[openib-general] [PATCH 0/6] osm: QoS policy parser

Hal Rosenstock halr at voltaire.com
Mon Jan 22 12:05:03 PST 2007


Hi Yevgeny,

On Sun, 2007-01-21 at 03:46, Yevgeny Kliteynik wrote: 
> Hi Sasha.
> 
> Sasha Khapyorsky wrote:
> > Hi Yevgeny,
> > 
> > On 17:01 Wed 17 Jan     , Yevgeny Kliteynik wrote:
> >> Hi Hal
> >>
> >> The following series of six patches implements QoS policy file parser:
> >>
> >> 1. QoS parser Lex file
> >> 2. QoS parser Lex-generated c file
> >> 3. QoS parser grammar (Yacc) file
> >> 4. QoS parser Yacc-generated grammar c and h file
> >> 5. QoS parser header file that defines parse tree data structures 
> >> 6. Changes in makefiles and configure.in file for compiling QoS parser files
> > 
> > Is there any description of proposed format and functionality?
> 
> The parser is based on QoS RFC sent by Eitan in May 2006, with a few
> minor modifications. You can find the RFC here:
> http://openib.org/pipermail/openib-general/2006-May/022336.html
> 
> > Also what about using human readable formats?
> 
> To me the xml-like format in the RFC looks pretty readable.
> It has very limited number of keywords (tags), so it's easy 
> to follow and/or to modify.

Putting aside the issue of plain text versus XML file formats for a
moment, can an example of the XML format be supplied ? What are the tags
used and their relationships ? I don't think there's been a discussion
on this yet.

Also, why were lex and yacc chosen to be used rather than some open
source XML parser (already written in C) ?

I also have some questions about the patches but I'll wait to see more
of the bigger picture here.

-- Hal

> -- Yevgeny
> 
> > Sasha
> > 





More information about the general mailing list