Hi,
Just wanted to check before I got to lengths programming (as testing is not 100% & annomolies can show up at any time)
Is using the FileSystemObject & .Readline 100% sure to read line by line in the order of the file. In my case a .csv
TransferText is not guaranteed as far as I can tell.
For ages always had problems trying to import CSV's or TXT files in the order they originally created. Often no field in true sort order, no autonumber or unique field.
I know the whole point of a database if to make sure you have these fields but not always easy with files from other sources.
Transfertext often works not not guaranteed. I read linking is good but still not so sure of this.
Best example is Bank Statements exported to be used in accounting or others.
Always have a date of course but if the same day, e.g. 30 on the same day they can get mixed up & trying to visually match from original source to imported is a pain.
I have other situations with text files also so would be great if I can rely on this
I can read line by line & add my own autonumber as it goes, also not including lines I don't.
Also seems pretty fast this way but not tested much yet.
Thanks I/A
Just wanted to check before I got to lengths programming (as testing is not 100% & annomolies can show up at any time)
Is using the FileSystemObject & .Readline 100% sure to read line by line in the order of the file. In my case a .csv
TransferText is not guaranteed as far as I can tell.
For ages always had problems trying to import CSV's or TXT files in the order they originally created. Often no field in true sort order, no autonumber or unique field.
I know the whole point of a database if to make sure you have these fields but not always easy with files from other sources.
Transfertext often works not not guaranteed. I read linking is good but still not so sure of this.
Best example is Bank Statements exported to be used in accounting or others.
Always have a date of course but if the same day, e.g. 30 on the same day they can get mixed up & trying to visually match from original source to imported is a pain.
I have other situations with text files also so would be great if I can rely on this
I can read line by line & add my own autonumber as it goes, also not including lines I don't.
Also seems pretty fast this way but not tested much yet.
Thanks I/A