[Bug] Spanish analyzer not normalizing all accented words.

[svera]

Hi,
I've been working with the spanish analyzer in order to index documents in spanish, when I found and issue with some accented words which are not normalised as they should be. This is easily reproducible using the Bleve text analysis wizard, choosing the es analyzer and putting fría or guía in the Text to analyze input box. The words are kept as they are, however, using plural forms of these words (guías and frías) works as expected.

Other accented words such as tentación or comeré are correctly stemmed to coleccion and comer, respectively.

[svera]

I've found where the issue comes from. Looking at bleve/analysis/lang/es/light_stemmer_es.go, the normalization of accented letters only happens if the input is larger than 5 characters, something that neither guía nor fría comply.
The solution would be to always execute the accented characters normalization, by moving it to a separate file just like it is done in the german analyzer, or maybe use the asciifolding filter.

[svera]

PR opened addressing this issue: #1957

[abhinavdangeti]

Thanks for raising the pull request @svera . The team will review soon.