DSS: Back Log Project Database Import

Modification Request #8 (C1 2008)

1. Introduction

There are a number of incomplete projects in Carl Bignell's database that are planned to be used as back log projects for DSS. This iteration will focus on extracting these projects from Carl's database and importing them into the DSS database. See DBaseDocs for a description of Carl's table structures.

2. Background

This is the beginning of the third task in the Testing table of the SoftwareAgileWorkBreakdownStructure (release 0.4) [refer to "summer 08 test database"]. See DSS project charter for more information.

2.1 Nomenclature

Three sets of terms exist for describing scheduling are required for this MR. The first set are the one's currently used by Carl and are reflected in his database and indicated by italics. The second set is that adopted and defined by DSS and are defined by links. The third set are the names of the tables in nell's database schema and when needed are indicated by bold print. Normally, this set is for programmer's use only, but since we will be describing possible implementations, we may need to occasionally make it explicit.

3. Requirements

The goal of this exercise to transfer only the essential information needed to successfully schedule these pending projects under the DSS scheduling mechanism targeted for the 08B trimester. It is in no way intended to replace or usurp the functionality of the current database.

Which projects are back log projects which need to be brought forward into trimester 08B must be selected by Carl.

3.1 Tasks

Complete the following tasks by February 29, 2008:

Export all incomplete projects from Carl's database to ascii file with a unique field delimiter.
Import ascii file dump into DSS database populating existing fields only and fully integrating the information.
Import frequency bands
Import canonical sky positions

"Round Tuits"

Evolve DSS database schema to handle additional fields.*
Modify CRUD to handle additional fields.*
Generate "reverse path" of resultant project progress for the 08B semester

* These items have been moved to their own MR.

4. Out of Scope

This is MR does not cover interfacing DSS with PST project data.

5. Design

See the Dynamic Scheduling Documentation.

5.1 Translations

An automatic translation from Carl's database to DSS's database is probably not poissible, but there do exist some rules of thumb that should allow at least some translation. Note that various types of projects partition their sessions differently between the two databases. For example, a monitoring project is fully represented by sessions with numerous descriptive fields indicating when and how they should/are scheduled; these will have to be repartitioned into Fixed and Windowed Allocations in the DSS.

Generally, using Carl's separation and obstype fields a session can be classified thus:

if separation > 0
    if obstype is monitoring
        Windowed
    else
        mya be large monitoring project crossing semesters
else if separation = 0
    if obstype is monitoring
        Open
    else
        if has date_request
            Fixed
        else
            Open

Projects

Name and time can be translated directly between the two. Used time in Carl's system will be represented by a bogus completed session in DSS to align the time remaining correctly.

Sessions

A DSS session is described by a database table which includes:

name (string)
total_hours (float) hours
min_duration (integer) hours (later to be quarter-hours)
max_duration (integer) hours (later to be quarter-hours)
frequency (float) GHz
ra (float) hours
dec (float) degrees
type (string)
original_id (integer)

Where type may be Fixed, Windowed, or Open (neither Fixed nor Windowed).

Carl's session is defined by:

perhaps a name
length
repeat count
lst range
perhaps a source list
separation
receiver band

All cases will require significant manual editing and additional information. We expect the following fields may be incomplete and require manual editing.

min_duration
max_duration
frequency
ra
dec

For Open sessions:

name = pcode
total_hours = length x repeat count
min_duration = max_duration = length
frequency = middle of the receiver band repersented by the short code
ra, dec = if sources then center position of source list else ra is center of lst range and dec of unknown
type = Open

Note that large ranges of sources or sky partition will need to be broken into multiple sessions in DSS.

For Fixed sessions:

name = pcode
total_hours = length
min_duration = max_duration = length
frequency = middle of receiver band repersented by the short code
ra, dec = if sources then center position of source list else ra is center of lst range and dec of unknown
start_time and duration = where is time kept?
type = Fixed

For Windowed sessions:

Each of the repeat count telescope slots is mapped to a DSS session where the time between is set by separation.

name = pcode
total_hours = length
min_duration = max_duration = length
frequency = middle of receiver band repersented by the short code
ra, dec = if sources then center position of source list else ra is center of lst range and dec of unknown
a list of start_times and durations = generated over a range that is 10% of separation
type = Windowed

Written	- MikeMcCarty - 06 Feb 2008
Checked	- MarkClark - 06 Feb 2008
Approved by Sponsor	- KarenONeil - 06 Mar 2008
Approved by CCC	- RonMaddalena - 11 Mar 2008
Accepted/Delivered by Sponsor	- KarenONeil - 18 Mar 2008

Symbols:

Use %X% if MR is not complete (will display )
Use %Y% if MR is complete (will display )

CCC Discussion Area

Topic ModificationRequest8C108 . { Edit | Attach | Ref-By | Printable | Diffs | r1.18 | > | r1.17 | > | r1.16 | More }

Revision r1.18 - 18 Mar 2008 - 10:30 GMT - KarenONeil Parents: PlanOfRecordC12008	Content copyright © 1999-2007 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.