<div dir="auto">For the rest of us, the OSM Relation Analyser is pretty good at showing any gaps in relations: <a href="http://ra.osmsurround.org/analyzeRelation?relationId=21895&noCache=true&_noCache=on">http://ra.osmsurround.org/analyzeRelation?relationId=21895&noCache=true&_noCache=on</a><div dir="auto"><br></div><div dir="auto"><br></div><div dir="auto">Regards,</div><div dir="auto">_Paul_</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Fri, 14 Jan 2022, 18:17 Dave F via Talk-GB, <<a href="mailto:talk-gb@openstreetmap.org">talk-gb@openstreetmap.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">I've tweaked it a bit further using this XML/XSLT/XPath parser to search <br>
for gaps & spurious spurs:<br>
<br>
Download the route relation as OSM/XML data<br>
curl -o NCN_17.osm -g <br>
<a href="https://overpass-api.de/api/interpreter?data=rel(id:21895);(._" rel="noreferrer noreferrer" target="_blank">https://overpass-api.de/api/interpreter?data=rel(id:21895);(._</a>;>;);out;<br>
<br>
Parse the data & output the results to file:<br>
java -jar C:\<Path to program>\SaxonHE9-7-0-2J\saxon9he.jar <br>
-s:NCN_17.osm -xsl:Route_Rel_Check.xsl -o:Route_Rel_Check_Upload.osm<br>
<br>
The XSL routine that performs the parsing, Basically it takes the first <br>
node of each way & compares it with the end node of all the other <br>
ways.If it doesn't find a match it outputs the node's id.<br>
The list of nodes starting with ' [not(@ref' are genuine end nodes to be <br>
ignored, such as the start/end nodes & where ways meet roundabouts <br>
mapped as a singular, circular way:<br>
<br>
<xsl:stylesheet xmlns:xsl="<a href="http://www.w3.org/1999/XSL/Transform" rel="noreferrer noreferrer" target="_blank">http://www.w3.org/1999/XSL/Transform</a>" <br>
version="2.0"><br>
<xsl:output omit-xml-declaration="yes" indent="yes" /><br>
<xsl:strip-space elements="*" /><br>
<xsl:template match="osm"><br>
<xsl:for-each-group select="way/nd[position() = (1, <br>
last())]" group-by="@ref"><br>
<xsl:sequence select=".[not(current-group()[2])] <br>
[not(@ref ='26908253' or @ref='269810730' or @ref='30394554' or <br>
@ref='33118766' or @ref='30394555')]"/><br>
</xsl:for-each-group><br>
</xsl:template><br>
</xsl:stylesheet><br>
<br>
I've find this routine very useful to maintain cycle route relations in <br>
my area<br>
Others may find this helpful all types of route relations.<br>
<br>
DaveF<br>
<br>
On 14/01/2022 15:46, Gregory Williams wrote:<br>
> I've subsequently conversed with a Sustrans Volunteer Ranger that I<br>
> know in the area and have cleared up the relation further. It should be<br>
> looking better now.<br>
><br>
> Gregory<br>
><br>
> On Fri, 2022-01-14 at 08:28 +0000, Gregory Williams wrote:<br>
>> I put some of that NCR17 mapping in. NCR17 has changed alignment over<br>
>> time, but has also had issues with bad signing.<br>
>><br>
>> I've removed some bits that I know are now wrong. I suspect that the<br>
>> portion in west Maidstone around Poplar Grove to Queen's Road is also<br>
>> wrong, but I think that there have been odd signs here that<br>
>> incorrectly<br>
>> suggest that it is NCR17. I seem to recall Maidstone having some<br>
>> signs<br>
>> that confused "17" and "(17)". I.e. is *on* vs. *leads to* NCR17.<br>
>><br>
>> I've copied Maidstone Cycle Forum, as they should be know their patch<br>
>> better than me! Perhaps they may be able to advise about the Medway<br>
>> portion too?<br>
>><br>
>> Gregory<br>
>><br>
>> On Thu, 2022-01-13 at 13:28 +0000, Chris Hodges wrote:<br>
>>> I'm not local so can't comment on that specific route, but it is<br>
>>> plausible for a few reasons:<br>
>>><br>
>>> - Some routes really do start and stop in silly places because land<br>
>>> access negotiations have stalled. Sometimes in multiple places.<br>
>>> NCN33<br>
>>> in Somerset is like that, with work finally beginning on connecting<br>
>>> it<br>
>>> to NCN26 and Clevedon after over 10 years of proposals and talk.<br>
>>><br>
>>> - Sustrans have been removing branding on (and support for) routes<br>
>>> that<br>
>>> don't fit their rather arbitrary standards. This can leave gaps -<br>
>>> or<br>
>>> sections that have NCN numbers but aren't put of the NCN (according<br>
>>> to<br>
