Closed tehnrd closed 3 years ago
Hm I guess there might be id's I can check instead of just going row by row
On Thu, Aug 13, 2020, 1:32 AM Jason Venable notifications@github.com wrote:
First, I don't necessarily think this is an issue with this library, but it would be great if this library would throw an exception in the situation in which there are more dbf entries than shape features. Yes, this is a bad/invalid shapefile but we don't have control over the source https://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/1270.0.55.003July%202013?OpenDocument .
Attached is a very small shapefile to reproduce the issue. badshp.zip https://github.com/calvinmetcalf/shapefile-js/files/5067255/badshp.zip
If you take that file and drop it here, http://leaflet.calvinmetcalf.com/#6/-33.596/138.076, you will see it cover Kangaroo Island and this is the correct location. If you click on the shape to display the info window you will see it has the incorrect data and the name is "No usual address" which is coming from the first row in the dbf file. [image: Screen Shot 2020-08-12 at 10 26 49 PM] https://user-images.githubusercontent.com/1007085/90097530-fccd6a80-dcea-11ea-90fc-886215d24467.png
Interestingly, if I load this in Qgis and select the feature, it shows the correct information. [image: Screen Shot 2020-08-12 at 10 28 16 PM] https://user-images.githubusercontent.com/1007085/90097634-41590600-dceb-11ea-8bd2-4fefc2bf87fa.png
If this was able to work like Qgis, great, but at the very least it should probably throw an exception as this can result in very correct data outputs.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/calvinmetcalf/shapefile-js/issues/134, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAITRHYXISJI4E44RV7K74DSAN3FZANCNFSM4P52J2YQ .
Row by row should work, as that is the spec, but looks like some files don't follow the spec. I may try to get a PR for this if I can squeeze in the time but at least wanted to get the issue logged in case anyone else came across the issue or felt ambitious about doing a PR.
ok cool, If I have some time at some point I can take a look
@yakomaru @calvinmetcalf thanks for merging this fix so quickly. It definitely helps us with some data corruption issues we were experiencing.
First, I don't necessarily think this is an issue with this library, but it would be great if this library would throw an exception in the situation in which there are more dbf entries than shape features. Yes, this is a bad/invalid shapefile but we don't have control over the source. There might actually be shape features but they have no coordinates/points.
Attached is a very small shapefile to reproduce the issue. badshp.zip
If you take that file and drop it here, http://leaflet.calvinmetcalf.com/#6/-33.596/138.076, you will see it cover Kangaroo Island and this is the correct location. If you click on the shape to display the info window you will see it has the incorrect data and the name is "No usual address" which is coming from the first row in the dbf file.
Interestingly, if I load this in Qgis and select the feature, it shows the correct information.
If this library was able to work like Qgis, great, but at the very least it should probably throw an exception as this can result in very corrupt data outputs.