htmltree.c.svn-base

这是一个用于解析xml文件的类库。使用这个类库

void xmlNsListDumpOutput(xmlOutputBufferPtr buf, xmlNsPtr cur);

/**
 * htmlDtdDumpOutput:
 * @buf:  the HTML buffer output
 * @doc:  the document
 * @encoding:  the encoding string
 * 
 * TODO: check whether encoding is needed
 *
 * Dump the HTML document DTD, if any.
 */
static void
htmlDtdDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc,
	          const char *encoding ATTRIBUTE_UNUSED) {
    xmlDtdPtr cur = doc->intSubset;

    if (cur == NULL) {
	htmlSaveErr(XML_SAVE_NO_DOCTYPE, (xmlNodePtr) doc, NULL);
	return;
    }
    xmlOutputBufferWriteString(buf, "<!DOCTYPE ");
    xmlOutputBufferWriteString(buf, (const char *)cur->name);
    if (cur->ExternalID != NULL) {
	xmlOutputBufferWriteString(buf, " PUBLIC ");
	xmlBufferWriteQuotedString(buf->buffer, cur->ExternalID);
	if (cur->SystemID != NULL) {
	    xmlOutputBufferWriteString(buf, " ");
	    xmlBufferWriteQuotedString(buf->buffer, cur->SystemID);
	} 
    }  else if (cur->SystemID != NULL) {
	xmlOutputBufferWriteString(buf, " SYSTEM ");
	xmlBufferWriteQuotedString(buf->buffer, cur->SystemID);
    }
    xmlOutputBufferWriteString(buf, ">\n");
}

/**
 * htmlAttrDumpOutput:
 * @buf:  the HTML buffer output
 * @doc:  the document
 * @cur:  the attribute pointer
 * @encoding:  the encoding string
 *
 * Dump an HTML attribute
 */
static void
htmlAttrDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc, xmlAttrPtr cur,
	           const char *encoding ATTRIBUTE_UNUSED) {
    xmlChar *value;

    /*
     * TODO: The html output method should not escape a & character
     *       occurring in an attribute value immediately followed by
     *       a { character (see Section B.7.1 of the HTML 4.0 Recommendation).
     */

    if (cur == NULL) {
	return;
    }
    xmlOutputBufferWriteString(buf, " ");
    if ((cur->ns != NULL) && (cur->ns->prefix != NULL)) {
        xmlOutputBufferWriteString(buf, (const char *)cur->ns->prefix);
	xmlOutputBufferWriteString(buf, ":");
    }
    xmlOutputBufferWriteString(buf, (const char *)cur->name);
    if ((cur->children != NULL) && (!htmlIsBooleanAttr(cur->name))) {
	value = xmlNodeListGetString(doc, cur->children, 0);
	if (value) {
	    xmlOutputBufferWriteString(buf, "=");
	    if ((cur->ns == NULL) && (cur->parent != NULL) &&
		(cur->parent->ns == NULL) &&
		((!xmlStrcasecmp(cur->name, BAD_CAST "href")) ||
	         (!xmlStrcasecmp(cur->name, BAD_CAST "action")) ||
		 (!xmlStrcasecmp(cur->name, BAD_CAST "src")))) {
		xmlChar *escaped;
		xmlChar *tmp = value;

		while (IS_BLANK_CH(*tmp)) tmp++;

		escaped = xmlURIEscapeStr(tmp, BAD_CAST"@/:=?;#%&,+");
		if (escaped != NULL) {
		    xmlBufferWriteQuotedString(buf->buffer, escaped);
		    xmlFree(escaped);
		} else {
		    xmlBufferWriteQuotedString(buf->buffer, value);
		}
	    } else {
		xmlBufferWriteQuotedString(buf->buffer, value);
	    }
	    xmlFree(value);
	} else  {
	    xmlOutputBufferWriteString(buf, "=\"\"");
	}
    }
}

