Problems upgrading to 4.6.1 or 4.8.0-be ta2 from 3.6.6 (Not about the foundatio
Doyon, Jean-Francois
Jean-Francois.Doyon at CCRS.NRCAN.GC.CA
Wed Nov 30 14:32:06 PST 2005
Steve,
Thanks for the quick reply, I realize you're busy with that other thing too
:P
1 & 2 are the really bizarre ones ... I've noted them on the same mapfile
for now, maybe they're related? I upgraded precisely for this particular
map we're working on! (Which will blow your mind too :) I'm wondering if
maybe I was testing with 4.6.0 before ... Hmmm. I did the DEBUG/LOG trick,
didn't tell me anything more.
3: I'll try to get to it tomorrow ... I'm not even sure what layer it is
yet, I just see a number of repeated identical errors "msEvalExpression():
General error message. Cannot evaluate expression, no item index defined."
The number of repetition varies between maps I think ... Looks like one per
class maybe?
4: I'll see what I can do :)
Thanks again!
J.F.
-----Original Message-----
From: Steve Lime [mailto:steve.lime at dnr.state.mn.us]
Sent: November 30, 2005 5:08 PM
To: Jean-Francois.Doyon at CCRS.NRCAN.GC.CA; MAPSERVER-USERS at LISTS.UMN.EDU
Subject: Re: [UMN_MAPSERVER-USERS] Problems upgrading to 4.6.1 or
4.8.0-beta2 from 3.6.6 (Not about the foundatio
Yeah! (on many counts)
1: I've tested REQUIRES a good bit and have had very good luck. I will test
that specific setup and let you know.
2: Again, easy to test on my end.
3: Specifics of when would really help, especially on where the item in
question is in the layer definition and what type of operation you are
attempting (query vs. browse)
4: Some addtional checks have been put in place to deal with corrupt
shapefiles or xbase files, but all *should* be accompanied by an error
message. Don't suppose you saved a version of the seemingly corrupt file?
We'll get you fixed up, don't worry. Will let you know...
Steve
>>> "Doyon, Jean-Francois" <Jean-Francois.Doyon at CCRS.NRCAN.GC.CA>
>>> 11/30/05 3:48 PM >>>
Hello,
So we've bitten the bullet and started the process to upgrade from 3.6.6 to
4.6.1 (And possibly 4.8).
Mostly, things came across as expected, knowing what has changed and so on.
Problem 1: REQUIRES.
I'm aware of the bug in 4.6.1, so I'd backported the fix into it, and things
were working ok for little while.
Now however, it just won't work :(
I've tried my version of 4.6.1 AND a "stock" 4.8.0-beta2, without success.
Here's an example:
Layer 1:
LAYER
NAME vegetation_250K_legend_zoom_3_4
STATUS OFF
GROUP base
TYPE POINT
MINSCALE 42001
MAXSCALE 300001
METADATA
"nolaydesclink" "2"
"lay_title" "Vegetation"
"checkable" "2"
"legend_order" "40"
END
CLASS
NAME "Cut Line: Firebreak"
SYMBOL cutline_firebreak_alt
SIZE 20
END
CLASS
NAME "Cut Line: Other"
SYMBOL cutline_other_alt
SIZE 20
END
CLASS
NAME "Wooded Area"
SYMBOL wooded_alt
SIZE 20
END
END
Layer 2:
LAYER
NAME vegetation_polygon_250K_map_zoom_3
STATUS DEFAULT
GROUP base
TYPE RASTER
MINSCALE 80001
MAXSCALE 300001
REQUIRES "[vegetation_250K_legend_zoom_3_4]"
METADATA
"legend_order" "-2"
"wms_srs" "EPSG:42304"
"wms_name"
"VEGETATION_2:BNDT_250K/NTDB_250K"
"wms_server_version" "1.1.0"
"wms_formatlist"
"image/gif,image/png,image/jpeg,image/wbmp"
"wms_format" "image/gif"
END
CONNECTIONTYPE WMS
CONNECTION "http://mydomain.com"
END
Unfortunately, turning on layer 1 does not make layer 2 appear, even though
it's "DEFAULT" ... The only way this works is if we comment out the REQUIRES
in Layer 2 ... Then it'll appear.
I have no idea what happened to either make my custom 4.6.1, or even worst,
the 4.8 beta, NOT work :(
I upgraded libcurl to 7.15.0 ... Could that be a problem? I also use
gdal/ogr 1.3.0 and gd 2.0.28 ...
Problem 2:
Pixmap symbols don't appear in the legend (An HTML legend). The symbols are
just white or transparent gif's (Of the right size, 20x20 in this case).
Vector symbology still works fine, other symbols are gif files, these are
the problematic ones.
Problem 3:
Intermittent "no index item defined" errors. I have to try and track this
down further ... But if anyone has already run into this?
I saw a few discussions on the topic, but nothing that seemed relevant in my
case.
One problem we had where mapserver would die without an error message (well,
an empty one) was solved by re-generating the data (shapefiles) ... Are
these versions of mapserver more data sensitive somehow?
Thanks in advance for any help! I'm dying here :P
Jean-François Doyon
Internet Service Development and Systems Support / Spécialiste de
dèveloppements internet et soutien technique Canada Centre for Remote
Sensing/Centre Canadien de télédétection Natural Resources Canada/Ressources
Naturelles Canada http://atlas.gc.ca
Tel./Tél.: (613) 992-4902
Fax: (613) 947-2410
More information about the MapServer-users
mailing list