[teiid-issues] [JBoss JIRA] (TEIID-1819) Reading multi entity data from a single data file

[ https://issues.jboss.org/browse/TEIID-1819?page=com.atlassian.jira.plugin... ] Steven Hawkins commented on TEIID-1819: --------------------------------------- Our internal format for compiling metadata from Teiid Designer is exactly in the format you're describing. A single character record type followed by a delimiter and the delimited record with the possibility for continuation records. So we're very familiar with this type of file. We'll add a SELECTOR argument to TEXTTABLE. The base implementation would be a string that matches the line prefix. Given that the value would be a literal, a compiled regex pattern would not have that much performance degradation, e.g. '^x' vs. a prefix comparison of 'x'. However if that complexity is not needed, we can stick with just a prefix pattern or go with something in the middle, such as a like pattern. Is you're preference to just match based upon a prefix string? Are you also requesting the ability to support continuation records? TEXTTABLE has support for quoted values and the use of an escape character. The data/time conversion, if their not the same as what is allowed by the built-in type conversion, would be specified in a select expression using a parseTime, parseDate, parseTimestamp function. And just as you say there's no need to overdo the TEXTTABLE function, as dealing with field length or referring to other rows could also be done after the primary extraction. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira