Should we rename INKSetCacheUrl() ?
-----------------------------------
Key: TS-520
URL: https://issues.apache.org/jira/browse/TS-520
Project: Traffic Server
Issue Type: Improvement
Components: TSAPI
Reporter: Leif Hedstrom
Fix For: 2.1.5
I think we should rename INKSetCacheUrl() to something more accurately
describing what it really does (which is to change the cache key). Perhaps
INKSetHttpCacheKey()
This is to distinguish it from the existing cache key APIs, which work on
non-HTTP cache keys. Perhaps we should even take it a step further, and allow
for the cache-key string to not necessarily be NULL terminated? This is more in
line with existing HTTP / URL / Header APIs. E.g.
INKSetHttpCacheKey(INKHttpTxn txnp, const char *url, int len);
where if len is <= 0, we'll do a strlen() on it (which is the current
behavior). There can never be a reason to set the cache key to a zero length
string.
Thoughts?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
