OraCIF EXPORTER  takes either a list of PDB accession codes generated from the Selector or a single PDB identifier supplied as a parameter and uses the stored procedures created by the Generator to extract the relevant data from the source database(s). 

INPUT.  

The Exporter needs to know either the PDB accession code to be handled  or a Selector job identifier. Some data from the OraCIF  parameters table determine  how the Exporter behaves (for instance whether the results are stored on the filesystem or as CLOB/XmlType columns  within database tables).

PROCESS. 

Exporting a  large list can be highly parallelised. Therefore the Exporter splits a list of accession codes in to small sublists and then farms these sublists out to separate processes. 

Each process is first split down into individual accession codes and then into individual mmCIF categories within a given PDB identifier.   The Exporter controls, coordinates and synchronizes   this   cascade and produces the correct outputs.

.

Depending on the input parameters  the results of the database stored procedures  may be  stored in special OraCIF tables. In this cases  the Exporter  will verify if the deposition has been updated within source database(s) and  can trigger the  processes necessary to update the  old records.

OUTPUT

The result of the Exporter depends on  the mode of use selected in the OraCIF parameter table. Either files files can be written to the file system directly or else the equivalent data can be stored in the database. 

In the current version  of OraCIF the Transformer step is integrated in to the Exporter so the results are fully formatted mmCIF or XML. However, in the near future, the extraction and the formatting steps will be decoupled to allow greater flexibility in the formatting.

.

Currently the mmCIF files returned by Exporter from the MSD Archive database (in ftp://ftp.ebi.ac.uk/pub/databases/msd/mmcif/) are used for data exchange between the EBI and the RCSB. The XML files returned by the Exporter from MSD Archive and Search databases are useful for reformatting by XSLT technologies and have been used to generate the Atlas pages  for MSDlite and MSDpro (see for example the  Atlas page and the  corresponding XML file for the PDB code "1get" ).