mnoGoSearch HTML parser

Tag parser

The tag parser understands these tag attribute notations:

• <tagname attribute=value ... >

• <tagname attribute="value" ... >

• <tagname attribute='value' ... >

HTML entities

indexer understands the following special HTML characters:

• &lt; &gt; &amp; &nbsp; &quot;

• All HTML-4 character entities: &auml; &uuml; and other.

• Characters in their Unicode notation: &#234;

META tags

mnoGoSearch HTML parser currently understands the following special META tags. Note that HTTP-EQUIV is also understood instead of NAME in all entries.

• <META NAME="Content-Type" Content="text/html; charset=xxxx"> - This tag is used to detect the document character set when it is not specified in the Content-type HTTP response header.

• <META NAME="REFRESH" Content="5; URL=http://www.somewhere.com"> - The URL value is inserted into the database.

• <META NAME="Robots" Content="xxx"> with the values ALL, NONE, INDEX, NOINDEX, FOLLOW, NOFOLLOW.

Links

mnoGoSearch HTML parser collects links from these tags:

• <A HREF="xxx">

• <IMG SRC="xxx">

• <LINK HREF="xxx">

• <FRAME SRC="xxx">

• <IFRAME SRC="xxx">

• <AREA HREF="xxx">

• <BASE HREF="xxx">

  Note: mnoGoSearch ignores not well-formed BASE HREF values and uses the current URL to compose relative links in case when a badly formed base URL met.

The links having the rel="nofollow" attribute are ignored by mnoGoSearch.

Example:

<a href="http://site/" rel="nofollow">

Comments

• Character data inside the <!-- .... --> tag is recognized as an HTML comment.

• You can use special purpose comment tags <!--UdmComment--> ... <!--/UdmComment--> to hide the character data and the mark-up in between from indexer. It may find useful to put these tags around the things such as the site navigation menus, advertisement blocks, etc.

• The <NOINDEX> ... </NOINDEX> tags are also understood as synonyms for <!--UdmComment--> and <!--/UdmComment-->.