#827 PO fuzzy counted as 100% instead of 60% in match statistics

[Didier Briel]

A PO document can contain "fuzzy" entries:

#, fuzzy
msgid "This can be a source text"
msgstr "This can be a totally unrelated target text"

While technically these entries may look translated and be 100% matches, most of the time the target text is loosely related (if at all) with the source text.

For this reason, OmegaT consider arbitratily these entries as 80% matches in the Fuzzy Matches pane, where they appear with a "fuzzy" prefix:

[Fuzzy]This can be a source text
This can be a totally unrelated target text
<80/80/80% test.po>

Match statistics doesn't take this into account and consider these entries as 100% matches.

This should be corrected, so that the percentage is the same as the one in the Fuzzy Matches pane.

Didier

[Jean-Christophe Helary]

This should be corrected and set to 60% since that's the threshold set by gettext to declare a previous string a fuzzy.

[Didier Briel]

As I wrote in the user group, having the match statistics behave consistently with the Fuzzy Matches pane and the actual percentage used are two different things (from a technical point of view).

You should create an RFE for that.

Didier

[Didier Briel]

This is implemented in RFE [#1258].

Didier

[Didier Briel]

Fixed in SVN (/trunk).

Didier

[Didier Briel]

Fixed in the released version 4.1.1 of OmegaT.

Didier