>>> their own map <a href="https://explore.osmaps.com/ncn" rel="noreferrer noreferrer" target="_blank">https://explore.osmaps.com/ncn</a> hosted by OS which I<br>
>>> used<br>
>>> to use as part of my route planning but don't trust)<br>
>>><br>
>>> - That same map doesn't always reflect reality as it ever happened,<br>
>>> nor<br>
>>> does it match the signs, which also don't match reality. We can and<br>
>>> do<br>
>>> make a better job of it in places.<br>
>>><br>
>>> For example NCN 45 into Bridgnorth from the north is correctly<br>
>>> mapped<br>
>>> on OSM (following the signs with a little interpolation where<br>
>>> they're<br>
>>> missing)<br>
>>> <a href="https://www.openstreetmap.org/search?query=bridgnorth#map=14/52.5333/-2.4153&layers=C" rel="noreferrer noreferrer" target="_blank">https://www.openstreetmap.org/search?query=bridgnorth#map=14/52.5333/-2.4153&layers=C</a><br>
>>> <br>
>>> but some odd sections that don't join up with anything are also<br>
>>> mapped.<br>
>>> They match Sustrans's own mapping, except that OSM is more sensible<br>
>>> and<br>
>>> doesn't show a route going across a school playing field, near but<br>
>>> not<br>
>>> following a public footpath. Some signs exist on those disconnected<br>
>>> sections. I think they had NCN route numbers on them when I was<br>
>>> there,<br>
>>> but I've also seen route numbers covered up<br>
>>><br>
>>><br>
>>> A glance at Sustrans's map of the Maidstone area, without detailed<br>
>>> inspection, suggests that much of what looks wrong could be right<br>
>>> or<br>
>>> at<br>
>>> least official, but not all the disconnected sections match that<br>
>>> map.<br>
>>> But I wouldn't trust it to be right, as well as the obvious<br>
>>> licensing<br>
>>> block on the data, i.e. even if it was allowed, copying the<br>
>>> "official"<br>
>>> data would increase the errors<br>
>>><br>
>>><br>
>>> Chris<br>
>>><br>
>>><br>
>>> On 13/01/2022 11:24, Andy Townsend wrote:<br>
>>>> This seems a bit odd:<br>
>>>><br>
>>>> <a href="https://www.openstreetmap.org/relation/21895" rel="noreferrer noreferrer" target="_blank">https://www.openstreetmap.org/relation/21895</a><br>
>>>><br>
>>>> shows a couple of disconnected sections around Maidstone and an<br>
>>>> odd<br>
>>>> spur towards Chatham.<br>
>>>><br>
>>>> <a href="https://osm.mapki.com/history/relation/21895" rel="noreferrer noreferrer" target="_blank">https://osm.mapki.com/history/relation/21895</a><br>
>>>><br>
>>>> suggests that at least one of the Maidstone "extras" has been<br>
>>>> there<br>
>>>> since 2008, suggesting that it's not a recent faux pas.<br>
>>>><br>
>>>> I suspect that given the number of avid cyclists on this list<br>
>>>> that<br>
>>>> someone will be immediately able to say "actually that's wrong"<br>
>>>> or<br>
>>>> "actually, that's correct because..."<br>
>>>><br>
>>>> Best Regards,<br>
>>>><br>
>>>> Andy<br>
>>>><br>
>>>><br>
>>>><br>
>>>> _______________________________________________<br>
>>>> Talk-GB mailing list<br>
>>>> <a href="mailto:Talk-GB@openstreetmap.org" target="_blank" rel="noreferrer">Talk-GB@openstreetmap.org</a><br>
>>>> <a href="https://lists.openstreetmap.org/listinfo/talk-gb" rel="noreferrer noreferrer" target="_blank">https://lists.openstreetmap.org/listinfo/talk-gb</a><br>
>>><br>
>>><br>
>>> _______________________________________________<br>
>>> Talk-GB mailing list<br>
>>> <a href="mailto:Talk-GB@openstreetmap.org" target="_blank" rel="noreferrer">Talk-GB@openstreetmap.org</a><br>
>>> <a href="https://lists.openstreetmap.org/listinfo/talk-gb" rel="noreferrer noreferrer" target="_blank">https://lists.openstreetmap.org/listinfo/talk-gb</a><br>
> _______________________________________________<br>
> Talk-GB mailing list<br>
> <a href="mailto:Talk-GB@openstreetmap.org" target="_blank" rel="noreferrer">Talk-GB@openstreetmap.org</a><br>
> <a href="https://lists.openstreetmap.org/listinfo/talk-gb" rel="noreferrer noreferrer" target="_blank">https://lists.openstreetmap.org/listinfo/talk-gb</a><br>
<br>
<br>
_______________________________________________<br>
Talk-GB mailing list<br>
<a href="mailto:Talk-GB@openstreetmap.org" target="_blank" rel="noreferrer">Talk-GB@openstreetmap.org</a><br>
<a href="https://lists.openstreetmap.org/listinfo/talk-gb" rel="noreferrer noreferrer" target="_blank">https://lists.openstreetmap.org/listinfo/talk-gb</a><br>
</blockquote></div>