Refactor CSV/SQL import structure slightly, and rewrite SQL import:

 - SQL import now reads and processes files in full streaming mode, running queries as they are encountered
 - Memory usage during import is significantly reduced, and should stay within a few megabytes; the significant memory use remaining is for query logging
 - The progress bar more accurately represents progress and is shown at once (this addresses Issue #320)

1 files changed, 2 insertions, 2 deletions

diff --git a/Source/SPSQLParser.h b/Source/SPSQLParser.h
index 022e17bb..73dfe493 100644
--- a/Source/SPSQLParser.h
+++ b/Source/SPSQLParser.h
@@ -54,14 +54,14 @@
  *
  * It is anticipated that characterAtIndex: is currently the parsing weak point, and that in future
  * this class could be further optimised by working with the underlying object/characters directly.
- * This class could also be improved by maintaining an internal parsedTo number to allow streaming
- * processing to occur without repetition.
  */
 
 @interface SPSQLParser : NSMutableString
 {
 	id string;
 	unichar *stringCharCache;
+	unichar parsedToChar;
+	long parsedToPosition;
 	long charCacheStart;
 	long charCacheEnd;
 	NSString *delimiter;