Timelines Infrastructure

The key elements of the timelines project are:

• parse_cmd_load_gen.pl: parse command load processing summaries from mission planning directories
• update_load_seg_db.py: update load segments and timelines tables

Helper elements include:

• fix_load_segments.py: module to implement “manual” database fixes for load_segments (that are not yet included in iFOT). If a load needs to be modified after it has rolled off of the current arc/iFOT load segment rdb, modify fix_load_segments.py and then call update_load_seg_db.py with a constructed rdb that includes the load segment that needs modification.
• fix_tl_processing.py: script to implement “manual” database fixes for the command load summary file parsing tables (tl_built_loads, tl_processing)
• timelines_test.py: package containing regression test elements. suitable for nose tests and the following scripts
• timelines_make_testdb.py: make a testing db for ... testing
• timelimes_check_testdb.py: compare testing db to telemetry

Regression Testing

The regression tests are packaged as nose tests in timelines_test. The end-to-end tests in test_loads() generate a test databases from load_segments and backstops and compare the created states with tables of states stored in the testing directory.

Tests may be run with:

nosetests timelines_test.py

(note that the script runs using sqlite test database by default. Set DBI=’sybase’ at the top of a local copy of timelines_test.py to test using aca_tstdb sybase database)

If there is a diff in the output:

.Checking t/july_fixed.rdb Made Test States in /tmp/tmpyOgXJT Diff in /tmp/tmpyOgXJT

look at the html diff (e.g. /tmp/tmpyOgXJT/diff.html) or manually diff the states with the expected states, for example:

meld /tmp/tmpyOgXJT/test_states.dat t/july_fixed.dat

If the diff is expected, due to a change in nonload commands since the creation of the fiducial data, update the fiducial data:

cp /tmp/tmpyOgXJT/test_states.dat t/july_fixed.dat

and commit the change in the timelines project.

The utility scripts “timelines_make_testdb.py” and “timelines_check_testdb.py” are provided to allow more manual testing:

• timelines_make_testdb

  For example:

  ./timelines_make_testdb.py –load_rdb t/replan.rdb –outdir replan

• timelines_check_testdb

  For example:

  ./timelines_check_testdb.py –load_rdb t/replan.rdb –outdir replan

Each script takes a load segment list in rdb format (as retrieved by arc or construct to look like it was retrieved that way).

timelines_make_testdb.py builds a testing area, creating a softlink tree to /data/mpcrit1/mplogs, parsing the files in that softlink tree, and constructing and populating testing database tables from that tree and the input load segment list.

timelines_check_testdb.py makes validation plots vs telemetry.