Hi there!
I’m having a hard time understanding a difference in operation while debugging a regular expression.
I’m trying to detect if the word “even” from the source is absent from the target. As you can guess, the word “even” has plenty of translations in my language (Spanish), so I ended up with a rather long regular expression:
Source term: "even"
Match whole word
Target term: -"(aún|aun(que)?|todavía|hasta|incluso|ni siquiera|plan(o|a)s?|parej(o|a)s?|igualad(o|a)s?|uniformes?|nivelad(o|a)s?|misma altura|regular(es)?|empatad(o|a)s?|par(es)?|lis(o|a)s?|alinead(o|a)s?|a mano|igual(es)?|a pesar|si bien|estables?|siquiera)"
Match whole word
This entry, however, throws a false positive with the following sentence:
ST: One or more of your products show as unavailable for purchase at checkout even though they’re displayed as being in stock on your online store.
TT: Uno o más de tus productos se muestran como no disponibles durante la confirmación de compra, aunque aparecen como en stock en la tienda en línea.
It’s a false positive because the TT contains “aunque”.
If I add < and > to enclose the target expression, the entry works correctly, but I want to understand why it doesn’t work to begin with. Perhaps I’ve been missing something and I might have to correct several entries I created in a similar way.
It also seems to be an issue with the “par(es)?” part of the expression, since it works normally if I remove it. The issue still happens even if I separate “par(es)?” into “par|pares”.
Can someone help me with this?
Thanks a lot!