From c05dcb8a5d5c7ab510b561dfd37bb846fdcd6b0c Mon Sep 17 00:00:00 2001
From: Nathan Yee <nyee32@schedmd.com>
Date: Thu, 24 Sep 2015 13:50:07 -0700
Subject: [PATCH] Modify node list parsing to take expressions
Modify the slurm_addto_char_list to process hostlist expressions
rather than only a comma-delimited list of host names. This
effects sreport, sacctmgr, and other commands
bug 1923
---
src/common/slurm_protocol_defs.c | 165 +++++++++++++++++++++----------
1 file changed, 112 insertions(+), 53 deletions(-)
diff --git a/src/common/slurm_protocol_defs.c b/src/common/slurm_protocol_defs.c
index 3d474d826db..8e66e78d2f0 100644
--- a/src/common/slurm_protocol_defs.c
+++ b/src/common/slurm_protocol_defs.c
@@ -188,16 +188,31 @@ extern List slurm_copy_char_list(List char_list)
return ret_list;
}
+static int _find_char_in_list(void *x, void *key)
+{
+ char *char1 = (char *)x;
+ char *char2 = (char *)key;
+
+ if (!strcasecmp(char1, char2))
+ return 1;
+
+ return 0;
+}
+
/* returns number of objects added to list */
extern int slurm_addto_char_list(List char_list, char *names)
{
- int i=0, start=0;
- char *name = NULL, *tmp_char = NULL;
+ int i = 0, start = 0;
+ char *name = NULL;
ListIterator itr = NULL;
char quote_c = '\0';
int quote = 0;
int count = 0;
+ bool brack_not = false;
+ bool first_brack = false;
+ char *this_node_name;
+ hostlist_t host_list;
if (!char_list) {
error("No list was given to fill in");
@@ -218,69 +233,113 @@ extern int slurm_addto_char_list(List char_list, char *names)
break;
else if ((names[i] == '\"') || (names[i] == '\''))
names[i] = '`';
- else if (names[i] == ',') {
- /* If there is a comma at the end just
- ignore it */
- if (!names[i+1])
- break;
+ else if (names[i] == '[')
+ /* Make sure there is a open bracket. This
+ * check is to allow comma sperated notation
+ * within the bracket ie. linux[0-1,2] */
+ first_brack = true;
+ else if (names[i] == ',' && !first_brack) {
+ /* Check that the string before , was
+ * not a [] notation value */
+ if (!brack_not) {
+ /* If there is a comma at the end just
+ * ignore it */
+ if (!names[i+1])
+ break;
- name = xstrndup(names+start, (i-start));
+ name = xstrndup(names+start,
+ (i-start));
+ //info("got %s %d", name, i-start);
+
+ /* If we get a duplicate remove the
+ * first one and tack this on the end.
+ * This is needed for get associations
+ * with qos.
+ */
+ if (list_find(itr,
+ _find_char_in_list,
+ name)) {
+ list_delete_item(itr);
+ } else
+ count++;
+
+ xstrtolower(name);
+ list_append(char_list, name);
+
+ list_iterator_reset(itr);
+
+ i++;
+ start = i;
+ if (!names[i]) {
+ info("There is a problem "
+ "with your request. "
+ "It appears you have "
+ "spaces inside your "
+ "list.");
+ count = 0;
+ goto endit;
+ }
+ } else {
+ brack_not = false;
+ /* Skip over the , so it is
+ * not included in the char list */
+ start = ++i;
+ }
+ } else if (names[i] == ']') {
+ brack_not = true;
+ first_brack = false;
+ name = xstrndup(names+start, ((i + 1)-start));
//info("got %s %d", name, i-start);
- while ((tmp_char = list_next(itr))) {
- if (!strcasecmp(tmp_char, name))
- break;
- }
- /* If we get a duplicate remove the
- * first one and tack this on the end.
- * This is needed for get associations
- * with qos.
- */
- if (tmp_char)
- list_delete_item(itr);
- else
- count++;
-
- xstrtolower(name);
- list_append(char_list, name);
-
- list_iterator_reset(itr);
-
- i++;
- start = i;
- if (!names[i]) {
- info("There is a problem with "
- "your request. It appears you "
- "have spaces inside your list.");
- count = 0;
- goto endit;
+ if ((host_list = hostlist_create(name))) {
+ while ((this_node_name =
+ xstrdup(hostlist_shift
+ (host_list)))) {
+ /* If we get a duplicate
+ * remove the first one and tack
+ * this on the end. This is
+ * needed for get associations
+ * with qos.
+ */
+ if (list_find(itr,
+ _find_char_in_list,
+ this_node_name)) {
+ list_delete_item(itr);
+ } else
+ count++;
+
+ xstrtolower(this_node_name);
+ list_append(char_list,
+ this_node_name);
+
+ list_iterator_reset(itr);
+
+ start = i + 1;
+ }
}
+ hostlist_destroy(host_list);
}
i++;
}
- name = xstrndup(names+start, (i-start));
- while ((tmp_char = list_next(itr))) {
- if (!strcasecmp(tmp_char, name))
- break;
+ if (i-start) {
+ name = xstrndup(names+start, (i-start));
+ /* If we get a duplicate remove the
+ * first one and tack this on the end.
+ * This is needed for get associations
+ * with qos.
+ */
+ if (list_find(itr, _find_char_in_list, name)) {
+ list_delete_item(itr);
+ } else
+ count++;
+
+ xstrtolower(name);
+ list_append(char_list, name);
}
-
- /* If we get a duplicate remove the
- * first one and tack this on the end.
- * This is needed for get associations
- * with qos.
- */
- if (tmp_char)
- list_delete_item(itr);
- else
- count++;
-
- xstrtolower(name);
- list_append(char_list, name);
}
endit:
list_iterator_destroy(itr);
-
return count;
}
--
GitLab