[jira] [Commented] (SANTUARIO-267) high memory and processor time cost while signing big files

["Scott Cantor (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SANTUARIO-267?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13009663#comment-13009663 ] 

Scott Cantor commented on SANTUARIO-267:
----------------------------------------

No, I have no plans to do work of that magnitude on this code, and my use cases are for small messages. If somebody else wants to join the project and rearchitect the code, they are of course very welcome.

This isn't terribly helpful to you, but I didn't want you to labor under the impression there was any chance of me working on it, and right now I'm the only one maintaining this library.

> high memory and processor time cost while signing big files
> -----------------------------------------------------------
>
>                 Key: SANTUARIO-267
>                 URL: https://issues.apache.org/jira/browse/SANTUARIO-267
>             Project: Santuario
>          Issue Type: Task
>          Components: C++
>    Affects Versions: C++ 1.6.0
>         Environment: OS: Windows 7 x64 
> Build: MS Visual Studio 2010 (x86 and x64)
> Xerces 3.1.1
>            Reporter: Maxim Makeev
>            Assignee: Scott Cantor
>
> Test file (36MB) is XML document.
> Signing and verifying signature on this file takes 20 seconds, but .NET XML Sigature library makes it ten times faster..
> Bottleneck is canonicalisation. TXFMC14n::readBytes gets too much processor's time while operation. Any canonicalization type works this way.
> Application allocates ~360MB (x86) and ~700MB(x64) while signing. Half of this memory allocates while parsing XML and half while canonisation...
> Its to much, because we need to work with big documents. Is it possible to optimize this operations? 

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira