[Kst] [Bug 111571] dirfile data source poorly handles error conditions.

Matthew Truch matt at truch.net
Wed Oct 12 04:17:25 CEST 2005

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

------- Additional Comments From matt truch net  2005-10-12 04:17 -------
On Tue, Oct 11, 2005 at 06:33:45PM -0400, Barth Netterfield wrote:
> All of these, (except for the famous memory allocation error... we've been 
> down that road several times...), were errors which can only happen due to an 
> internal bug in GetData.   Neither faulty data, nor faulty function input/use 
> can ever trigger any of them.  I had the 'exit(0)' and the error message 
> there for identifying bugs when I wrote the code several years ago.  None of 
> them have ever been triggered.  So the bug is an 'invalid' as far as I am 
> concerned.

When I wrote the bug report, I wasn't thinking of bugs internal to
getdata; I was thinking of when people write their own format files
which contain typos/invalid entries.  We've been seeing alot of these
since we started data analysis on BLAST and people are creating their
own raw data and adding (by hand) entries in the format file.  If these
entries are mistyped or invalid or missing (both the binary data missing
or the format entry missing) kst silently fails.

More information about the Kst mailing list