Never mind - found it in svgLayers() and pushed the tracks back into the
background.
My solution is
- have the track in the background
- make it wider than the ways
- color it an ugly cyan with 0.3 opacity
This way it is easily visible under the ways but does not obscure the
imagery. You might
Yes, that was my point.
So it was better originally but got moved into the foreground intentionally?
Is there a way to switch the order of those layers (in the code) to put it
into the back again?
--
Sent from: http://gis.19327.n8.nabble.com/Developer-Discussion-f5233107.html
I was testing the option to include GPX tracks in the map and I noted that
the layer is shown* in front of the* OSM data.
I would have expected the GPX track to be behind the OSM data so it is not
obscured and you can easily draw your ways on top of the GPX track.
Potlatch2 does it this way.
Is
Bryan Housel wrote
> Somebody made:
> https://github.com/openstreetmap/iD/pull/3140
> https://github.com/openstreetmap/iD/pull/3140;
> which I probably would have just accepted if it didn’t conflict with the
> existing "selected" style.
> (whatever we do needs to look and feel different than how
Eventually I managed to persuade iD to use my own icon images instead of the
maki library. Actually it took less than a dozen lines of code to implement
the patch - it just was very difficult to find those lines.
The taginfo service and the nominatim service still don't work. A mystery as
the osm
Bryan Housel wrote
>> I am not quite sure why the nominatim service does not work. It also
>> appears
>> that the taginfo service does not work either. Those fields which are
>> supposed to get their values from taginfo also remain empty. I assume
>> that
>> is has something to do with the test
Hi,
I have been trying to add a missing preset to id, for
tourism=trail_riding_station.
It is defined with "fields": [ "name", "address", "phone", "website",
"description" ]
There are two problems:
1. below the expected fields, there is an additional field appearing
"tourism = attraction",
Hi!
I strongly advice against a general mass messaging API. If there is a
messaging API it can be used for spamming. That we already have minor
spamming issues is no argument for making them worse.
A mechanism where a user has to explicitly opt-in into receiving mass
messages when authorizing
Hi!
On a particular machine, osmosis is throwing the following exception:
Okt 20, 2013 2:30:36 PM org.openstreetmap.osmosis.core.Osmosis run
INFO: Osmosis Version 0.39
Okt 20, 2013 2:30:36 PM org.openstreetmap.osmosis.core.Osmosis main
SEVERE: Execution aborted.
java.lang.NullPointerException
tmcw wrote
Is it really a terrible weight to update JOSM to recognize the new format?
It is not just about JOSM, the change breaks *every application, web service
or browser plugin* that creates a reference to the osm main page or that
could parse an OSM url.
That's a little bit more :-).
Derick Rethans wrote
* barrier nodes?
Yes
This actually causes some issues. In lots of the UKs county side mapping
gates and stiles are mapped on when you need to climb over a fence or
something. They are often just mapped with barrier=gate or
barrier=stile.
Acutally, GH already
Kai Krueger wrote
I was curious to see how it compares in speed and quality of calculated
routes to the other engines, so I took the liberty to add it to
http://apmon.dev.openstreetmap.org/routing
I took the liberty of comparing the result on that page to the direct call.
How realistic are the
Hi!
The editor iLOE is creating empty tags, probably because of some bug or
missing check.
See http://www.openstreetmap.org/browse/node/2341541875 for an example.
Unfortunately, the author has only given forms of contact that force
membership on the respective sites. Does anybody know his OSM
Hi!
Tonight a very stupid question. I got interested in the new ID editor and
I've been trying to find out more about it, but without success.
Potlatch2 has a wiki page with getting started info, how to set it up,
customize etc. and a mailing list. JOSM has its own Wiki with technical
stuff and
Hi!
I created a trac ticket #8644 for a wrong preset - but then I noted that
there were some pretty old and unresolved tickets in the list already. So I
am in doubt whether that was the right thing to do to get that mistake
fixed. Any suggestions?
bye, Nop
--
View this message in context:
Hi!
In the German forum, a Mapper was investigating missing data which he had
mapped. Tracing it back he found that the current planet file was 8 GB
smaller than the previous version.
