Seven parts,
one database
Data Wrangler is one application over a DataView database. Each part does a narrow job and hands off to the next, so you can adopt one without adopting all of them.
Below is what each part is for, taken from the platform overview. Five have a full component document as well — the same documents the software was built against, covering what each component does, the decisions behind it, and what it deliberately refuses to do.
The database
Everything else writes into this.
Getting data in
Three routes, chosen by what the data looks like rather than where it came from.
File Catalog
An extension of the DataView schema — tables for the inventory of files on a shared drive, and for cataloguing and extracting the tables of data inside them. Structured: LAS, DLIS, LIS, SEG-Y, P190, Esri shapefile. Unstructured: Word, Excel and PDF. What it extracts can be loaded into DataView, or into another database entirely.
Read the documentation →Data Assistant
Bulk loading for flat files. Data goes to staging, the table and column mapping is presented for review, values are normalised and validated, and any violated foreign key is put in front of you to resolve before the data is promoted. The mapping is then remembered — each file's column shape is kept as a fingerprint with a synonym store beside it, so the second file of that format takes a few clicks, and so does a whole directory of them.
Read the documentation →Document Assistant
Trains the system to recognise tables in a document by their shape. Once a shape has been seen, the File Catalog passes that table straight through. You inspect what the recogniser found, correct any mapping it got wrong, and claim the columns it missed — and those corrections join the knowledge base, so the vocabulary accumulates as more documents are read. The technique is not petroleum-specific; it works on any document with tables in it.
Read the documentation →Getting answers out
Once the data is in, three ways to ask it questions.
Mapping
A drop-down query tool and an AI query tool for the harder questions, scout tickets for the wells you select, and every document attached to them — final well report, LAS files and the rest. No SQL required. It has viewers for log data (LAS, LIS, DLIS), seismic (SEG-Y) and the usual desktop formats, and exports the selection as scout tickets, CSV, multi-page Excel, Petrel, GeoGraphix or OSDU.
Read the documentation →DB Explorer
Query, browse and export any table in the database directly.
Component document to followReference Tables
The controlled vocabularies loading checks against — well status, well type, units, depth datum, data source.
Component document to follow