That commit is from the 05th November.
If we consider the detections introduced by rule HÁ-ATRÁS, there was 43 hits on the regression test of 5th November. But they don’t change results since they change only suggestions and whitespaces in the pattern.
So… 113 matches that you do not even know from where they come from and you still think it is a good solution having 120k or random words added? It seams to me that it is not reviewable even by you since you can not find the errors it introduces and I believe it is not up to me, to find them, is it?
I told you before, and I reaffirm:
Put your name in the file, if you want to, but revert your changes.
They do not had nothing that is not done in a easier to maintain way, by a rule.
More, if you find any way to compress my rules into a lower number of total rules, please, tell me and I will also do it.
@tiagosantos
Tiago, I am very stressed and can’t think properly right now.
Could you please revert my changes in the 120k compounds then?
In January I will try to add compounds to it but only words and not verbs.
I will have to analyse the words in the pt_PT speller to see which ones are not verbs and it will take a long time.
Right now I need to dedicate more time to the PhD project since next week I will be a few days in the North with the PhD coordinator and won’t have the chance to do much.
Tiago, I noticed the other day that one of the redundancy rules you added already existed added by me months ago, the “hemorragia de sangue” which means there are two rules for the same.
Sure Marco. I will just replace the word list and leave the header as you saw fit.
I am certain you will add many more words in due time. The verbs are already covered.
This is an extra, so focus on the PhD. No need to do much. Just consistent improvement over time. We will get there.
No worries. I will review coverage and comment out the redundant one.