We need a routine that can run in Windows that can go through a text file of billingual definitions that looks like this, for example:
abbreviate abréger, abrégez, abrégent, rogner, abrégeons
assignation, enfranchisement, lending out attribution
aggrandize, augment propager, augmenter
And make it look like this:
lending out attribution
In other words, take any line where a word in one of the two languages has more than one word as a definition in the other language and split them into unique records. Also note the last line in the example where there is more than one word on each side of the definition (in this example two on each side, leading to four unique records).