Re: [xml] Global error (xmlLastError) not cleaned up if xmlCleanupParser() is called on a different thread!

[Daniel Veillard <veillard redhat com>]

On Tue, Jun 10, 2014 at 09:38:48AM +0100, Alex Bligh wrote:
> On 10 Jun 2014, at 09:14, Daniel Veillard wrote:
>
> > > * For this specific case (error cleanup), I strongly believe that the
> > > problem could be corrected without affecting either the API or ABI, but I
> > > don't have time to investigate more since the workaround `xmlResetLastError`
> > > does work OK for me, as I only make libxml2 calls from C++ wrappers anyway.
> >
> > Windows may have special ways to reclaim the memory, the problem is that
> > there is no safe way to do the cleanup in general. It's thread local
> > storage too, and then behaviour is OS dependent (possibly compiler
> > dependant). I take patches but it's a very difficult area....
>
> FWIW last time this all came up (possibly my fault) I was going to send
> you a patch,
>
> What that would have done was:
>
> a) provide 2 new calls:
>    xmlContextPtr xmlNewContext()
>    xmlFreerContext(xmlContextPtr ctx)
>
> b) make these two increment a static variable, and if the variable was
> zero on entry, do what xmlInitParser does; and decrement a static
> variable, and if it reaches zero, do xmlCleanupParser. Make them return
> and free an opaque xmlContext struct that currently holds nothing.

  what happen is a sublibrary or a plugin uses libxml2 ? Then you crash
the problem is that this assume a very controlled context and
unfortunately that's not the case libxml2 is reused by a number of
libraries.

> c) mark xmlInitParser and xmlCleanupParser as deprecated
>
> This at least allows for multiple inits and frees between libraries,
> but does not (yet) solve the problem of global variables. Eventually
> we'd then make an xmlContext contain an xmlGlobalState and add
> variants of each API call that allowed for an xmlContext to be
> passed as the first argument; this could be relatively easily
> done with some nasty perl on the source code.

  This would work if everybody started writing fresh code around
libxml2, but that does not reflect reality unfortunately.

Daniel

-- 
Daniel Veillard      | Open Source and Standards, Red Hat
veillard redhat com  | libxml Gnome XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | virtualization library  http://libvirt.org/