This can be easily confirmed on http://planet.openstreetmap.org/planet/2013/
planet-130213.osm.bz2
Hi!
Brett Henderson wrote
I believe it was fixed in Osmosis 0.38. But I'd recommend using 0.41 (ie.
the latest) if possible.
Unfortunately, it is only partially fixed. I tried Osmosis 0.41. It works if
you use the indexType Dynamic, but it fails with a MAXINT exception if you
try to use
Hi!
So which of the two statements is true?
1. 64bit IDs are only supported in 64bit tracker, so 0.42 will change
nothing and it's just a matter of documentation. The 64bit wiki page is
wrong.
2. 64bit IDs are supported in all index types in 0.42 as stated on the 64bit
wiki page.
bye, Nop
Hi!
The osm2pgsql on my server, which has been working nicely for a long time,
is now crashing with a segmentation fault every time and I would like to aks
you for some help and ideas on how to find the reason.
Operationg system is a 64bit debian. The osm2pgsql was compiled from the
then-current
Hi!
Ok, here we go again without markup:
The osm2pgsql on my server, which has been working nicely for a long time,
is now crashing with a segmentation fault every time and I would like to aks
you for some help and ideas on how to find the reason.
Operationg system is a 64bit debian. The
Hi!
I can give you an indication about the distribution of tile requests by zoom
levels on my hiking map server at wanderreitkarte.de.
=z8z9 z10 z11 z12 z13 z14 z15 z16 z17
z18
19% 3% 4% 5% 6% 10% 14% 18% 8% 5%
Hi!
When you observe the db server, it appears that the whole planet import done
by osm2pgsql is one huge transaction.
On one hand, this is not required in my case as the import is done into
shadow tables and activated later at a defined time. On the other hand, the
db server allocates large
Hi!
Small update: After moving from the test installation to the live instance,
the locale does work. Though I have no idea why as nothing has changed in
the embedding. Maybe some flash caching issue?
Now that it basically works, I noted that the translation of the UI is still
incomplete.
Hi!
Time for the next round of feedback.
I did another update, using random.dev this time again. It seems current and
my instance is working.
I tried the assets feature and it seems to work nicely.
Only the locale remains a problem. I copied from osm.org and used
args[locale] = de_DE;
but
Hi!
I played at migrating to P2.3 again and ran into a few problems.
1. getting the locales. I downloaded potlatch2.swf from osm.org and took the
resources from a git zip. But there are no locale files easily accessible. I
assume that you need to run p2 with a matching set of locales and
Hi!
Richard Fairhurst wrote:
NopMap wrote:
How would I go about updating a custom instance to P2.3?
Wiki instructions and random.dev appear to be outdated.
The config files have also changed a little (particularly
map_features.xml). If you use Safari's Activity Viewer
Hi!
How would I go about updating a custom instance to P2.3?
Wiki instructions and random.dev appear to be outdated.
bye
Nop
--
View this message in context:
http://gis.638310.n2.nabble.com/Potlatch-dev-Potlatch-2-3-tp6848644p6898615.html
Sent from the Potlatch mailing list
Hi!
Looks pretty good on the main site.
Is there a binary with example environment somewhere to update custom
instances with? random.dev still seems to be using Potlatch 2.2 and you
can't read the dirs on the main site.
bye
Nop
--
View this message in context:
It's good to know that it should be basically possible. However, my
knowledge of linux communication mechanisms, pearl and python is not good
enough to build something similar from scratch, let alone setting up python
in an efficient way.
So I'd still be looking for an existing python skeleton
Hi!
My server is using tirex and a fairly standard mapnik rendering backend to
create map tiles. There are also tiles with elevation contour lines which
are pre-rendered offline and uploaded. Not exactly state of the art.
I would like to put tirex in control of rendering those contours also.
sylvain letuffe wrote:
Are you using mod_tile ? does it permit all that ? have you an other queue
management/penalty/rate limiting software ?
The server uses mod_tile, but I am not using any of the throttling
mechanisms.
I have a seperate application that analyses the tile requests in
You know about the throttling of download speed in mod_tile, I guess.
