[Rd] R, AIX 64-bit builds - trying to understand root cause for message: "Error: Line starting 'Package: tools ...' is malformed!"

Michael Felt aixtools at gmail.com
Thu Dec 31 17:34:40 CET 2015


On 2015-12-31 17:02, Michael Felt wrote:
> On 2015-12-31 16:14, Michael Felt wrote:
>> - literal (&, M-^?) (38, 65535), pos 0, sub -1, 0 tags
>> + literal (&, M-^?) (38, -1), pos 0, sub -1, 0 tags 
> Getting down to these two lines - so I think it is now "before" 
Made a few changes - mainly in tre-ast.c

+135 static void
+136 tre_do_print(FILE *stream, tre_ast_node_t *ast, int indent)
+137 {
+138 tre_cint_t code_min, code_max;
+139 size_t pos;
+140 int num_tags = ast->num_tags;
+141 tre_literal_t *lit;

Lines 138 and 139 are now typed to what I 'think' the correct type is:

And, to better see the values:


+182 else
+183 {
+184 fprintf(stream, "literal (%c, %c) (%010x, %010x), pos %d, sub %d, "
+185 "%d tags\n", code_min, code_max, code_min, code_max, pos,
+186 ast->submatch_id, num_tags);
+187 }

changed %d to %010x for code_min and code_max

And the comparison is now shows:

64_bit: literal (&, ▒) (0000000026, 00ffffffff), pos 0, sub -1, 0 tags
32_bit: literal (&, ▒) (0000000026, 000000ffff), pos 0, sub -1, 0 tags


So, if -elsewhere- a test is being made for code_max - there is a 
difference. Strings and lists are not my thing. I do hope someone else 
has enough to find the 'type dependent' test or constant.

Michael



More information about the R-devel mailing list