Skip to content

Milestones

List view

  • Writing Python and GAP scripts for data extraction and cleanup from sources

    The data has two sources: (1) the [HTML tables](http://www2.informatik.uni-halle.de/da/hedtke/tpp/) which contain data about maximal-by-size subgroup TPP triples for *almost* groups of order up to 1000, with some important [exceptions](http://www2.informatik.uni-halle.de/da/hedtke/tpp/README.txt) and (2) the small groups of order up to 1000 contained in the [GAP small groups library (SGL)](http://www.icm.tu-bs.de/ag_algebra/software/small/). The Python scripts will extract data from the HTML tables, but the extracted data needs to be partly cleaned up by means of GAP scripts before writing to CSV files - this is because currently the HTML tables contain data about the components of a subgroup TPP triple that are actually GAP statements and are not in a serialised format suitable for storage in database. The combination of these scripts will produce CSV files that store the data about groups and subgroup TPP triples consistent with the database schema.

    Overdue by 11 year(s)
    •
    Due by February 12, 2015
    0% complete0 open0 closed

  • Developing and testing a serialising/deserialising method for subgroup TPP triples in GAP

    A subgroup TPP triple of a (finite) group in GAP is an ordered triple (a list of length 3) of subgroups which satisfy the TPP. The components of this triple are GAP objects as is the list itself, and the triple could be represented in GAP as a record. To store this subgroup TPP triple in a relational database table it is necessary to serialise its GAP representation to a format suitable for storage as a record in a relational database table. Currently there are several options that GAP provides via specific packages, including [OpenMath](http://alexk.host.cs.st-andrews.ac.uk/openmath/), [IO](https://github.com/gap-system/io), [SCSCP](http://alexk.host.cs.st-andrews.ac.uk/scscp/) and an experimental package [json](https://github.com/gap-system/json). Of these the SCSCP seems the best - it contains the [IO_PickleToString](http://alexk.host.cs.st-andrews.ac.uk/scscp/doc/chap9.html#X84F055ED860120D5) and [IO_UnpickleFromString](http://alexk.host.cs.st-andrews.ac.uk/scscp/doc/chap9.html#X813EACD27C218E19) methods which appear be highly compact. Ideally the serialisation method should transform GAP subgroup TPP triples to positive integers that uniquely identify them with respect to the groups with which they are associated.

    Overdue by 11 year(s)
    •
    Due by February 5, 2015
    0% complete0 open0 closed

  • Designing the app database schema

    Describe the schema for the core tables, of which there will be at least three initially - `groups`, `subgroup_tpp_triples` and `mbs_tpp_stats` - and the table relationships, views, indexes and stored procedures.

    Overdue by 11 year(s)
    •
    Due by January 21, 2015
    0% complete0 open0 closed

  • Developing a high-level architecture for the application

    Overdue by 11 year(s)
    •
    Due by January 13, 2015
    0% complete0 open0 closed