On my server I use a tile limit per zoom level (automated tile scrapers show
an exponential increase with higher zoom levels) and a penalty for
missing/fake agents. Excessive IPs are blocked for 24h.
Actually, a good strategy
Frederik Ramm wrote:
I'm surprised that could happen; even if the master was frozen, then the
most that should happen is that every mod_tile attempt to speak to the
master would fail, but normal Apache operations (not talking to master)
should continue normally.
I have no idea how it
Frederik Ramm wrote:
I don't normally bother doing tile expiry; I just
have a cronjob running that stuffs the oldes 10k tiles into the queue
for re-rendering once per hour. This means that my low-priority queue is
*always* full, so tirex will *always* keep maxproc CPUs of the lowest
I don't know whether it will become important, but here's a quick heads-up:
The application GlobalMapper has released an update yesterday which includes
built-in OpenStreetMap world maps for tile downloading.
A quick check showed some attribution, no mentioning of a licence.
bye
Till Harbaum / Lists wrote:
Then firefox is violating the license as well since i can enter a tile url
and firefox will display
it without the license.
That is purely theoretical. There is a difference between a single tile
(experimental, practically useless) and a slippy map/map
I tried Locus on my phone and traced the accesses in the server log. I
positively is the culprit.
Needless to say that the program contains no proper attribution and no
license hint whatsoever. :-(
I have just banned it from my server.
by
Nop
--
View this message in context:
Igor Brejc wrote:
Leaving legality, morality and attribution aside, the application still
doesn't properly identify itself when contacting any server and performs
actions which the developer knows very well is not considered a good
practice.
That is exactly the point. From the blog and
Hi!
Paging the admins of the OSM tile servers. I'd like to get in touch about
the topic of mass downloaders. It appears that some problems first become
visible on small tile servers like my own, but hit the main tile servers a
while later, so it might be a good thing to share experiences. E.g.
Grant Slater wrote:
I too am perplexed by the MIDP user-agent. I believe the user-agent is
from Java Mobile Edition. Some? BlackBerry's prefix their user-agent
eg: BlackBerry9800/6.0.0.246 Profile/MIDP-2.0 Configuration/CLDC-1.1.
Basic tile.osm.org user-agent stats:
I can only speak for my own server, but those devs who have discussed the
matter and limited their applications to a reasonable amount of data have
gotten undamped access within those bounds.
Unfortunately, most folks never had the notion to ask whether a few million
tiles are ok - and some even
Ben Supnik wrote:
But I do feel that the area as an OSM primitive is badly needed. Here
even _with_ external semantics it is sometimes difficult to determine
whether an OSM way represents a polyline or a closed polygon.
Agreed 100%. :-) If we end up with a real area primitive and nothing
Markus Wagner wrote:
pointers. You think I can sufficiently improve the situation by adding
one or two more drives (and getting faster ones) to a software
RAID-0?! Or does it have to be a hardware RAID?
I have gotten very good results by adding a 128 MB SSD for the db. They are
Mike Collinson wrote on dev:
... the License Working Group intends implementing Phase 3 of the
license change implementation plan [1]. This involves blocking edits
with HTTP Forbidden messages until the individual contributor has
Accepted/Declined the new terms
Is Potlatch already equipped to
While talking about osm2pgsql: Is there a current windows build available
somewhere?
I have not heard of any. Is it possible to build a windows version at all or
is this known to be broken?
bye
Nop
--
View this message in context:
Hi!
Henning Scholland wrote:
I thought, I told you about this error in German forum...
I remember that post. It just didn't sound like the same thing. From that I
would have expected the reader to break off and throw an error message, not
quietly drop half the file. Also, it never occured
After playing around a bit, I am even more confused what needs to go where.
I put gpx.css into stylesheets/ and now it is recognized for tracks and the
tracks are drawn - the drop down box in the vector file list in the GUI
remains empty however.
If I move my main (custom) stylesheet rwk.css
Richard Fairhurst wrote:
It does, but you need to put it in the stylesheets/ directory.
It used to work with the style sheet simply in the same directory as p2.
Does it need all stylesheets in a subdir? The main stylesheet is still
working from the main dir.