/**
 * htmlAttrListDumpOutput:
 * @buf:  the HTML buffer output
 * @doc:  the document
 * @cur:  the first attribute pointer
 * @encoding:  the encoding string
 *
 * Dump a list of HTML attributes
 */
static void
htmlAttrListDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc, xmlAttrPtr cur, const char *encoding) {
    if (cur == NULL) {
	return;
    }
    while (cur != NULL) {
        htmlAttrDumpOutput(buf, doc, cur, encoding);
	cur = cur->next;
    }
}



/**
 * htmlNodeListDumpOutput:
 * @buf:  the HTML buffer output
 * @doc:  the document
 * @cur:  the first node
 * @encoding:  the encoding string
 * @format:  should formatting spaces been added
 *
 * Dump an HTML node list, recursive behaviour,children are printed too.
 */
static void
htmlNodeListDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc,
	               xmlNodePtr cur, const char *encoding, int format) {
    if (cur == NULL) {
	return;
    }
    while (cur != NULL) {
        htmlNodeDumpFormatOutput(buf, doc, cur, encoding, format);
	cur = cur->next;
    }
}

/**
 * htmlNodeDumpFormatOutput:
 * @buf:  the HTML buffer output
 * @doc:  the document
 * @cur:  the current node
 * @encoding:  the encoding string
 * @format:  should formatting spaces been added
 *
 * Dump an HTML node, recursive behaviour,children are printed too.
 */
void
htmlNodeDumpFormatOutput(xmlOutputBufferPtr buf, xmlDocPtr doc,
	                 xmlNodePtr cur, const char *encoding, int format) {
    const htmlElemDesc * info;

    xmlInitParser();

    if (cur == NULL) {
	return;
    }
    /*
     * Special cases.
     */
    if (cur->type == XML_DTD_NODE)
	return;
    if (cur->type == XML_HTML_DOCUMENT_NODE) {
	htmlDocContentDumpOutput(buf, (xmlDocPtr) cur, encoding);
	return;
    }
    if (cur->type == HTML_TEXT_NODE) {
	if (cur->content != NULL) {
	    if (((cur->name == (const xmlChar *)xmlStringText) ||
		 (cur->name != (const xmlChar *)xmlStringTextNoenc)) &&
		((cur->parent == NULL) ||
		 ((xmlStrcasecmp(cur->parent->name, BAD_CAST "script")) &&
		  (xmlStrcasecmp(cur->parent->name, BAD_CAST "style"))))) {
		xmlChar *buffer;

		buffer = xmlEncodeEntitiesReentrant(doc, cur->content);
		if (buffer != NULL) {
		    xmlOutputBufferWriteString(buf, (const char *)buffer);
		    xmlFree(buffer);
		}
	    } else {
		xmlOutputBufferWriteString(buf, (const char *)cur->content);
	    }
	}
	return;
    }
    if (cur->type == HTML_COMMENT_NODE) {
	if (cur->content != NULL) {
	    xmlOutputBufferWriteString(buf, "<!--");
	    xmlOutputBufferWriteString(buf, (const char *)cur->content);
	    xmlOutputBufferWriteString(buf, "-->");
	}
	return;
    }
    if (cur->type == HTML_PI_NODE) {
	if (cur->name == NULL)
	    return;
	xmlOutputBufferWriteString(buf, "<?");
	xmlOutputBufferWriteString(buf, (const char *)cur->name);
	if (cur->content != NULL) {
	    xmlOutputBufferWriteString(buf, " ");
	    xmlOutputBufferWriteString(buf, (const char *)cur->content);
	}
	xmlOutputBufferWriteString(buf, ">");
	return;
    }
    if (cur->type == HTML_ENTITY_REF_NODE) {
        xmlOutputBufferWriteString(buf, "&");
	xmlOutputBufferWriteString(buf, (const char *)cur->name);
        xmlOutputBufferWriteString(buf, ";");
	return;
    }
    if (cur->type == HTML_PRESERVE_NODE) {
	if (cur->content != NULL) {
	    xmlOutputBufferWriteString(buf, (const char *)cur->content);
	}
	return;
    }

