Commit b5e9e46d authored by Hugo Buddelmeijer's avatar Hugo Buddelmeijer
Browse files

+ Cleaned up the todo list somewhat.

parent 885b0bea
......@@ -4,31 +4,12 @@ TODO list for SourceCollection classes.
comments in the individual files.
+ The comments here only refer to the current implementation of the
SourceCollection classes. Possible future extensions are not included.
+ There is no explicit ranking in importance.
+ There is no explicit ranking of the items here.
+ The main points that need to be changed before the implementation can
be moved from 'experimental' to 'main' are
- DATAMODEL
*** [DATASTORING] Storing/Caching of Data
+ Using an actual SourceList for the sourcelist_data and sourcelist_sources
is not optimal. It might be better to design separate data structures for
this in the future.
+ The sourcelist_sources, sourcelist_data and all_data_stored
cannot yet be changed for persistent objects.
+ The USER_14_i columns etc. should be used if necessary.
+ Perhaps initialize the SCID for SCs when a sourcelist_data is made so it
can be put in the name of the sourcelist_data.
+ Decide whether ingest_data() should call create_empty_sourcelist_data().
+ Ingesting data should be possible without having to commit
the sourcelist_data first.
+ The 'ORA-01458' error should be fixed, see SC.ingest_data().
*** [DATAMODEL] Datamodel
+ The RelabelSources could use an 'other_sourcecollection' persistent property.
+ The query of the FilterSources has a maximum length of 297, perhaps this
should be increased.
+ The description of the AttributeCalculatorParameter is a list by accident.
+ There are a couple of SourceCollections that Jeffrey uses that are given a
special threatment. These should be fixed, see SC.is_made().
......@@ -39,18 +20,21 @@ TODO list for SourceCollection classes.
+ Idem for SC.store_sources().
*** [DATAFORMAKE] The SC.make() method should be improved.
+ SourceCollection.make() should have an optimize argument as well.
- With optimize=False it should assume all data it requires has been made and
loaded into Python if necessary. With optimize=True, an SCT is created that
is optimized to minimize the creation of data. (Currently this has to be
done explicitely.) This SCT will then ensure the data is available.
- With optimize=True, making data should be done recursively.
+ How does the SCT know which data it has to retrieve? Other than the parents.
+ The AttributeCalculator will fetch the data for the parent at the moment.
*** [DETERMINESOURCES] The process to determine the sources in an SC.
+ It should also be possible by creating an SA with just [SLID,SID]
as attribute and optimizing this.
*** [DATASTORING] Storing/Caching of Data
+ Using an actual SourceList for the sourcelist_data and sourcelist_sources
is not optimal. It might be better to design separate data structures for
this in the future.
+ The sourcelist_sources, sourcelist_data and all_data_stored
cannot yet be changed for persistent objects.
+ The USER_14_i columns etc. should be used if necessary.
+ Perhaps initialize the SCID for SCs when a sourcelist_data is made so it
can be put in the name of the sourcelist_data.
+ Ingesting data should be possible without having to commit
the sourcelist_data first.
+ The 'ORA-01458' error should be fixed, see SC.ingest_data().
*** [SQLPYTHON] Python v.s. SQL
The interplay between Python and SQL should be improved.
......@@ -87,17 +71,10 @@ The interplay between Python and SQL should be improved.
- What should SourceCollectionTree.replace_in_tree() return?
*** [RELATIONS] Relations Handling
+ Consistent use of 'quick=True' in get_source_relations():
- Add parents if they are not there by passing quick=True
- Filter the relations.
+ Move the cross section and union adding code from
the CA and CS into the SetRelations classes.
+ Do something sensible when the relations might become to
+ Do something sensible when the relations might become too
large, e.g. in the CS or when moving an SS/FS up through an SS/FS.
+ How relations are handled in the modify_* functions is not consistent.
- Should they be created if not given?
- only if they are required for the modification?
- as a temporary copy?
*** [SCTTRACKING] SourceCollection Tracking in SCT
The tracking of SourceCollections within an SCT should be improved:
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment