> + continue;
> + String key =
> header.getKey().toString().toLowerCase(Locale.getDefault());
> + // Ignore any headers that are not particularly interesting.
> + if (key.equalsIgnoreCase(HttpHeaders.CONTENT_TYPE) ||
> key.equalsIgnoreCase("Content-MD5")
> + || key.equalsIgnoreCase(HttpHeaders.HOST) ||
> key.startsWith(HEADER_TAG)) {
> + canonicalizedHeaders.put(key, header.getValue());
> + }
> + }
> + return canonicalizedHeaders;
> + }
> +
> + private String buildCanonicalizedHeadersString(Multimap<String, String>
> canonicalizedHeadersMap) {
> + StringBuilder canonicalizedHeadersBuffer = new StringBuilder();
> + for (Entry<String, String> header : canonicalizedHeadersMap.entries())
> {
> + String key = header.getKey();
> + canonicalizedHeadersBuffer.append(String.format("%s:%s\n",
> key.toLowerCase(), header.getValue()));
Kind of an anti-pattern here, this compiles into another StringBuffer
allocation. Instead:
```
canonicalizedHeadersBuffer.append(key.toLowerCase())
.append(':').append(header.getValue()).append('\n');
```
---
Reply to this email directly or view it on GitHub:
https://github.com/jclouds/jclouds-labs-aws/pull/3/files#r12570927