    /*
     * Get specific HTML info for that node.
     */
    if (cur->ns == NULL)
	info = htmlTagLookup(cur->name);
    else
	info = NULL;

    xmlOutputBufferWriteString(buf, "<");
    if ((cur->ns != NULL) && (cur->ns->prefix != NULL)) {
        xmlOutputBufferWriteString(buf, (const char *)cur->ns->prefix);
	xmlOutputBufferWriteString(buf, ":");
    }
    xmlOutputBufferWriteString(buf, (const char *)cur->name);
    if (cur->nsDef)
	xmlNsListDumpOutput(buf, cur->nsDef);
    if (cur->properties != NULL)
        htmlAttrListDumpOutput(buf, doc, cur->properties, encoding);

    if ((info != NULL) && (info->empty)) {
        xmlOutputBufferWriteString(buf, ">");
	if ((format) && (!info->isinline) && (cur->next != NULL)) {
	    if ((cur->next->type != HTML_TEXT_NODE) &&
		(cur->next->type != HTML_ENTITY_REF_NODE) &&
		(cur->parent != NULL) &&
		(cur->parent->name != NULL) &&
		(cur->parent->name[0] != 'p')) /* p, pre, param */
		xmlOutputBufferWriteString(buf, "\n");
	}
	return;
    }
    if (((cur->type == XML_ELEMENT_NODE) || (cur->content == NULL)) &&
	(cur->children == NULL)) {
        if ((info != NULL) && (info->saveEndTag != 0) &&
	    (xmlStrcmp(BAD_CAST info->name, BAD_CAST "html")) &&
	    (xmlStrcmp(BAD_CAST info->name, BAD_CAST "body"))) {
	    xmlOutputBufferWriteString(buf, ">");
	} else {
	    xmlOutputBufferWriteString(buf, "></");
            if ((cur->ns != NULL) && (cur->ns->prefix != NULL)) {
                xmlOutputBufferWriteString(buf, (const char *)cur->ns->prefix);
                xmlOutputBufferWriteString(buf, ":");
            }
	    xmlOutputBufferWriteString(buf, (const char *)cur->name);
	    xmlOutputBufferWriteString(buf, ">");
	}
	if ((format) && (cur->next != NULL) &&
            (info != NULL) && (!info->isinline)) {
	    if ((cur->next->type != HTML_TEXT_NODE) &&
		(cur->next->type != HTML_ENTITY_REF_NODE) &&
		(cur->parent != NULL) &&
		(cur->parent->name != NULL) &&
		(cur->parent->name[0] != 'p')) /* p, pre, param */
		xmlOutputBufferWriteString(buf, "\n");
	}
	return;
    }
    xmlOutputBufferWriteString(buf, ">");
    if ((cur->type != XML_ELEMENT_NODE) &&
	(cur->content != NULL)) {
	    /*
	     * Uses the OutputBuffer property to automatically convert
	     * invalids to charrefs
	     */

            xmlOutputBufferWriteString(buf, (const char *) cur->content);
    }
    if (cur->children != NULL) {
        if ((format) && (info != NULL) && (!info->isinline) &&
	    (cur->children->type != HTML_TEXT_NODE) &&
	    (cur->children->type != HTML_ENTITY_REF_NODE) &&
	    (cur->children != cur->last) &&
	    (cur->name != NULL) &&
	    (cur->name[0] != 'p')) /* p, pre, param */
	    xmlOutputBufferWriteString(buf, "\n");
	htmlNodeListDumpOutput(buf, doc, cur->children, encoding, format);
        if ((format) && (info != NULL) && (!info->isinline) &&
	    (cur->last->type != HTML_TEXT_NODE) &&
	    (cur->last->type != HTML_ENTITY_REF_NODE) &&
	    (cur->children != cur->last) &&
	    (cur->name != NULL) &&
	    (cur->name[0] != 'p')) /* p, pre, param */
	    xmlOutputBufferWriteString(buf, "\n");
    }
    xmlOutputBufferWriteString(buf, "</");
    if ((cur->ns != NULL) && (cur->ns->prefix != NULL)) {
        xmlOutputBufferWriteString(buf, (const char *)cur->ns->prefix);
	xmlOutputBufferWriteString(buf, ":");
    }
    xmlOutputBufferWriteString(buf, (const char *)cur->name);
    xmlOutputBufferWriteString(buf, ">");
    if ((format) && (info != NULL) && (!info->isinline) &&
	(cur->next != NULL)) {
        if ((cur->next->type != HTML_TEXT_NODE) &&
	    (cur->next->type != HTML_ENTITY_REF_NODE) &&
	    (cur->parent != NULL) &&
	    (cur->parent->name != NULL) &&
	    (cur->parent->name[0] != 'p')) /* p, pre, param */
	    xmlOutputBufferWriteString(buf, "\n");
    }
}

