Kouhei Sutou
null+****@clear*****
Thu Mar 30 11:25:07 JST 2017
Kouhei Sutou 2017-03-30 11:25:07 +0900 (Thu, 30 Mar 2017) New Revision: 528c32afba096c477348112e222f69071e7b6fde https://github.com/groonga/groonga/commit/528c32afba096c477348112e222f69071e7b6fde Message: grn_pat_scan: fix a bug that garbage leading space is added It's occurred when keyword doesn't start with single byte character. Added files: test/command/suite/select/function/highlight_html/space_before_target.expected test/command/suite/select/function/highlight_html/space_before_target.test Modified files: lib/pat.c Modified: lib/pat.c (+6 -3) =================================================================== --- lib/pat.c 2017-03-30 10:24:13 +0900 (5b168ff) +++ lib/pat.c 2017-03-30 11:25:07 +0900 (8e20fde) @@ -1,5 +1,6 @@ /* -*- c-basic-offset: 2 -*- */ -/* Copyright(C) 2009-2016 Brazil +/* + Copyright(C) 2009-2017 Brazil This library is free software; you can redistribute it and/or modify it under the terms of the GNU Lesser General Public @@ -2180,13 +2181,15 @@ grn_pat_scan(grn_ctx *ctx, grn_pat *pat, const char *str, unsigned int str_len, if ((tid = grn_pat_lcp_search(ctx, pat, sp, se - sp))) { const char *key; uint32_t len; + int first_key_char_len; key = _grn_pat_key(ctx, pat, tid, &len); sh[n].id = tid; sh[n].offset = (*cp > 0) ? offset : offset0; + first_key_char_len = grn_charlen(ctx, key, key + len); if (sh[n].offset > 0 && GRN_CHAR_IS_BLANK(tp[-1]) && - grn_charlen(ctx, key, key + len) == 1 && - key[0] != ' ') { + ((first_key_char_len == 1 && key[0] != ' ') || + first_key_char_len > 1)){ /* Remove leading spaces. */ const char *original_str = str + sh[n].offset; while (grn_charlen(ctx, original_str, str + str_len) == 1 && Added: test/command/suite/select/function/highlight_html/space_before_target.expected (+37 -0) 100644 =================================================================== --- /dev/null +++ test/command/suite/select/function/highlight_html/space_before_target.expected 2017-03-30 11:25:07 +0900 (8cbf283) @@ -0,0 +1,37 @@ +table_create Entries TABLE_NO_KEY +[[0,0.0,0.0],true] +column_create Entries body COLUMN_SCALAR ShortText +[[0,0.0,0.0],true] +table_create Terms TABLE_PAT_KEY ShortText --default_tokenizer TokenBigram --normalizer NormalizerAuto +[[0,0.0,0.0],true] +column_create Terms document_index COLUMN_INDEX|WITH_POSITION Entries body +[[0,0.0,0.0],true] +load --table Entries +[ +{"body": "Groonga 高速！"} +] +[[0,0.0,0.0],1] +select Entries --output_columns --match_columns body --query '高速' --output_columns 'highlight_html(body)' +[ + [ + 0, + 0.0, + 0.0 + ], + [ + [ + [ + 1 + ], + [ + [ + "highlight_html", + null + ] + ], + [ + "Groonga <span class=\"keyword\">高速</span>！" + ] + ] + ] +] Added: test/command/suite/select/function/highlight_html/space_before_target.test (+14 -0) 100644 =================================================================== --- /dev/null +++ test/command/suite/select/function/highlight_html/space_before_target.test 2017-03-30 11:25:07 +0900 (480a29b) @@ -0,0 +1,14 @@ +table_create Entries TABLE_NO_KEY +column_create Entries body COLUMN_SCALAR ShortText + +table_create Terms TABLE_PAT_KEY ShortText --default_tokenizer TokenBigram --normalizer NormalizerAuto +column_create Terms document_index COLUMN_INDEX|WITH_POSITION Entries body + +load --table Entries +[ +{"body": "Groonga 高速！"} +] + +select Entries --output_columns \ + --match_columns body --query '高速' \ + --output_columns 'highlight_html(body)' -------------- next part -------------- HTML����������������������������...Download