Dotter: replaced messalloc/messfree by g_malloc/g_free. Removed some unused items from dotter_.h and removed use of dotter_.h from belvu because it doesn't seem to contain anything that is being used.

Added a prefix to the 'organism' data field so that a narrow column width can be used that displays just the prefix

Fixed a couple of bugs with squash matches: go-to-next was highlighting the wrong frame for protein matches; MSPs for both strands were being added to each tree.

Removed the need for the -F command line option when a single input file is used. If blixem is called with one input file, it is assumed to contain both alignment data and the reference sequence. If it is called with two files, the first is assumed to contain the reference sequence and the second the alignment data. The -F option has been left in for now for backwards compatibility.

Fixed a couple of bugs with the GFF parser: sequence name was not being read from the header; fixed a problem with constructing introns

Added two status bars: one for messages and one for feeding back info about the currently moused-over item. Fixed a bug with parsing peptide sequence data from seqbl files. Fixed some bugs with negative offset: some termini were being shown as splice sites; go-to-next match was sometimes messing up.

Added support for grouping exons by parent transcript ID (i.e. as passed in GFF format). Added full support for CDS/UTR types. Calculate the reading frame, using the given phase, if applicable. Changed score and ID to be decimal values rather than integers and allow grid scale to show fractions of an integer per cell. Reduced exon size and padding when the exon view is bumped. Added GT-AG to list of canonical splice sites. Removed incorrect splice-site highlighting which was being shown at the termini.

Removed a spurious popup message that happens when pfetching full entries on startup fails (because pfetching just the fasta sequence may still work fine)

Fixed a bug with the clipped-match marker where it was drawn the wrong side of the base when the display is reversed

Added pfetch of additional EMBL info for pfetch-http mode. Added marker to indicate if a match has a missing section where it has been clipped to the reference sequence range. Created a replacement for the old acedb message system. Made relevant dialogs persistent and added functionality to make them refresh their data when necessary.

Removed more of the acedb package from blixem. Fixed a bug in the new pfetch code where it was getting data in the wrong order when data was loaded from the settings dialog rather than at startup.

Added a warning if pfetch net id and port are not provided in the command line, env vars or config file.

Fixed an issue where pfetch port and net_id were not being set if user changes to the pfetch mode from a different default mode.

Added functionality to parse additional info from the EMBL files such as organism and tissue-type. Added a command-line option to fetch this on startup and an option in the settings dialog to populate the data later, since it may be slower fetching the full EMBL files. (Although for nucleotide sequences it does not seem to be slower so in that case it always fetches the extra info.). Also fixed a bug where a crash was possible if the input s coords were out of the range of the given s sequence.

Fixed a bug where middle-clicking in the tree headers was sometimes not working. Improved middle-clicking so that it now works on empty trees as well.

Added option to select fetch mode. Fixed an issue where the background color changed slightly if you hit ok on the settings dialog.

RT58135: When pfetching a protein variant, pfetch with the -F option fails, so re-try without the -F option so that at least we get the fasta sequence. Also fixed a bug where the show-unaligned-sequence option was not working when Blixem was run in pfetch-http mode

Fixed a bug where pfetch was sometimes failing to display the results because it was trying to apply pango markup