Opened 3 years ago
Last modified 3 years ago
#20485 new defect
Inconsistent handling of fixme tags
Reported by: | mdk | Owned by: | team |
---|---|---|---|
Priority: | normal | Milestone: | |
Component: | Core validator | Version: | |
Keywords: | template_report fixme node | Cc: |
Description
What steps will reproduce the problem?
- load fixme.osm
- run validator
What is the expected result?
5 "Other/fixme" validator hints.
What happens instead?
I got only 3 for the fixme/FIXME tags on ways and relation, but none for the fixme/FIXME tags on nodes.
Please provide any additional information below. Attach a screenshot if possible.
The hints for fixme tags are very helpful for finding them in a downloaded array. As far as I remember this is a regression.
While I created the example file, I saw also, that the standard mappaint style is drawing an "F" for both "fixme" and "FIXME" on nodes, but the tag list for the selected objects only show "Annotation/Fixme" entry for "fixme", but not for "FIXME" (nodes and ways).
URL:https://josm.openstreetmap.de/svn/trunk Repository:UUID: 0c6e7542-c601-0410-84e7-c038aed88b3b Last:Changed Date: 2021-02-12 09:50:35 +0100 (Fri, 12 Feb 2021) Build-Date:2021-02-13 02:30:51 Revision:17491 Relative:URL: ^/trunk Identification: JOSM/1.5 (17491 en_GB) Windows 10 64-Bit OS Build number: Windows 10 Pro 2004 (19041) Memory Usage: 1566 MB / 3616 MB (510 MB allocated, but free) Java version: 1.8.0_201-b09, Oracle Corporation, Java HotSpot(TM) 64-Bit Server VM Look and Feel: com.sun.java.swing.plaf.windows.WindowsLookAndFeel Screen: \Display0 1920×1200 (scaling 1.00×1.00) \Display1 1920×1200 (scaling 1.00×1.00) Maximum Screen Size: 1920×1200 Best cursor sizes: 16×16→32×32, 32×32→32×32 Dataset consistency test: No problems found Plugins: + ColumbusCSV (35640) + OpeningHoursEditor (35640) + RoadSigns (35640) + apache-commons (35524) + apache-http (35589) + buildings_tools (35669) + contourmerge (v0.1.6) + imagery-xml-bounds (35640) + imagery_offset_db (35640) + jna (35662) + pt_assistant (2.1.10-80-g7d9bba3) + reverter (35688) + terracer (35640) + utilsplugin2 (35691) Map paint styles: + https://josm.openstreetmap.de/josmfile?page=Styles/Lane_and_Road_Attributes&zip=1 + https://josm.openstreetmap.de/josmfile?page=Styles/Maxspeed&zip=1 - https://josm.openstreetmap.de/josmfile?page=Styles/PublicTransport&zip=1
Attachments (2)
Change History (11)
by , 3 years ago
comment:1 by , 3 years ago
Component: | Core → Core validator |
---|---|
Keywords: | fixme node added |
comment:3 by , 3 years ago
I don't fully understand what your goal is. I think it makes not much sense to produce a validator warning for a fixme, but I assume it is done because it difficult to render it. This problem doesn't exist with the node.
If I want to find objects with a fixme I use search.
comment:4 by , 3 years ago
An informal warning about fixme/FIXME=*
is what this rule is about. Searching for all nodes with this tag will only return a list of all objects but not each by its own. I can probably take a look at this next week.
comment:5 by , 3 years ago
Right. The goal is that all fixme tags are handled the same same way - no special handling for nodes with only one (the fixme) tag.
comment:6 by , 3 years ago
The question is, if we want to hide duplicate warnings and which one exactly. E.g. we have warnings:
- about nodes:
Unconnected nodes without physical tags
>Has tag containing 'fixme'
Waterway ends without a connection to another waterway or the direction of the waterway is wrong.
(only inside download area)- maybe:
Way end node near other (high/rail/water)way
- about ways:
Untagged ways (commented)
- plus maybe:
Unnamed ways
Have a look at attached sample session's files.
I tend to hide the "informal warning" in favor of the higher level warning, at least, for the first warning for each object type.
comment:7 by , 3 years ago
I am actually working on bad mapped areas in Madagascar. Typically you found the pattern of a waterway stub where the "source" node is tagged with fixme=continue. In the example is only the other case, where the "drain" node is tagged (and add an additional "waterway ends without a connection..." warning).
I also find highways, where the direction doesn't matter.
Both, waterways and highways normally has actually no name, so I would exclude this case.
Also the examples are not so appropriate, because they have fixme tags on both, the way and and a node. "fixme" on ways are producing always (?) informal warnings. But I'm missing the informal warnings if only a node is tagged.
In general: A fixme/FIXME tag is explicitly set by a mapper, so I see no reason for hiding any of them.
comment:8 by , 3 years ago
Owner: | changed from | to
---|---|
Status: | new → assigned |
With the oneway example I had #18364 in mind. The waterway warning is not run on drain
and for the first way node we might have an own warning about missing waterway=source
. But that are future thoughts.
I still feel like hiding in favor of Unconnected nodes without physical tags
> Has tag containing 'fixme'
and Untagged ways (commented)
.
comment:9 by , 3 years ago
Owner: | changed from | to
---|---|
Status: | assigned → new |
I had a deeper look and I am not able to fix it as the tests are written in java in TagChecker.java, sorry.
Seems to me that there needs to be one more tag.
fixme=*
+highway=bus_stop
produces the expected info warning. For ways and relations we have higher level warnings if there is onlyfixme/FIXME=*
but not for nodes.