Sender: |
|
Date: |
Thu, 3 Mar 2011 10:59:46 -0500 |
MIME-version: |
1.0 (Apple Message framework v936) |
Reply-To: |
|
Content-type: |
text/plain; charset=US-ASCII; format=flowed; delsp=yes |
Subject: |
|
From: |
|
In-Reply-To: |
|
Content-transfer-encoding: |
7bit |
Parts/Attachments: |
|
|
Ah, I think you're right, non \u escapes would signal a failure.
Thanks Ben.
Sean
On Mar 2, 2011, at 4:50 PM, Ben Stabile wrote:
> Encoding/decoding of strings was not working correctly with escape
> characters for me. I had to change the decoding switch case somewhat
> because it was falling into the "unicode sequence" trap for no
> apparent reason.
>
> The following change was necessary to make this work (but see the
> warning below before making this change without checking your own
> sources):
>
> [Code.java, line 479]
>
> else if (c=='\\') // escape
> {
> x++;
> if ( x >= len )
> { d.type = DecodeReturn.T_ERROR; d.s =
> "Unterminated String"; return; }
> if (dat.charAt(x)!='u' && inUnicode)
> { d.type = DecodeReturn.T_ERROR; d.s =
> "Escape character in Unicode sequence"; return; }
> switch (dat.charAt(x))
> ...
>
> CAVEAT EMPTOR: I am working with significantly altered sources at
> the moment, so this may not show up for others the way it has for
> me. I am referencing the original source code linenumber above. If
> you don't use the "Code" class for this kind of thing, you might
> want to just ignore this. ;-)
>
> Regards,
> Ben S.
>
>
>
|
|
|