id
int64 0
5.38k
| issuekey
stringlengths 4
16
| created
stringlengths 19
19
| title
stringlengths 5
252
| description
stringlengths 1
1.39M
| storypoint
float64 0
100
|
|---|---|---|---|---|---|
2,399 |
DM-11800
|
09/05/2017 10:59:06
|
Update meas_mosaic to new transform-based API for afw::cameraGeom
|
With the merge of DM-5922, {{meas_mosaic}} is currently broken with errors like: {code} src/mosaicfit.cc:2500:38: error: 'lsst::afw::cameraGeom::CameraTransformMap' has not been declared afw::cameraGeom::CameraTransformMap::Transforms newTr; ^ {code} Please update {{meas_mosaic}} to the new {{afw::cameraGeom}} API.
| 1 |
2,400 |
DM-11802
|
09/05/2017 11:49:55
|
bump eups version to 2.1.4
|
Update the eups version to {{2.1.3}} for an important py3 compatibility fix.
| 0.5 |
2,401 |
DM-11809
|
09/05/2017 15:49:25
|
Explicitly enable pep8-naming tests
|
Currently, {{flake8}} is installed with minimal plugins and does not include {{pep8-naming}}. This is problematic on two counts: # The developer guide explicitly specifies naming tests with a set of ignored codes. These codes are not currently checked. # If a developer happens to have {{pep8-naming}} installed their build can break because the flake8 testing in CI and in default builds is not testing what it should be. I am going to add {{pep8-naming}} eups package and enable name testing. This will also require fixing any flake8 problems in the packages that have automated flake8 testing. I do not believe an RFC is required to add this package as the developer guide already implies we are using it.
| 1 |
2,402 |
DM-11862
|
09/07/2017 17:50:11
|
writeFitsCatalogStorage calls obj.writeFits incorrectly
|
{{writeFitsCatalogStorage}} calls {{obj.writeFits(logLoc.locString(), flags=flags)}}, where {{obj}} is "the object to be written". If {{obj}} is an {{afw.Persistable}}, this is an invalid call (Persistable.writeFits doesn't take {{flags}}), so I cannot persist PhotoCalibs in jointcal: {code} dataRef.put(photoCalib, 'photoCalib') File "/home/parejkoj/lsst/lsstsw/stack/Linux64/daf_persistence/13.0-34-g674f0d6/python/lsst/daf/persistence/butlerSubset.py", line 216, in put self.butlerSubset.butler.put(obj, datasetType, self.dataId, doBackup=doBackup, **rest) File "/home/parejkoj/lsst/lsstsw/stack/Linux64/daf_persistence/13.0-34-g674f0d6/python/lsst/daf/persistence/butler.py", line 1429, in put location.getRepository().write(location, obj) File "/home/parejkoj/lsst/lsstsw/stack/Linux64/daf_persistence/13.0-34-g674f0d6/python/lsst/daf/persistence/repository.py", line 189, in write return butlerLocationStorage.write(butlerLocation, obj) File "/home/parejkoj/lsst/lsstsw/stack/Linux64/daf_persistence/13.0-34-g674f0d6/python/lsst/daf/persistence/posixStorage.py", line 264, in write writeFormatter(butlerLocation, obj) File "/home/parejkoj/lsst/lsstsw/stack/Linux64/daf_persistence/13.0-34-g674f0d6/python/lsst/daf/persistence/posixStorage.py", line 713, in writeFitsCatalogStorage obj.writeFits(logLoc.locString(), flags=flags) TypeError: writeFits(): incompatible function arguments. The following argument types are supported: 1. (self: lsst.afw.table.io.persistable.Persistable, fileName: str, mode: str='w') -> None 2. (self: lsst.afw.table.io.persistable.Persistable, manager: lsst.afw.fits._fits.MemFileManager, mode: str='w') -> None Invoked with: PhotoCalib(spatially constant with mean: 7.30334e+12 err: 2.15887e+09), '/home/parejkoj/lsst/jointcal/jointcal/tests/.test/JointcalTestCFHTMinimal/run/jointcal-results/06AL01/D3/2006-05-20/r/0/photoCalib-849375-12.fits_mcaoaru'; kwargs: flags=0 {code}
| 1 |
2,403 |
DM-11863
|
09/07/2017 19:52:14
|
Fix bug in strict monotonicity
|
There is a bug in the strict monotonicity algorithm that occurs when a faint object is close enough to a bright object that its initial SED is a poor choice, and is closer to the SED of its brighter neighbor. In this case the algorithm completely fails. To test the projection, I ran a deblend using only the symmetry operator, which resulted in the first attached image. I then ran the projection, which resulted in the 2nd image, which is not strictly monotonic. This ticket is to investigate the origin of this strange behavior.
| 1 |
2,404 |
DM-11867
|
09/08/2017 15:25:22
|
Update developer guide with flake8 testing instructions
|
Now that flake8 testing can be enabled as part of normal tests, we should document how to do that.