Is there any way to find
So the information on
http://wiki.openstreetmap.org/wiki/Potlatch_2/Deploying_Potlatch_2 is
totally obsolete? That's what I used to set up my instance and nothing has
changed there.
Do you actually mean code? As I don't have a flash build system, code would
be rather useless to me.
rather
The problem still persists in build 25557.
--
View this message in context:
http://gis.638310.n2.nabble.com/Bug-Selection-broken-after-save-tp6092692p6164105.html
Sent from the Potlatch mailing list archive at Nabble.com.
___
Potlatch-dev mailing
When you select the menu item in the P2 on wanderreitkarte.de, Flash throws
an error message:
TypeError: Error #1009: Der Zugriff auf eine Eigenschaft oder eine Methode
eines null-Objektverweises ist nicht m�ch.
at net.systemeD.halcyon.styleparser::RuleSet/parse()
at
Richard Fairhurst wrote:
I think that's a sufficiently huge bug that we'd have noticed if it were
the case for everyone. :)
So it would seem :-)
Richard Fairhurst wrote:
Can you provide steps to reproduce? What's the URL you're invoking P2
with (lat/long/zoom)? What are you doing?
Now I installed the content debugger and the problem disappeared.
Before debugger installation it was 100% reproducible.
After debugger installation it worked several times.
This would explain why you never observed it. But now we have the problem
how to debug something that is changed by the
Hi!
Richard Fairhurst wrote:
Interesting. On wanderreitkarte it fails for me on FP 10.2, no matter
whether I'm using the debug version or the non-debug version. But
random.dev.openstreetmap.org works fine. I wonder if there's something
in your stylesheets that's unearthing a bug in
The subject says it all.
bye
Nop
--
View this message in context:
http://gis.638310.n2.nabble.com/Can-P2-handle-WMS-imagery-tp6069931p6069931.html
Sent from the Potlatch mailing list archive at Nabble.com.
___
Potlatch-dev mailing
Hi!
Just to explain the situation: I am running an (unmodified) P2, but I have
built a custom map_feature file from scratch. Of course it would be good to
translate both.
So it would be desirable to have seperate language resources for strings
hardcoded in P2 and for strings from customizable
I have done this a year ago. Don't have the source anymore, but I did it with
an independent java application and rsync. Generate_tiles just produced a
tile list to process. That way I could use both cores in parallel. One hint
that might be useful: Run pngnq on the tiles before you upload them.
Hi!
It sounds like the select statements for z1-10 take longer than the tirex
timeout, so the renderer is killed and the postgres process is orphaned.
Those zoomlevels are the worst if the render layers and DB are not optimized
as a lot of data is gathered from the huge areas. Under bad
Stephen Westrip wrote:
I will increase the tirex timeout and see whether the tiles ever render,
although I do know already that tiles render properly as zoom level 12 and
above are successful.
That's what I expect, but sometimes it's a problem in a layer that is only
rendered for the
Steve Bennett-3 wrote:
Anyway think the element names could be refined slightly. Are there
any other tags that work this way, apart from the lifecycle ones? Do
different tags have different lifecycles (I seem to recall that
railways have more states). Should I just hard-code it all?
I
Hi!
Steve Bennett-3 wrote:
Btw, do you have any idea why there are 60,000 route=hiking *nodes* in
taginfo?
I have had a look at some of them. It appears that a few people are actually
putting that tag to every single node of the hiking route. :-)
I did not find anything meaningful
Hi!
I have played around with this feature and it works really nicely on nodes
and ways.
Would it be possible to do a similar thing for relations?
There are multiple taggings for hiking route relations, some are route=foot,
some are route=hiking. At the moment, I need to add two relation
Steve Bennett-3 wrote:
In this particular case, what do you want to happen though: add two
relations (one route=foot, one route=hiking)? Or interpret both tags,
but always save as route=foot?
I would like one map_features entry that shows only one drop-down box with a
combined list of
Hi!
Wouldn't a parameter --dense or --sparse that controls such parameters be
much more straightforward?
I am pretty positive users will happily give that indication if it improves
performance.
bye
Nop
--
View this message in context:
Hi!
We recently added a 128GB SSD drive to a render server and maybe some of you
might find the results useful for their own hardware considerations.
