ProteoLens software version history

1.1.1
- Database support:
  Support the network of 10000 nodes retriving from Oracle DB.
- Network drawing:
  Re-calculating the network layout,and seting the nodes always upon the edges.
- Memory allocation problem:
  Supporting large network *.jpg/*.png saving.
- Minor bug fixes

1.1
- Layout support
  nodes can be now drawn above crossing edges
- general code cleanup
- scalability bug fixes

1.08

- Database support:
  Improvements to formation of SQL queries preliminary to a 1.1 release.

1.06

- Database support:
  Access time significantly improved in some situations. 1) in the data browser (when 'view' 
  operation is called upon a DB table in the filesystem browser), the actual submitted query
  uses LIMIT clause to show small sample of the actual table data in the Table pane (use 'Retrive all'
  menu option to override this behavior and retrieve complete result of query execution into 
  the table pane); 2) when annotations are created/applied, only the relevant data are 
  retrieved from the database table using the IN operator (i.e. selection is now performed on 
  the database side by default).
- Minor bug fixes

1.05

- Database support:
  Database support extended to allow connections to PostgreSQL databases. Database
  connection dialog now offers selection from Oracle (thin or oci8 drivers) and
  PostgreSQL databases to connect to; only fully qualified DB connection parameters
  (host, port, database - same parameters as required for Oracle thin connection) 
  are accepted for PostgreSQL. 

- Database support:
  Bug in Oracle DB support fixed: non-standard (not fully capitalized) table
  names are properly quoted in automatically generated queries (Oracle JDBC 
  otherwise refuses to recognize MYSCHEMA.MyTable table, must be MYSCHEMA."MyTable")  

- Loading/Saving networks: 
  File type detection by file extension is abandoned. Load network and Save 
  network menus now have sub-menu options enforcing saving to/loading from 
  a particular format ('Save as -> GML file...', 'Save as -> List of node IDs...' 
  etc). Thus the target file does not have to have the suggested extension anymore.

- Loading networks:
  Simple GML format verifier implemented. A very high-level and incomplete format 
  check is performed for this release; the main purpose is to guard against an apparent
  bug in the underlying 3rd party GML handler (y-Files), which freezes the system
  if the file it attempts to read is seriously malformed. An error message pop-up is 
  now displayed if the file claimed to be GML does not have top-level 'graph' token or
  if brackets are not balanced. More subtle problems with file format are successfully caught
  and reported by the generic IO handler.

- Filter loading: 
  Multiple file selection is allowed when loading filter list from file(s);
  node IDs loaded from multiple selected files are merged into the single filter list.

- XML session configuration reader: 
  if a file specified as the source for a data association does not exist at the time 
  of reading the configuration, the XML loader will pause and offer the user an option
  of selecting the missing source file manually. 

- XML session configuration reader/writer:
  If a data association is defined on top of a local file, *and* if this file is located
  in the local filesystem hierarchy below dedicated 'data' folder, then relative URI 
  (with respect to this data folder) is stored instead of the absolute one. Correspondingly, 
  when such data association is loaded from the XML configuration, the source data file URI is
  resolved against the dedicated 'data' directory configured in the application. 
  This feature simplifies data transfer: as long as the data files are placed into the 
  same subfolder hierarchy under the dedicated 'data' folder, they can
  be distributed along with ready-to-use XML configuration files defining data associations.
  NOTE 1: data folder is not configurable in version 1.05 and is set to be 'data' in the
  proteolens base installation directory.
  NOTE 2: if the data folder is not defined/does not exist in the application
  that reads XML configuration file but source file is specified relative to such
  non-existent data folder, then the reader attempts to resolve the relative file locations
  against the "user.dir" directory (working directory, or the directory where application
  was started from). If the sought file does not exist in that directory either, the user will
  be asked to locate file manually (see previous item in the history list).