summaryrefslogtreecommitdiff
path: root/doc
diff options
context:
space:
mode:
authorShougo Matsushita <Shougo.Matsu@gmail.com>2016-12-08 07:55:08 +0900
committerShougo Matsushita <Shougo.Matsu@gmail.com>2016-12-08 07:55:08 +0900
commit2211f12e1dad2f453a3cf6a78ddb6d7d5162d569 (patch)
treeb37f7f861a0f621e46b9cc1a0b80fb18e826751e /doc
parent51fec4e233de05e7d8794e9beec64fde32a00b9b (diff)
downloaddeoplete.nvim-2211f12e1dad2f453a3cf6a78ddb6d7d5162d569.zip
Improve the documentation
Diffstat (limited to 'doc')
-rw-r--r--doc/deoplete.txt10
1 files changed, 8 insertions, 2 deletions
diff --git a/doc/deoplete.txt b/doc/deoplete.txt
index ee85303..51fd759 100644
--- a/doc/deoplete.txt
+++ b/doc/deoplete.txt
@@ -523,8 +523,14 @@ EXAMPLES *deoplete-examples*
SOURCES *deoplete-sources*
buffer *deoplete-source-buffer*
- It collects keywords from current buffer and the buffers which
- have same 'filetype'.
+ It collects keywords from current buffer and the opened
+ buffers which have same 'filetype'.
+
+ Note: It does not collect keywords from not loaded buffer.
+ For example, if you open file by "nvim A B". A is already
+ loaded, but B is not loaded. So buffer source cannot collect
+ the keywords in B until you switch to buffer B.
+
Note: It takes time to get the candidates in the first time if
you want to edit the large files(like Vim 22000 lines eval.c).