Opened 9 years ago

Closed 9 years ago

Last modified 7 years ago

#4500 closed defect (fixed)

ogr2ogr: Segmentation fault when using both: -select and -where

Reported by: peifer Owned by: warmerdam
Priority: normal Milestone:
Component: OGR_SF Version: 1.8.0
Severity: normal Keywords:
Cc:

Description

"-select SITECODE" and "-where fid=1" work as expected when used alone, but ogr2ogr segfaults when both options are used in combination. See below. This looks like a bug to me.

> ogrinfo -al -so my.shp
INFO: Open of `my.shp'
      using driver `ESRI Shapefile' successful.

Layer name: my
Geometry: Polygon
Feature Count: 10
Extent: (16.175822, 46.993819) - (17.063785, 47.999705)
Layer SRS WKT:
GEOGCS["GCS_WGS_1984",
    DATUM["WGS_1984",
        SPHEROID["WGS_84",6378137,298.257223563]],
    PRIMEM["Greenwich",0],
    UNIT["Degree",0.017453292519943295]]
SITECODE: String (9.0)
SITENAME: String (240.0)
RELEASE_DA: Date (10.0)
MS: String (2.0)
SITETYPE: String (1.0)
POINT_X: Real (19.11)
POINT_Y: Real (19.11)
Shape_Leng: Real (19.11)
Shape_Area: Real (19.11)

> ogr2ogr -f kml /dev/null my.shp -where fid=1
> ogr2ogr -f kml /dev/null my.shp -select SITECODE
> ogr2ogr -f kml /dev/null my.shp -select SITECODE -where fid=1
Segmentation fault

Attachments (1)

my1.zip (72.2 KB) - added by peifer 9 years ago.

Download all attachments as: .zip

Change History (7)

comment:1 Changed 9 years ago by peifer

I forgot to mention that -select FID,SITECODE -where fid=1 -dialect ogrsql doesn't help much, as I end up with:

Field 'FID' not found in source layer.

comment:2 Changed 9 years ago by Even Rouault

Could you attach the test my.shp ? I tried to reproduce with one of my shapefiles, but didn't observe anything wrong when running both -select and -where fid=1

(The -select FID,SITECODE is a completely unrelated topic. FID isn't a regular column, so -select cannot find it. FID is only recognized as a special column when using -sql or -where options)

comment:3 Changed 9 years ago by peifer

Resolution: worksforme
Status: newclosed

Hmm. The testfile my.shp is gone and I can't reproduce the segfault with other shapefiles that I have around. I will close the ticket now. I will re-open it should I manage to re-produce the error.

comment:4 Changed 9 years ago by peifer

Resolution: worksforme
Status: closedreopened

I managed to re-produce the segfault. Looks to me like some encoding issue.

$ ogr2ogr -f kml /dev/null my1.shp -select SITECODE -where fid=0
Shape: DBF Codepage = LDID/87 for my1.shp
Shape: Treating as encoding 'ISO-8859-1'.
OGR: OGROpen(my1.shp/0x213f0e0) succeeded as ESRI Shapefile.
KML: Attempt to create: /dev/null
OGRCT: Source: +proj=laea +lat_0=52 +lon_0=10 +x_0=4321000 +y_0=3210000 +ellps=GRS80 +units=m +no_defs
OGRCT: Target: +proj=longlat +ellps=WGS84 +towgs84=0,0,0,0,0,0,0 +no_defs
Segmentation fault

Changed 9 years ago by peifer

Attachment: my1.zip added

comment:5 Changed 9 years ago by Even Rouault

Component: UtilitiesOGR_SF
Milestone: 1.8.2
Resolution: fixed
Status: reopenedclosed
Version: svn-trunk1.8.0

r23939 /trunk/ (2 files in 2 dirs): Fix segfault when running 'ogr2ogr target src -select AFIELD -where FID=xxx' when the source layer has more than 5 fields (#4500)

r23940 /branches/1.9/gdal/ogr/ogrfeaturequery.cpp: Fix segfault when running 'ogr2ogr target src -select AFIELD -where FID=xxx' when the source layer has more than 5 fields (#4500)

r23941 /branches/1.8/gdal/ogr/ogrfeaturequery.cpp: Fix segfault when running 'ogr2ogr target src -select AFIELD -where FID=xxx' when the source layer has more than 5 fields (#4500)

comment:6 Changed 7 years ago by Even Rouault

Milestone: 1.8.2

Milestone 1.8.2 deleted

Note: See TracTickets for help on using tickets.