Data in Glast are stored in several different ways and in different types of files. Depending on your analysis, you will have to access the data in different ways. This is a brief primer to guide you in your adventure. More details can be found in the Data Analysis Primer (html) (Word) and the SAS Offline Workbook.
From the online to the offline world the data chain is the following: LDF - FITS - Digi - Recon - Merit and SVAC ntuples
Looking at the Merit and SVAC ntuples is the easiest. It can either be done with ROOT or Hippodraw. And it can be done either interactively, with the Root TBrowser or the Hippodraw Inspector, or with Root macros/Hippodraw Python scripts. More details can be found in the data analysis examples
The Merit and SVAC root tuples only contain a summary of the available information in the digi and recon root files. Depending on your analysis you may have to access these files directly. The data format in both the digi and recon files is not as straightforward as in the root tuples and you can not access them in the same way. Two possibilities exist: