Ticket #9998 (closed: fixed)
Modify SaveSPE to write canonical ASCII files compartible with old data readers
Reported by: | Alex Buts | Owned by: | Alex Buts |
---|---|---|---|
Priority: | major | Milestone: | Release 3.3 |
Component: | Direct Inelastic | Keywords: | |
Cc: | r.coldea@… | Blocked By: | |
Blocking: | Tester: | Dan Nixon |
Description
Current SaveSPE writes symbolic string "NaN" if data have such values. Canonical SaveSPE writes -1e+30 instead.
The column width should also be checked and modified according to old SPE file agreement.
As nobody objected against this change, I will modify the algorithm accordingly.
Change History
comment:2 Changed 6 years ago by Alex Buts
refs #9998 Should fix it though the tests still needed.
Changeset: 0bedc84f2043d64234af87ce95fa2f5ae26caee0
comment:3 Changed 6 years ago by Alex Buts
refs #9998 missing format specification
Changeset: f512952043ddb60aaa237e83240d45bf6739c6da
comment:4 Changed 6 years ago by Nick Draper
- Status changed from new to assigned
- Component changed from Framework to Direct Inelastic
comment:5 Changed 6 years ago by Alex Buts
- Status changed from assigned to inprogress
refs #9998 Doxygen error
Changeset: 1b38089f1b243cfe18d10ca41543ec8e165f154a
comment:6 Changed 6 years ago by Alex Buts
- Status changed from inprogress to verify
- Resolution set to fixed
The data (rather bulky) to test this ticket can be found at ftp://ftp.nd.rl.ac.uk/scratch/abuts/Ticket_9998/
Load nxspe file sitting there (it contains NaN-s) and try to SaveSPE -- you will find 1.#QOE+00 values written to spe (e.g around row N4641 -- the detector table there is rather extensive).
Apply the patch and the repeat the procedure: -1E+30-s would appear instead.
I have troubles identifying standard SPE format for spe files. -- existing Mantid produces something similar to Libisis output but I have troubles starting Libisis now, so I produced sample SPE file using Mslice -- this seems works fine but produces a bit different result. I have changed the format of the Mantid SPE file to fit this format (slightly different from Libisis). The example of Mslice output is provided in the test folder (MAP_mslice_produced.zip). Tester can verify that the new format corresponds the format, written by Mslice (in signals and errors tables -- detector's azimuthal angle is not written by Mslice, so it contains 0-s there)
Mantid happily reads files written in this changed spe format, so I am leaving it up to users to complain if anything is wrong with it.
Changes are both in Mantid and System Tests branch as indirect inelastic system test verify exact correspondence of the spe file format rather then the derived values.
comment:7 Changed 6 years ago by Alex Buts
refs #9998 Typo in doxygen definition
Changeset: 2ac865f356aeb8e85476efd16e110108ec915e89
comment:8 Changed 6 years ago by Alex Buts
refs #9998 Changes to IndirectInelastic system test
which explicitly checks spe file format. Verified that the data in the file are physically the same.
Changeset: bf348801eab7eaf8335b16d80ff6a735d039180f
comment:9 Changed 6 years ago by Alex Buts
I have spoken with Indirect inelastic people in ISIS and the complain was about 3 significant digits in the data. The changes make it 4 significant digits so this is apparently what was requested.
They are also aware of where to come if for some mysterious reason their software stops working.
comment:10 Changed 6 years ago by Dan Nixon
- Status changed from verify to verifying
- Tester set to Dan Nixon
comment:11 Changed 6 years ago by Dan Nixon
- Status changed from verifying to closed
Merge remote-tracking branch 'origin/bugfix/9998_SaveSPE_format'
Full changeset: dd98da15d1a6b1aaf212d82534f0e1bc48343983
comment:12 Changed 6 years ago by Dan Nixon
Merge remote-tracking branch 'origin/bugfix/9998_SaveSPE_format'
Full changeset: 3695270b5b5d4981f4bc15a80d994142b1e00139
comment:13 Changed 5 years ago by Stuart Campbell
This ticket has been transferred to github issue 10840