Fix pgbench lexer's "continuation" rule to cope with Windows newlines.

tglsfdc · tglsfdc · commit 3ea7e015f37a · 2018-09-17T12:11:43.000-04:00
Our general practice in frontend code is to accept input with either Unix-style newlines (\n) or DOS-style (\r\n). pgbench was mostly down with that, but its rule for line continuations (backslash-newline) was not. This had been masked on Windows buildfarm machines before commit 0ba06e0 by use of Windows text mode to read files. We could have fixed it by forcing text mode again, but it's better to fix the parsing code so that Windows-style text files on Unix systems don't cause problems. Back-patch to v10 where pgbench grew line continuations. Discussion: https://postgr.es/m/17194.1537191697@sss.pgh.pa.us
diff --git a/src/bin/pgbench/exprscan.l b/src/bin/pgbench/exprscan.l
@@ -67,7 +67,7 @@ nonspace		[^ \t\r\f\v\n]
 newline			[\n]
 
 /* Line continuation marker */
-continuation	\\{newline}
+continuation	\\\r?{newline}
 
 /* Exclusive states */
 %x EXPR
@@ -104,8 +104,12 @@ continuation	\\{newline}
 	 * a continuation marker just after a word:
 	 */
 {nonspace}+{continuation}	{
-					/* Found "word\\\n", emit and return just "word" */
-					psqlscan_emit(cur_state, yytext, yyleng - 2);
+					/* Found "word\\\r?\n", emit and return just "word" */
+					int		wordlen = yyleng - 2;
+					if (yytext[wordlen] == '\r')
+						wordlen--;
+					Assert(yytext[wordlen] == '\\');
+					psqlscan_emit(cur_state, yytext, wordlen);
 					return 1;
 				}
 
