Have you ever had problems with source code formating and xml files with large blocks of text content? Here is the way to go to go:

Are you using iBATIS sqlmaps or any other technology heavily based on sql statements in XML files? With ibatis I quite often create complex sql queries in PL/SQL Developer, format them nicely there and paste them into my ibatis mapping file. The xml file looks quite nice after that but what happens if you use the source code formatting feature of eclipse (Ctrl+Shift+F)? Correct - the sql formatting is lost!

The solution to this problem is the usage of the CDATA element:

<select id="getTagsWithTagCategory" resultMap="Kk51TagWithKk50Result"> <![CDATA[ select kk51.KK51_TAG_SQ as kk51_KK51_TAG_SQ, kk51.TAG_CONTENT_1 as kk51_TAG_CONTENT_1, kk51.RANK as kk51_RANK, kk51.TAG_DATE as kk51_TAG_DATE, kk51.TAG_TITLE as kk51_TAG_TITLE, kk51.TAG_DATE_YEAR as kk51_TAG_DATE_YEAR, kk51.TAG_DATE_MONTH as kk51_TAG_DATE_MONTH, kk51.TAG_DATE_DAY as kk51_TAG_DATE_DAY from kk50_tag_category kk50, kk51_tag kk51 where kk50.kk01_list_sq = #kk01ListSq:BIGINT# and kk50.tag_category_id = #tagCategoryId:VARCHAR# and kk50.kk50_tag_category_sq = kk51.kk50_tag_category_sq ]]> </select> The ibatis mapping example stays the same even after the document is reformated. Withouth the CDATA it would look like this after formatting: <select id="getTagsWithTagCategory" resultMap="Kk51TagWithKk50Result"> select kk51.KK51_TAG_SQ as kk51_KK51_TAG_SQ, kk51.TAG_CONTENT_1 as kk51_TAG_CONTENT_1, kk51.RANK as kk51_RANK, kk51.TAG_DATE as kk51_TAG_DATE, kk51.TAG_TITLE as kk51_TAG_TITLE, kk51.TAG_DATE_YEAR as kk51_TAG_DATE_YEAR, kk51.TAG_DATE_MONTH as kk51_TAG_DATE_MONTH, kk51.TAG_DATE_DAY as kk51_TAG_DATE_DAY from kk50_tag_category kk50, kk51_tag kk51 where kk50.kk01_list_sq = #kk01ListSq:BIGINT# and kk50.tag_category_id = #tagCategoryId:VARCHAR# and kk50.kk50_tag_category_sq = kk51.kk50_tag_category_sq </select>