I am trying to parse csv file and visualize it as geotiff using gdal_grid but I keep getting this error:
C:\work\testgdal1>gdal_grid -ot Float32 -l test test.vrt dona.tif
ERROR 1: Failed to open datasource `test.csv'.
Unable to open input datasource "test.vrt".
Failed to open datasource `test.csv'.
Both test.csv and test.vrt are in the same folder.
What can possibly go wrong and how to fix it?
Environment:
Windows 7 x64
osgeo4w as of 2 May 2011
test.vrt
test.csv
test
WGS84
wkbPoint
2 lines of test.csv
200010 207 020311 40658.5 406593 52 344927.31 7100203.50 -26.2078720 127.4491855 345060.64 7100369.14 26.4 650.3 628.0 55471.293 20.168 55648.817 55637.523 -146.062
200010 207 020311 40658.6 406594 52 344932.31 7100203.50 -26.2078726 127.4492355 345065.64 7100369.14 27.2 650.3 627.2 55456.719 20.172 55648.814 55637.520 -160.629
Answer
Ah yes, I feel slightly to blame for this... I've just re-read the page on OGR's handling of CSV and it says that it only accepts commas, semi-colons, and tabs as field delimiters, but your file is fixed-width.
What you will need to do is convert each run of spaces to a comma. In Linux you can use a sed script, but as you're in Windows your options are a bit more limited. A simple way would be to load the ASCII file into a spreadsheet program, telling it to use fixed-width fields, then save it out telling it to use commas as separators. Not ideal, I grant you, but it is a technique I've used in the past.
No comments:
Post a Comment