I am new here, so maybe that is a former design decision but ./testrules.sh does not recognize <url></url> tags.
Probably to workaround this regression Portuguese grammar.xml also no longer has that information. That feature was working perfectly in version 3.5.
Should these type of regressions be fixed by removing contents?
What exactly do you mean? What error do you get or what error don’t you get that should be shown?
Running XML validation for pt/grammar.xml... cvc-complex-type.2.4.a: Invalid content was found starting with element 'url'. One of '{example}' is expected. Problem found at line 3562, column 8. Exception in thread "main" java.io.IOException: Cannot load or parse '/org/languagetool/rules/pt/grammar.xml'
Line 3562 is the one with the URL and the rule in question:
<!-- Concordance error person - I + Verb --> <!-- Created by Tiago F. Santos, Portuguese rule, 2016-10-16 --> <rule id="ERRO_DE_CONCORDNCIA_DO_NMERO_DO_VERBO" name="Erro de concordância do número do verbo"> <pattern> <marker> <token>eu</token> <token min="0" max="2" postag='CS|RG|RN' postag_regexp='yes'/> <token postag='VM[CIS][CFIMPS][23]S0|VM[CIS][CFIMPS][123]P0' postag_regexp='yes'> <exception postag='VM[CIS][CFIMPS]1S0' postag_regexp='yes'></exception></token> </marker> </pattern> <message>Erro de concordância verbal.</message> <example correction='Eu passeio'><marker>Eu passeiam</marker> na praia.</example> <url>http://www.infoescola.com/portugues/concordancia-verbal/</url> </rule>
The tags must appear in a certain order: <url> must always appear before the <example>s and after the <message>.
Ok. Another thing that somehow works on LO and does not in LT. I will change accordingly.
Thank you Daniel.