Can you file an RFE? This issue has come up several times in the past
and I think we should make it optional or settable via a property.
Thanks,
Sean
Michael McIntosh wrote:
> Perhaps I was not clear. I know I can change the code. But due to
> constraints imposed by other organizations I cannot easily redistribute
> the changed code or binaries produced from it. I'd prefer that this code
> be changed in the Apache source. In the Java source there is currently no
> way to make these '\n's optional. The function looks like:
>
> public static void addReturnToElement(Element e) {
> Document doc = e.getOwnerDocument();
> e.appendChild(doc.createTextNode("\n"));
> }
>
> and calls to it are not protected by any if statements. On my system I've
> replaced it with this:
>
> public static void addReturnToElement(Element e) {
> if (null ==
> System.getProperty("org.apache.xml.security.util.XMLUtils.ignoreAddReturnToElement"))
>
> {
> Document doc = e.getOwnerDocument();
> e.appendChild(doc.createTextNode("\n"));
> }
> }
>
> I suspect there is a more efficient way that is more consistent with
> similar things in the XML Security project, but I'm not familiar enough
> with the code. Perhaps someone that is familiar code could recommend how
> this should be done, and even better could commit the change. Another
> responder suggested not changing the function behavior, but since it is
> called in so many places (66 times in 17 files) that would require too
> many changes.
>
> Thanks,
> Mike
>
> [EMAIL PROTECTED] wrote on 04/04/2007 05:01:35 PM:
>
>> I think, you can change the code as much as you want.
>> But there is some feature to make this return optionals. The only
>> place where the standard force it are in the Base64 representation
>> of the digest and signature value, that must be split at 80 chars
>> (apache xmlsec does not have any problem with this, but perhaps
>> other implementations relay on this)
>>
>> Regards,
>>
>> Raul
>>
>
>> On 4/4/07, Michael McIntosh <[EMAIL PROTECTED]> wrote:
>> I was refering to the Java version - but my notation has betrayed my C++
>> roots ;-)
>>
>> [EMAIL PROTECTED] wrote on 04/04/2007 03:31:49 PM:
>>
>>> Hi Mike,
>>> With your notation I think you are referring to C++ version. Don't
> you?
>>> On 4/4/07, Michael McIntosh < [EMAIL PROTECTED]> wrote:
>>> I know what XMLUtils::addReturnToElement does, by why do it?
>>> I acknowledge that adding the occasional line to an XML document makes
>> it
>>> more readable in certain circumstances, but I'd really like to be able
>> to
>>> turn it off.
>>> Actually I'd like to leave it alone, but a system I need to
> interoperate
>>> with cannot accept Signatures with whitespaces in certain places.
>>> No need to tell me its should be allowed - I know - but I cannot
> change
>>> their code.
>>> I have a fix - which is to change the source code for the funciton to
>> turn
>>> it into a no-op when a system property is set, but I'd like to not
> need
>> to
>>> redistribute my modified Apache source code.
>>>
>>> Thanks,
>>> Mike
>>>
>>> Michael McIntosh
>>> Java and Web Services Security Group
>>> Security, Privacy, and Extensible Technologies Department
>>> IBM Research
>>>
>>>
>>>
>>> --
>>> http://r-bg.com
>>
>>
>> --
>> http://r-bg.com