/**
 * htmlNodeDumpOutput:
 * @buf:  the HTML buffer output
 * @doc:  the document
 * @cur:  the current node
 * @encoding:  the encoding string
 *
 * Dump an HTML node, recursive behaviour,children are printed too,
 * and formatting returns/spaces are added.
 */
void
htmlNodeDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr doc,
	           xmlNodePtr cur, const char *encoding) {
    htmlNodeDumpFormatOutput(buf, doc, cur, encoding, 1);
}

/**
 * htmlDocContentDumpFormatOutput:
 * @buf:  the HTML buffer output
 * @cur:  the document
 * @encoding:  the encoding string
 * @format:  should formatting spaces been added
 *
 * Dump an HTML document.
 */
void
htmlDocContentDumpFormatOutput(xmlOutputBufferPtr buf, xmlDocPtr cur,
	                       const char *encoding, int format) {
    int type;

    xmlInitParser();

    /*
     * force to output the stuff as HTML, especially for entities
     */
    type = cur->type;
    cur->type = XML_HTML_DOCUMENT_NODE;
    if (cur->intSubset != NULL) {
        htmlDtdDumpOutput(buf, cur, NULL);
    }
    if (cur->children != NULL) {
        htmlNodeListDumpOutput(buf, cur, cur->children, encoding, format);
    }
    xmlOutputBufferWriteString(buf, "\n");
    cur->type = (xmlElementType) type;
}

/**
 * htmlDocContentDumpOutput:
 * @buf:  the HTML buffer output
 * @cur:  the document
 * @encoding:  the encoding string
 *
 * Dump an HTML document. Formating return/spaces are added.
 */
void
htmlDocContentDumpOutput(xmlOutputBufferPtr buf, xmlDocPtr cur,
	                 const char *encoding) {
    htmlDocContentDumpFormatOutput(buf, cur, encoding, 1);
}

/************************************************************************
 *									*
 *		Saving functions front-ends				*
 *									*
 ************************************************************************/

/**
 * htmlDocDump:
 * @f:  the FILE*
 * @cur:  the document
 *
 * Dump an HTML document to an open FILE.
 *
 * returns: the number of byte written or -1 in case of failure.
 */
int
htmlDocDump(FILE *f, xmlDocPtr cur) {
    xmlOutputBufferPtr buf;
    xmlCharEncodingHandlerPtr handler = NULL;
    const char *encoding;
    int ret;

    xmlInitParser();

    if (cur == NULL) {
	return(-1);
    }

    encoding = (const char *) htmlGetMetaEncoding(cur);

    if (encoding != NULL) {
	xmlCharEncoding enc;

	enc = xmlParseCharEncoding(encoding);
	if (enc != cur->charset) {
	    if (cur->charset != XML_CHAR_ENCODING_UTF8) {
		/*
		 * Not supported yet
		 */
		return(-1);
	    }

	    handler = xmlFindCharEncodingHandler(encoding);
	    if (handler == NULL)
		return(-1);
	}
    }

