Arne Deutsch 3bc181024e [#69] Fix content assist if lexer returns multiple invalid tokens. ... If the lexer returns multiple invalid tokens in succession only the last one is taken into account for content assist. Especially if there are keywords only (no ID or similar) this is an issue because only the last character is used as prefix. The fix assumes that ALL tokens with syntax errors that come in succession are relevant for content assist. It seems not to be useful to execute content assist otherwise. This can never be reproduced with any grammar that includes standart "Terminals" because of the ID and the ANY_OTHER rule. The ID rule makes sure that the start is usually read as only ONE token, whereas the ANY_OTHER rule ensures there is never a lexer error. Signed-off-by: Arne Deutsch <Arne.Deutsch@itemis.de>		2018-06-25 17:50:24 +02:00
..
.settings	[JavaFormatting] set auto-wrap trigger to 120 chars	2017-08-15 17:30:15 +02:00
META-INF	[release] Increment version to 2.15.0	2018-05-22 23:53:44 +02:00
src/org/eclipse/xtext/ide	[#69] Fix content assist if lexer returns multiple invalid tokens.	2018-06-25 17:50:24 +02:00
xtend-gen/org/eclipse/xtext/ide	added missing since annotations	2018-06-15 19:15:40 +02:00
.classpath	Refreshed projects with Buildship 2.0	2017-01-12 10:32:06 +01:00
.project	gradleprojectbuilder added by Buildship project import	2017-07-01 23:09:36 +02:00
about.html	Moved projects to root of repository	2016-06-14 08:30:25 +02:00
build.gradle	added ecore.change to dependencies of xtext.ide	2017-08-15 17:30:15 +02:00