The table ingest service scans an ASCII file and attempts to interpret it as a table with coordinates or object names (which are then processed through the NED/SIMBAD name resolvers). The service is fairly forgiving, but it is still the responsibility of the user adhere to reasonable formatting and column identification rules.

The "corrected" data can be retrieved in column-delimited ASCII or VOTable format and will always have two columns ("ra" and "dec") which contain the coordinates for each record in decimal degrees (Equatorial J2000). It is possible that the coordinates will be unavailable for some records (either because no coordinate info was given or because an object name could not be found in NED or SIMBAD). In these cases, the RA,Dec columns will be blank.

This document outlines the functionality of the service and and provides examples of both good and bad input tables. The service is not fool-proof, so users should pay particular attention to the examples of what not to do.

Functionality

• Reformats input table headers and data whose values are separated by commas (CVS), pipes (vertical bars), spaces, or tab stops to fixed width columns separated by pipes (headers) and spaces (data). (I)
  

• Removes carriage returns and blank lines from table records. (II)
  

• Corrects misaligned records and variable length records to fixed width records. (III)
  

• Interprets tables without column names as a list of object names or positions written as full coordinate strings (e.g. "3h27m11.2s +12d16m34s Equ J2000" or "234.5 -11.2 ga" or just "10. 40.", which is assumed to be decimal degrees Equatorial J2000). (IV)
  

• Handles VOTables and Vizier TSV (Output layout: Tab-Separated-Values) formats as special cases.
  

• Determines what columns are best to use for coordinates or as object names and adds (if necessary) decimal-degree RA and Dec columns (J2000) to the output.
  

Common Mistakes

• Some tables cannot be unambiguously parsed but will still pass the service and therefore return incorrectly formatted output. (I)
  

• Some input is just uninterpretable. (II)
  

• Of course binary files (e.g.Excel spreadsheets) are just gibberish to the service. (III)
  

• Some input is otherwise perfectly understandable but fails because it is impossible to determine what information represents coordinates. (IV)