The server is a quadcore, debian linux with a tirex - mapnik - PostGIS
rendering chain. Before the upgrade, tiles and database were located on a
Hi!
Frederik Ramm wrote:
I had that once and found it was a problem where somehow the master
could not message back to tirex-batch that it had completed a tile. I
thought I had fixed the problem in r23241.
I believe it mostly happened in a scenario where you would input one
batch
Problem solved: Ther content type was turned around: plain/text instead of
text/plain.
So much for blindly taking over lines from examples on the web. :-)
bye
Nop
--
View this message in context:
Nick Whitelegg-2 wrote:
1. Complete guess but are you outputting anything - including spaces -
before the header() line? (in which case the header() line will have no
effect).
2. Have you tried using XML or JSON as the data format instead?
1. Yes, I checked for that trip wire. The
Hi!
Nick Whitelegg-2 wrote:
You don't really need a text file, you can generate XML directly from the
PHP and have OpenLayers read that. One good format to use for the XML is
GeoRSS as OpenLayers already has a parser for that.
The OpenLayers parser uses GeoRSS with Atom rather than RSS
Hi!
Stephan Knauss wrote:
Firebug shows a return code 200, but displays it in red.
does it state some response text when you open the detail view?
No, the details view is completely empty - but when I select Open in new
tab for the request, I get the correct POI list from the server.
Hi!
Yes, the description was a little bit vague. I had hoped that maybe this was
a known problem. Anyway, here's the details.
The JS is pretty much straight from the examples:
var pois = new OpenLayers.Layer.Vector(POIs, {
strategies: [new
Hi!
I am trying to establish an overlay with clickable POIs over a map. I
started from the OL dynamic POI example [1].
What I have so far:
- a Vector layer with BBOX strategy, that shows the POIs when handed a text
file
- a server side PHP script that creates a short text file from the
Hi!
Andy Allan wrote:
The vectors.xml, imagery.xml and stylesheets.xml files are all
requested at initialisation, so you could (if you had a need) provide
custom per-user versions of any of them.
The myGPX dialog (which is practically unfindable now) talks only to
the same OSM server
Richard Fairhurst wrote:
I've added it in r24590 (and uploaded a new swf to geowiki.com should
you want to play with it for now). You can pass in the URL using the
gpx_url parameter.
Thanks. I'd really like to play with it, but I don't understand where that
parameter is and how to call
Hi!
Tom Hughes-3 wrote:
Looking in I18N for it is fairly high on my to do list - it shouldn't be
hard to do as Flex has I18N support built in.
That was for messages builtin to Potlatch though - I hadn't looked at
the question of the presets, which come from an external XML file. I do
Andy Allan wrote:
You can set 1 or more GPX files to be loaded at startup using the
vectors.xml config file, if that's what you're meaning:
Not really. That file obviously is global for all users.
My use case is:
- A user has a GPX track on his local drive that he wants to use for
Hi!
I just tried to delete a superfluous way in P2. But when I try to upload
that change, I get a persistent
Error 2032.
bye
Nop
--
View this message in context:
http://gis.638310.n2.nabble.com/Unable-to-delete-a-way-tp5724094p5724094.html
Sent from the Potlatch mailing list archive
Hi!
I extended my map_features.xml with some new tags and icons. For some
strange reason I cannot see my update, when I call P2, it still shows the
old features.
I have cleared the Flash cache and the Browser cache and there's definitely
only the new version on the server. I'm at a loss where
Richard Fairhurst wrote:
Just fixed that in r23987 (which will make its way to the various
deployments in due course). You now get a little 'x' button in the route
'icon'.
I guess there's still no way to check wich build a .swf is from.
bye
Nop
--
View this message in
Hi!
Sebastian Klein wrote:
You are probable interested in this version:
http://josm.openstreetmap.de/export/2813/trunk/data/defaultpresets.xml
Yes, exactly.
thanks
Nop
--
View this message in context:
Hi!
After switching to a recent version of JOSM, I noted that the presets have
changed to a different structure. I liked some of the old presets better,
e.g. the ability to set a track/tracktype with one click - which has become
much more complicated with the new presets.
Are the previous
81 matches
Mail list logo
