[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Intel's patch submission...
From: |
Mark Wielaard |
Subject: |
Re: Intel's patch submission... |
Date: |
18 Oct 2002 22:49:09 +0200 |
Hi,
On Sat, 2002-09-28 at 21:18, Brian Jones wrote:
> All, I've reformatted the Intel patch submission using a few command
> line tricks. I've attached the result here, let me know if you plan
> to work on any part of this.
I merged in the first batch:
2002-10-18 Mark Wielaard <address@hidden>
Merged patches from Wu Gansha <address@hidden>:
* gnu/java/io/PlatformHelper.java (INITIAL_MAX_PATH): new constant.
(toCanonicalForm): Use new constant to optimize StringBuffer usage.
Return lowercase path when isWindows == true.
* gnu/java/lang/SystemClassLoader.java (classpath): New static field.
(pathinfos): Likewise.
(getSystemResource): Refactored, cache partial result for each path
item in classpath.
* gnu/java/net/protocol/file/Handler.java (toExternalForm): New
method implementation specific to file protocol.
* gnu/java/net/protocol/http/HttpURLConnection.java
(buffered_out_stream): New field.
(connect): Don't set in_stream, wrap out_stream BufferedOutputStream.
Move sending request and receiveing reesponce to new methods.
(sendRequest): New method.
(receiveReply): Likewise.
(setRequestMethod): Accept "POST".
(getInputStream): When in_stream == null, call SendRequest() and
ReceiveReply().
(getOutputStream): New method.
* gnu/java/net/protocol/jar/Handler.java (toExternalForm): New
optimized method.
* gnu/java/net/protocol/jar/JarURLConnection.java (getContentLength):
New method.
* java/net/URL.java (URL): Remove toLowerCase() when calling this().
* java/net/URLConnection.java (req_props): Made final, set in
constructor using Hashtable(Map).
(setRequestProperty): Not synchronized.
(getRequestProperties): New method.
* java/net/URLStreamHandler (toExternalForm): Use StringBuffer.
No new Mauve regressions (after fixing a small thing in file Handler)
and a few new PASSes in the URLTest. But this was mostly optimization
code. Will look at the rest of java.net next and then the java.lang
things.
Cheers,
Mark
- Re: Intel's patch submission...,
Mark Wielaard <=