[jira] [Updated] (EMAIL-205) Possible Dead Code in EmailUtils.encodeURL()?

["Adam Emerson (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/EMAIL-205?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Adam Emerson updated EMAIL-205:
-------------------------------
    Description: 
Hello,

I was doing some testing on the library for a university project and came across what may be some dead code in the {{EmailUtils.encodeURL()}} function.

The code in question is the if block at line 298:
{code:java}
for (final byte c : input.getBytes(US_ASCII))
{
    int b = c;
    if (b < 0)
    {
        b = 256 + b;
    }
...{code}
I realize that the {{b = 256 + b; }}condition is there as an attempt to handle the fact that bytes are signed in Java, however, when I look at the implementation of{{{} input.getBytes(US_ASCII){}}}, I don't see any case that it can return a negative byte value.  {{getBytes(US_ASCII)}} returns a '?' character for any non-ascii character, thus preventing an overflow that may result in a negative byte ('?' has a value of 63).  This would mean that the the contents of the if statement are unreachable.

If anyone with more experience has the time to take a look at this I would greatly appreciate it. For my own education, I would love to see a test case in which a negative byte value would be returned here.

Thanks!

 

  was:
Hello,

I was doing some testing on the library for a university project and came across what may be some dead code in the {{EmailUtils.encodeURL()}} function.

The code in question is the if block at line 298:
{code:java}
for (final byte c : input.getBytes(US_ASCII))
{
    int b = c;
    if (b < 0)
    {
        b = 256 + b;
    }
...{code}
I realize that the {{b = 256 + b; }}condition is there as an attempt to handle the fact that bytes are signed in Java, however, when I look at the implementation of {{{}input.getBytes(US_ASCII){}}}, I don't see any case that it can return a negative byte value.  {{getBytes(US_ASCII)}} returns a '?' character for any non-ascii character, thus preventing an overflow that may result in a negative byte ('?' has a value of 63).  This would mean that the the contents of the {{if }}statement are unreachable.

If anyone with more experience has the time to take a look at this I would greatly appreciate it. For my own education, I would love to see a test case in which a negative byte value would be returned here.

Thanks!

 


> Possible Dead Code in EmailUtils.encodeURL()?
> ---------------------------------------------
>
>                 Key: EMAIL-205
>                 URL: https://issues.apache.org/jira/browse/EMAIL-205
>             Project: Commons Email
>          Issue Type: Improvement
>            Reporter: Adam Emerson
>            Priority: Minor
>
> Hello,
> I was doing some testing on the library for a university project and came across what may be some dead code in the {{EmailUtils.encodeURL()}} function.
> The code in question is the if block at line 298:
> {code:java}
> for (final byte c : input.getBytes(US_ASCII))
> {
>     int b = c;
>     if (b < 0)
>     {
>         b = 256 + b;
>     }
> ...{code}
> I realize that the {{b = 256 + b; }}condition is there as an attempt to handle the fact that bytes are signed in Java, however, when I look at the implementation of{{{} input.getBytes(US_ASCII){}}}, I don't see any case that it can return a negative byte value.  {{getBytes(US_ASCII)}} returns a '?' character for any non-ascii character, thus preventing an overflow that may result in a negative byte ('?' has a value of 63).  This would mean that the the contents of the if statement are unreachable.
> If anyone with more experience has the time to take a look at this I would greatly appreciate it. For my own education, I would love to see a test case in which a negative byte value would be returned here.
> Thanks!
>  



--
This message was sent by Atlassian Jira
(v8.20.10#820010)