An example file, data.olf, is included in the Samples directory.

In OLIF, there is no provision to declare a file-wide source or
target language. Rather, these data categories are expressed implicitly
throughout the file: Each mono is in one of only two languages, each mono
in the source language bears a transfer to one in the target language,
and no mono in the target language bears a transfer. The language code
must be expressed in an xml:lang attribute of the canForm, not solely
in a language attribute. If these requirements are violated, the file
is not convertible.

Subject field is expressed as in all OLIF. It must be uniform across
the whole file for convertibility. Per the OLIF 3.0 specification,
subject fields outside the specified list must be prefixed with x and
a hyphen. This prefix is removed when converting into other formats.

A glossary-wide note is expressed as in all OLIF.

Source and target terms must be matched by transfers, not by the use
of shared concept IDs; the multidirectional semantics of the concept
grouping are not convertible. For the same reason, no mono may bear more
than one transfer; conversion accommodates exactly one source and one
target term per concept. (An unknown term may be represented by a mono
with a blank canForm, but not by an absent mono.)

Part of speech is required on all monos, as in all OLIF. Convertible
values are adj, adv, noun, verb, and x-properNoun.

Note on an entry must be expressed within the generalDC of the source
mono for convertibility, even if the note actually concerns the target
term, a transfer relationship, etc.

Definition, source of definition, contextual example, and source of
contextual example are expressed using the definition and example
elements, as in all OLIF. Within these elements, though, additional
wrapper elements are used (borrowed from the TBX namespace and standard),
to clearly separate the substance of the definition or example from its
source citation.

No data category may take a value containing a tab character; such values
are not convertible.

OLIF mandates two non-convertible data categories: creator ID and
language ID use. When OLIF is selected as the output format, the
creator ID will be set to 'user', and the language ID use will be set to
'region_standard'. These fields may be edited by hand if desired.
