ARCADE
Word track - Format of results

Format of results

The results will be provided in tabular form. All results (3724 lines of results + 1 header line) will be gathered in the same plain text file. A file naming convention will be distributed later. Column separators will be tabs.

The result file will be composed of four columns:

Cat	grammatical category of word
NoOccur	occurrence number
French	French list of tokens
English	English list of tokens

The two first columns must be identical to those of the word list distributed (see example of the dry run). Remember that the combination of the two fields uniquely identify each occurrence.

The column French contains all the tokens contained in the French part of the match. If the expression is multi-token, the tokens must be separated by a slash ("/"). Do not add any space or any other symbol. If the expression is composed of discontinous pieces, do not include the intermediary tokens.

The column English contains all the tokens contained in the English par of the match, with the same format convention for multi-token expressions. If the French word is not translated this field should be left empty.

The file must contain a one-line header with the name of the fields.

Example (corresponding to the verbs of the dry run):

Cat NoOccur French English

V 1 apporte is/provided

V 2 apportent represent

V 3 apporte/la/preuve demonstrates

V 4 apportent make

V 5 apporte contains

V 6 apporte provided

V 7 apporte has/been/providing

A full example is given here in text format (exact file that should be returned for the words of the dry run).