🚀 go-pugleaf

All,

A very short newsletter this month as most of us have been away...but are back
now and ready for action....

cheers Ed
-- 
 ------------------------------------------------------------------------
| Ed Griffiths, Acedb development, Informatics Group,                    |
|        Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus,  |
|               Hinxton, Cambridge CB10 1SA, UK                          |
|                                                                        |
| email: edgrif@sanger.ac.uk  Tel: +44-1223-494780  Fax: +44-1223-494919 |
 ------------------------------------------------------------------------


----------------------------------------------------------------------------
----------------------------------------------------------------------------

ACEDB User Group Newsletter - August 2003
#########################################

If you want to have this newsletter mailed to you _or_ you want to make
comments/suggestions about the format/content then send an email to
acedb@sanger.ac.uk.

-----------------------------------------------

Not much to report this month as most of the acedb developers have been away
on their holidays and are of course now refreshed to hack away at acedb with
renewed vigour.

-----------------------------------------------

Articles
********

Acedb and MS Windows
++++++++++++++++++++

Porting to other platforms is not easy and although the task of porting
acedb to MS Windows is considerably eased by our use of the Cygwin libraries
which provide a unix system call interface on a windows machine, its still
not all plain sailing.

Recent problems include:

   * Modifying code to for temporary directories to match the new Cygwin
     style including adding support for the $TEMP environment variable so
     user can override the default location of the temp directory.
   * If the stack size for the acedb process is very low, acedb attempts to
     raise it but this doesn't work on Windows so the code must handle any
     error and continue.
   * Various manglings of include files and macros: "moved definition of
     COBJMACROS and removed #include to get IMalloc_Free to work properly on
     Win32" and "#include for new cygwin version and manually #defined
     IMalloc_Free as it's not happening properly" and so on, and so on.

For all this the system generally works very well and means that acedb truly
runs on Unix, Mac and MS Windows machines.

SMap and FMap
+++++++++++++

Originally the objects that made the hierachy from which FMap constructed
its view of a sequence had to all belong to the Sequence class. The
introduction of the SMap code into acedb now allows for hierachies to be
made up of any objects as long as they contain the SMap tags and data.

While this system is very much more flexible and brings many benefits it has
led to some problems in the FMap code. A recent example from the C. elegans
database was a bug which meant that if a user double clicked on a
"Transcript" class object in the FMap it was not Tree displayed as would
normally be expected. This was caused by an assumption in the code that the
object should have been of the Sequence class.

While this bug has been fixed it is likely that there will be other areas of
FMap that will have to be fixed in the future since much of the code
contains often subtle assumptions that objects will be of the Sequence
class. This is work in progress so if you run into problems with FMap that
you suspect are caused by SMap related changes then please email the acedb
developers.

-----------------------------------------------

Bugs Fixed
**********

Tree display
++++++++++++

Fixed bug in tree display, lookDraw->stackText must be large enough under
all circumstances since it gets passed to graphTextEntry with a buffer
length of 2500.

Blixem bug when showing gapped homologies
+++++++++++++++++++++++++++++++++++++++++

(Simon Kelley srk@sanger.ac.uk wins the "Developers award for Persistence
under Fire" for sorting this one out.)

Previous versions of acedb did not pass gap information to blixem, the
latest monthly now does this, but.... The process by which acedb passes
sequence/clone information to blixem is not straight forward as account must
be taken of orientation (e.g. was the sequence reverse complemented in fmap
?), whether blixem is being run as part of xace or as a separate program,
what format the gap information should be in and so on and so on. Sadly this
has led to a few bugs. Simon has manfully tracked these down and fixed them
one by one.

You should now find that gapped alignments from fmap are now correctly
displayed in blixem.

-----------------------------------------------

Developers Corner
*****************

Aquila documentation
++++++++++++++++++++

Rob has been adding to and improving documentation for the aquila system,
see documentation on the website. He has improved the sections on how to add
and run tests and how to interpret the results.

-----------------------------------------------

August monthly build now available.
***********************************

You can pick up the monthly builds from:

Sanger users
     ~acedb/RELEASE.DEVELOPMENT
External users
     http://www.acedb.org/Software/Downloads/monthly.shtml

-----------------------------------------------

Next User Group Meeting - D319, 3.00pm, Thursday 11th September 2003
********************************************************************

----------------------------------------------------------------------------
----------------------------------------------------------------------------
Ed Griffiths <edgrif@sanger.ac.uk>
Last modified: Wed Sep 3 11:30:10 BST 2003