    /*
     * Fallback to HTML or ASCII when the encoding is unspecified
     */
    if (handler == NULL)
	handler = xmlFindCharEncodingHandler("HTML");
    if (handler == NULL)
	handler = xmlFindCharEncodingHandler("ascii");

    buf = xmlOutputBufferCreateFile(f, handler);
    if (buf == NULL) return(-1);
    htmlDocContentDumpOutput(buf, cur, NULL);

    ret = xmlOutputBufferClose(buf);
    return(ret);
}

/**
 * htmlSaveFile:
 * @filename:  the filename (or URL)
 * @cur:  the document
 *
 * Dump an HTML document to a file. If @filename is "-" the stdout file is
 * used.
 * returns: the number of byte written or -1 in case of failure.
 */
int
htmlSaveFile(const char *filename, xmlDocPtr cur) {
    xmlOutputBufferPtr buf;
    xmlCharEncodingHandlerPtr handler = NULL;
    const char *encoding;
    int ret;

    xmlInitParser();

    encoding = (const char *) htmlGetMetaEncoding(cur);

    if (encoding != NULL) {
	xmlCharEncoding enc;

	enc = xmlParseCharEncoding(encoding);
	if (enc != cur->charset) {
	    if (cur->charset != XML_CHAR_ENCODING_UTF8) {
		/*
		 * Not supported yet
		 */
		return(-1);
	    }

	    handler = xmlFindCharEncodingHandler(encoding);
	    if (handler == NULL)
		return(-1);
	}
    }

    /*
     * Fallback to HTML or ASCII when the encoding is unspecified
     */
    if (handler == NULL)
	handler = xmlFindCharEncodingHandler("HTML");
    if (handler == NULL)
	handler = xmlFindCharEncodingHandler("ascii");

    /* 
     * save the content to a temp buffer.
     */
    buf = xmlOutputBufferCreateFilename(filename, handler, cur->compression);
    if (buf == NULL) return(0);

    htmlDocContentDumpOutput(buf, cur, NULL);

    ret = xmlOutputBufferClose(buf);
    return(ret);
}

/**
 * htmlSaveFileFormat:
 * @filename:  the filename
 * @cur:  the document
 * @format:  should formatting spaces been added
 * @encoding: the document encoding
 *
 * Dump an HTML document to a file using a given encoding.
 * 
 * returns: the number of byte written or -1 in case of failure.
 */
int
htmlSaveFileFormat(const char *filename, xmlDocPtr cur,
	           const char *encoding, int format) {
    xmlOutputBufferPtr buf;
    xmlCharEncodingHandlerPtr handler = NULL;
    int ret;

    xmlInitParser();

    if (encoding != NULL) {
	xmlCharEncoding enc;

	enc = xmlParseCharEncoding(encoding);
	if (enc != cur->charset) {
	    if (cur->charset != XML_CHAR_ENCODING_UTF8) {
		/*
		 * Not supported yet
		 */
		return(-1);
	    }

	    handler = xmlFindCharEncodingHandler(encoding);
	    if (handler == NULL)
		return(-1);
            htmlSetMetaEncoding(cur, (const xmlChar *) encoding);
	}
    } else {
	htmlSetMetaEncoding(cur, (const xmlChar *) "UTF-8");
    }

    /*
     * Fallback to HTML or ASCII when the encoding is unspecified
     */
    if (handler == NULL)
	handler = xmlFindCharEncodingHandler("HTML");
    if (handler == NULL)
	handler = xmlFindCharEncodingHandler("ascii");

    /* 
     * save the content to a temp buffer.
     */
    buf = xmlOutputBufferCreateFilename(filename, handler, 0);
    if (buf == NULL) return(0);

    htmlDocContentDumpFormatOutput(buf, cur, encoding, format);

    ret = xmlOutputBufferClose(buf);
    return(ret);
}

/**
 * htmlSaveFileEnc:
 * @filename:  the filename
 * @cur:  the document
 * @encoding: the document encoding
 *
 * Dump an HTML document to a file using a given encoding
 * and formatting returns/spaces are added.
 * 
 * returns: the number of byte written or -1 in case of failure.
 */
int
htmlSaveFileEnc(const char *filename, xmlDocPtr cur, const char *encoding) {
    return(htmlSaveFileFormat(filename, cur, encoding, 1));
}

#endif /* LIBXML_OUTPUT_ENABLED */

#endif /* LIBXML_HTML_ENABLED */