: the filename or URI
 *
 * Callback used in the I/O Input API to open the resource
 *
 * Returns an Input context or NULL in case or error
 */
typedef void * (XMLCALL *xmlInputOpenCallback) (char const *filename);
/**
 * xmlInputReadCallback:
 * @context:  an Input context
 * @buffer:  the buffer to store data read
 * @len:  the length of the buffer in bytes
 *
 * Callback used in the I/O Input API to read the resource
 *
 * Returns the number of bytes read or -1 in case of error
 */
typedef int (XMLCALL *xmlInputReadCallback) (void * context, char * buffer, int len);
/**
 * xmlInputCloseCallback:
 * @context:  an Input context
 *
 * Callback used in the I/O Input API to close the resource
 *
 * Returns 0 or -1 in case of error
 */
typedef int (XMLCALL *xmlInputCloseCallback) (void * context);

/**
 * xmlRegisterDefaultInputCallbacks:
 *
 * Registers the default compiled-in I/O handlers.
 */
void
xmlRegisterDefaultInputCallbacks(void) {
    if (xmlInputCallbackInitialized)
	return;

    xmlRegisterInputCallbacks(xmlFileMatch, xmlFileOpen,
	                      xmlFileRead, xmlFileClose);
#ifdef LIBXML_ZLIB_ENABLED
    xmlRegisterInputCallbacks(xmlGzfileMatch, xmlGzfileOpen,
	                      xmlGzfileRead, xmlGzfileClose);
#endif /* LIBXML_ZLIB_ENABLED */
#ifdef LIBXML_LZMA_ENABLED
    xmlRegisterInputCallbacks(xmlXzfileMatch, xmlXzfileOpen,
	                      xmlXzfileRead, xmlXzfileClose);
#endif /* LIBXML_LZMA_ENABLED */

#ifdef LIBXML_HTTP_ENABLED
    xmlRegisterInputCallbacks(xmlIOHTTPMatch, xmlIOHTTPOpen,
	                      xmlIOHTTPRead, xmlIOHTTPClose);
#endif /* LIBXML_HTTP_ENABLED */

#ifdef LIBXML_FTP_ENABLED
    xmlRegisterInputCallbacks(xmlIOFTPMatch, xmlIOFTPOpen,
	                      xmlIOFTPRead, xmlIOFTPClose);
#endif /* LIBXML_FTP_ENABLED */
    xmlInputCallbackInitialized = 1;
}

/**
 * xmlFileMatch:
 * @filename:  the URI for matching
 *
 * input from FILE *
 *
 * Returns 1 if matches, 0 otherwise
 */
int
xmlFileMatch (const char *filename ATTRIBUTE_UNUSED) {
    return(1);
}

/**
 * xmlFileOpen:
 * @filename:  the URI for matching
 *
 * Wrapper around xmlFileOpen_real that try it with an unescaped
 * version of @filename, if this fails fallback to @filename
 *
 * Returns a handler or NULL in case or failure
 */
void *
xmlFileOpen (const char *filename) {
    char *unescaped;
    void *retval;

    retval = xmlFileOpen_real(filename);
    if (retval == NULL) {
	unescaped = xmlURIUnescapeString(filename, 0, NULL);
	if (unescaped != NULL) {
	    retval = xmlFileOpen_real(unescaped);
	    xmlFree(unescaped);
	}
    }

    return retval;
}

/**
 * xmlFileOpen_real:
 * @filename:  the URI for matching
 *
 * input from FILE *, supports compressed input
 * if @filename is " " then the standard input is used
 *
 * Returns an I/O context or NULL in case of error
 */
static void *
xmlFileOpen_real (const char *filename) {
    const char *path = filename;
    FILE *fd;

    if (filename == NULL)
        return(NULL);

    if (!strcmp(filename, "-")) {
	fd = stdin;
	return((void *) fd);
    }

    if (!xmlStrncasecmp(BAD_CAST filename, BAD_CAST "file://localhost/", 17)) {
#if defined (_WIN32) || defined (__DJGPP__) && !defined(__CYGWIN__)
	path = &filename[17];
#else
	path = &filename[16];
#endif
    } else if (!xmlStrncasecmp(BAD_CAST filename, BAD_CAST "file:///", 8)) {
#if defined (_WIN32) || defined (__DJGPP__) && !defined(__CYGWIN__)
	path = &filename[8];
#else
	path = &filename[7];
#endif
    } else if (!xmlStrncasecmp(BAD_CAST filename, BAD_CAST "file:/", 6)) {
        /* lots of generators seems to lazy to read RFC 1738 */
#if defined (_WIN32) || defined (__DJGPP__) && !defined(__CYGWIN__)
	path = &filename[6];
#else
	path = &filename[5];
#endif
    }

    /* Do not check DDNAME on zOS ! */
#if !defined(__MVS__)
    if (!xmlCheckFilename(path))
        return(NULL);
#endif

#if defined(_WIN32) || defined (__DJGPP__) && !defined (__CYGWIN__)
    fd = xmlWrapOpenUtf8(path, 0);
#else
    fd = fopen(path, "r");
#endif /* WIN32 */
    if (fd == NULL) xmlIOErr(0, path);
    return((void *) fd);
}

/**
 * xmlFileRead:
 * @context:  the I/O context
 * @buffer:  where to drop data
 * @len:  number of bytes to write
 *
 * Read @len bytes to @buffer from the I/O channel.
 *
 * Returns the number of bytes written or < 0 in case of failure
 */
int
xmlFileRead (void * context, char * buffer, int len) {
    int ret;
    if ((context == NULL) || (buffer == NULL))
        return(-1);
    ret = fread(&buffer[0], 1,  len, (FILE *) context);
    if (ret < 0) xmlIOErr(0, "fread()");
    return(ret);
}

/**
 * xmlAllocParserInputBuffer:
 * @enc:  the charset encoding if known
 *
 * Create a buffered parser input for progressive parsing
 *
 * Returns the new parser input or NULL
 */
xmlParserInputBufferPtr
xmlAllocParserInputBuffer(xmlCharEncoding enc) {
    xmlParserInputBufferPtr ret;

    ret = (xmlParserInputBufferPtr) xmlMalloc(sizeof(xmlParserInputBuffer));
    if (ret == NULL) {
	xmlIOErrMemory("creating input buffer");
	return(NULL);
    }
    memset(ret, 0, (size_t) sizeof(xmlParserInputBuffer));
    ret->buffer = xmlBufCreateSize(2 * xmlDefaultBufferSize);
    if (ret->buffer == NULL) {
        xmlFree(ret);
	return(NULL);
    }
    xmlBufSetAllocationScheme(ret->buffer, XML_BUFFER_ALLOC_DOUBLEIT);
    ret->encoder = xmlGetCharEncodingHandler(enc);
    if (ret->encoder != NULL)
        ret->raw = xmlBufCreateSize(2 * xmlDefaultBufferSize);
    else
        ret->raw = NULL;
    ret->readcallback = NULL;
    ret->closecallback = NULL;
    ret->context = NULL;
    ret->compressed = -1;
    ret->rawconsumed = 0;

    return(ret);
}