Hi Ticker,
my concern was not about the number of additional bytes on the disk, but the
size limits in IMG format. Anyway, since nobody else commented we probably only
find out when someone hits that limit, so I've committed the patch (first v5,
than changes from v6, sorry for that) with
Version mkgmap-r4599 was committed by gerd on Sun, 24 Jan 2021
improve handling of options when creating the overview map (now really with
overview-v6.patch by Ticker Berkin)
(last commit was overview-v5.patch)
http://www.mkgmap.org.uk/websvn/revision.php?repname=mkgmap=4599
Version mkgmap-r4598 was committed by gerd on Sun, 24 Jan 2021
improve handling of options when creating the overview map (overview-v6.patch
by Ticker Berkin)
- apply --order-by-decreasing-area to overview map (increases the overview map
a bit)
- don't create empty NOD or NET files
Hi Gerd
Given that the overview map is for use on computers with 100s of G of
disk space and the main map will be a G or so, can an extra few 10K or
so in the overview map really be a problem for anyone?
Building British-and-ireland, with default style and just --gmapsupp
(ie no index or
Hi Gerd,
In OSM/Mapnik the park was rendered green so I (wrongly) assumed that the
OSM-data must be OK.
Should have checked in JOSM, sorry...
I am glad to hear you could (re)use the test case and fix a related bug so
thankfully it wasn't a complete waste of time :)
Thanks for the fix!
Ranko
Hi Ticker,
outch, sorry! It seems I created my patch without any testing :(
I'm still not happy with the handling of the --order-by-decreasing-area option.
I wonder why nobody else comments on this. Either nobody cares about the size
of the overview map or nobody else tried any of the patches.
Hi Gerd
I hadn't rebuilt after taking your patch. You had removed a declaration
of bounds, exposing a different one.
Patch attached.
Ticker
On Sat, 2021-01-23 at 10:27 +, Gerd Petermann wrote:
> Hi Ticker,
>
> I used mkgmap r4597 patched with overview-v5.patch on the output of
>
Hi Ticker,
I used mkgmap r4597 patched with overview-v5.patch on the output of splitter
with Rankos split command:
E:\rj-mp-split>java -jar d:\mkgmap\dist\mkgmap.jar 8014.osm.pbf
Time started: Sat Jan 23 11:23:51 CET 2021
Number of MapFailedExceptions: 0
Exception in thread "main"
Hi Gerd
I've just tried this and don't get a problem (with or without -ea)
I used Ranko's osm.pbf, splitter-r595 and my development mkgmap which
is trunk + overview patch + couple of trivial changes.
Where were you getting the exception?
Ticker
On Sat, 2021-01-23 at 07:18 +, Gerd
Hi Ticker,
seems you are right regarding the 0x4a polygons. The (unpatched) code doesn't
work as I expected. I must confess that I don't know for sure how it is
expected to work, though. We have code in the polish reader to use 0x4a
polygons "as is" and that seems to work, but it is meant for
Hi Ranko,
I've found a bug in mkgmap that caused this. Fixed with r4597, so thanks for
reporting this :)
I've also fixed the wrong multipolygon in OSM.
ciao,
Gerd
Von: mkgmap-dev im Auftrag von Gerd
Petermann
Gesendet: Samstag, 23. Januar 2021 07:55
11 matches
Mail list logo
