Bug report #17681
Negative values for age in expressions
| Status: | Closed | ||
|---|---|---|---|
| Priority: | High | ||
| Assignee: |  Sandro Santilli | ||
| Category: | Expressions | ||
| Affected QGIS version: | 2.18.14 | Regression?: | No |
| Operating System: | Mac OS High Sierra | Easy fix?: | No |
| Pull Request or Patch supplied: | Yes | Resolution: | fixed/implemented |
| Crashes QGIS or corrupts data: | No | Copied to github as #: | 25578 |
Description
In calculating the difference between two dates from a long list of starting and ending dates (from medieval period to now), some of the outcomes produce negative values:
DATEfrom DATEto AGE (in days)
1380-01-01 1515-12-31 -37
1100-01-01 1225-12-31 -3689
580-01-01 665-12-31 -18299
1612-01-01 1697-12-31 -18299
Whereas most other dates do provide correct ages (>90%):
1001-01-01 1199-12-31 23339
1814-01-01 2016-12-31 24434
The expression used here is: day(age("DATEto","DATEfrom")). Is this a bug?
Associated revisions
Allow age expression to represent time intervals above 130 years (#5914)
Uses a 64bit integer instead of 32bit one for seconds,
meaning about 584 billions of years rather than 136...
Closes #17681
Includes test
History
#1
Updated by Andreas Neumann about 7 years ago
- Priority changed from Normal to High
Seems to be a serious issue to me.
Calculations that provide wrong results are always bad.
#2
Updated by Sandro Santilli about 7 years ago
- Assignee set to Sandro Santilli
- Status changed from Open to In Progress
I'm looking at this, starting from the existing unit test tests/src/core/testqgsexpression.cpp (run via output/bin/qgis_expressiontest).
The test in there puts later day first, but I'm not sure yet about the test code semantic:
day(age(to_date('2004-03-22'),to_date('2004-03-12')))
#3
Updated by Sandro Santilli about 7 years ago
I'm guessing it's an overflow of `int` type used to compute seconds between the dates.
The first age you report as working has 4291574400, max integer is 4294967296.
The first non-working age has 6279724800 seconds, so turns to negative.
I'm working on adding an automated test for this.
#4
Updated by Sandro Santilli about 7 years ago
- Pull Request or Patch supplied changed from No to Yes
Pull request ready for test: https://github.com/qgis/QGIS/pull/5914
Andreas: can you give that a try ?
#5
Updated by Sandro Santilli about 7 years ago
- Status changed from In Progress to Closed
- % Done changed from 0 to 100
Applied in changeset qgis|336995dc30f4409520b65a0007e2cea6966f7599.
#6
Updated by Giovanni Manghi almost 7 years ago
- Resolution set to fixed/implemented