Call for vote on MS-RFC-28: LOG/DEBUG output mechanism

[Jeff McKenna]

+1 for this new MS_DEBUGLEVEL env variable and its useful settings.

As one of those users similar to Ed, I'm constantly looking at DEBUG 
messages in layers/maps for performance.  Sometimes (like my sample 
debug output below) the messages are not that useful, showing only a few 
layer's drawing time and the total time.  So if RFC#28 improves the 
output, I'm all for it.

   msDrawMap(): Layer 0 (Ontario_Landcover), 0.141s
   msDrawMap(): Layer 1 (Conservation_Area), 0.000s
   msDrawMap(): Layer 2 (Provincial_Parks), 0.000s
   msDrawMap(): Layer 3 (National_Parks), 0.000s
   msDrawMap(): Layer 4 (Municipal_Boundary), 0.000s
   msDrawMap(): Layer 5 (Municipal_Boundary_Lower), 0.00
   msDrawMap(): Layer 6 (Geographic_Lot_Fabric), 0.000s
   msDrawMap(): Layer 7 (Geographic_Township), 0.000s
   msDrawMap(): Layer 8 (NTS_Railway), 0.031s
   msDrawMap(): Layer 9 (DMTI_Major_Roads), 0.000s
   msDrawMap(): Drawing Label Cache, 0.000s
   msDrawMap() total time: 30.406s

Oh, and I don't know of anyone who uses the LOG option, which is maybe 
why it is so poorly documented.

--
jeff




Daniel Morissette wrote:
> Hi everyone,
> 
> I have made a last round of updates to MS-RFC-28 to add a section about 
> an optional MS_DEBUGLEVEL environment variable which can be used to turn 
> on debug before a map is loaded, mostly useful to figure initialization 
> and map loading time (following a comment from Ed I think).
> 
> The way the debug levels work vs the MS_ERRORFILE are unchanged since I 
> didn't get any vote in favor of option 2 following my email from yesterday.
> 
> With this email I would like to call for a vote to adopt MS-RFC-28 to be 
> included in MapServer 5.0:
> http://mapserver.gis.umn.edu/development/rfc/ms-rfc-28/
> 
> I start with my +1.
> 
> Daniel