| 1 |
2,405 |
DM-11870
|
09/09/2017 11:06:54
|
weekly release w_2017_36 failed
|
The build / eups distrib tag / tarball completed but the pipeline died in the container "build from eups distrib" before building the jupyterlabdemo image. Presumably, the failure was because {{lsst_distrib}} can no longer be built in the docker default 10GiB root volume size. The easiest solution would be to replace the eupspkg build with a tarball install. {code:java} docker-centos-7: [ 72/111 ] afw 13.0-97-g247b2c20e ... scons: building terminated because of errors. docker-centos-7: IOError: [Errno 28] No space left on device: docker-centos-7: File "/opt/lsst/software/stack/stack/miniconda3-4.2.12-7c8e67/Linux64/scons/2.5.0.lsst2+1/lib/scons/SCons/Script/Main.py", line 1346: docker-centos-7: _exec_main(parser, values) docker-centos-7: File "/opt/lsst/software/stack/stack/miniconda3-4.2.12-7c8e67/Linux64/scons/2.5.0.lsst2+1/lib/scons/SCons/Script/Main.py", line 1309: docker-centos-7: _main(parser) docker-centos-7: File "/opt/lsst/software/stack/stack/miniconda3-4.2.12-7c8e67/Linux64/scons/2.5.0.lsst2+1/lib/scons/SCons/Script/Main.py", line 1091: docker-centos-7: nodes = _build_targets(fs, options, targets, target_top) docker-centos-7: File "/opt/lsst/software/stack/stack/miniconda3-4.2.12-7c8e67/Linux64/scons/2.5.0.lsst2+1/lib/scons/SCons/Script/Main.py", line 1283: docker-centos-7: jobs.run(postfunc = jobs_postfunc) docker-centos-7: File "/opt/lsst/software/stack/stack/miniconda3-4.2.12-7c8e67/Linux64/scons/2.5.0.lsst2+1/lib/scons/SCons/Job.py", line 113: docker-centos-7: postfunc() docker-centos-7: File "/opt/lsst/software/stack/stack/miniconda3-4.2.12-7c8e67/Linux64/scons/2.5.0.lsst2+1/lib/scons/SCons/Script/Main.py", line 1280: docker-centos-7: SCons.SConsign.write() docker-centos-7: File "/opt/lsst/software/stack/stack/miniconda3-4.2.12-7c8e67/Linux64/scons/2.5.0.lsst2+1/lib/scons/SCons/SConsign.py", line 109: docker-centos-7: syncmethod() docker-centos-7: File "/opt/lsst/software/stack/stack/miniconda3-4.2.12-7c8e67/Linux64/scons/2.5.0.lsst2+1/lib/scons/SCons/dblite.py", line 119: docker-centos-7: self._pickle_dump(self._dict, f, 1) docker-centos-7: Exception IOError: (28, 'No space left on device') in <bound method dblite.__del__ of <SCons.dblite.dblite object at 0x350e490>> ignored docker-centos-7: + exit -5 {code}
| 2 |
2,406 |
DM-11871
|
09/10/2017 08:38:04
|
afw.table column access is slow via__getitem__ (as compared to get)
|
See https://community.lsst.org/t/afwtable-performance-degradation-using-brackets-vs-explicit-get-for-fields.
| 1 |
2,407 |
DM-11873
|
09/11/2017 11:55:01
|
Optimize QA code for speed in accessing/assigning afw table columns
|
The QA scripts in {{pipe_analysis}} do a lot of column accessing & assigning of afw tables. A recent [community post|https://community.lsst.org/t/afwtable-performance-degradation-using-brackets-vs-explicit-get-for-fields/2236] highlighted the inefficient nature of accessing catalog elements with {{catalog\[fieldName\]}} vs {{catalog.get(fieldName)}} (or, even better, {{catalog.get(filedNameKey)}}). Please update the code for speed efficiency according to the findings and recommendations of [this community post|https://community.lsst.org/t/afwtable-performance-degradation-using-brackets-vs-explicit-get-for-fields/2236].
| 2 |
2,408 |
DM-11895
|
09/12/2017 11:21:21
|
Support getting data for current card of FitsChan
|
AST {{astGetFits<X>}} functions ({{astGetFitsS}}, {{astGetFitsF}}, etc.) take a name to specify which FITS card to read; if the name (a {{char *}}) is {{NULL}} then the current card is read. astshim does not yet support the ability to read the current card, and it is very useful; in particular it would be a big help for DM-10765. I propose to add this functionality as follows: - if {{name}} is "" then read the current card - {{name}} will default to "" - In the pybind11 wrapper allow name=None, which will be the default In addition, the most recent version of AST enhances {{astTestFits}} to support {{name=NULL}} for the current card. Add similar support to astshim, as above. An obvious alternative is to add an overload for each {{FitsChan.getFIts<X>}} that has no {{name}} argument. That is arguably a bit more C++-like, but in my opinion it adds too many new member functions to be worthwhile. Plus overloads make the Python interface a bit harder to understand.
| 3 |
2,409 |
DM-11902
|
09/13/2017 16:58:52
|
LSST Eigen installs are not discoverable by cmake
|
Our Eigen package has an all-too-clever eupspkg.cfg.sh that "emulates" the CMake build that comes with Eigen - and fails to install files that other CMake packages can use to find it. This was discovered while testing an external PR on ndarray that improves and cleans up its CMake build. My planned solution is to simply remove the eupspkg.sh and let Eigen's CMake build do its thing.
| 1 |
2,410 |
DM-11904
|
09/13/2017 17:17:59
|
"namespace" all newinstall.sh functions
|
Virtually all {{newinstall.sh}} functions should be namespaced to facilitate using the script as a "library" from other scripts. At present, only the most recently modified functions have been namespaced under {{n8l::}}.
| 2 |
2,411 |
DM-11909
|
09/14/2017 09:43:33
|
Test functionality of deblender task
|
DM-11329 involved creating a deblender task to run the new deblender on HSC data. This ticket is designed to run that task on a few HSC fields to test its performance before the larger task of running the deblender on a large patch of data and analyzing the results (DM-11330).
| 5 |
2,412 |
DM-11916
|
09/14/2017 12:49:22
|
Warnings "Extent2I object has no attribute getWidth" when running meas_mosaic, jointcal
|
Running {{MosaicTask}} with {{w_2017_36}} and current {{meas_mosaic}} gives many warnings like this: {code:java} root WARN: Failed to read {...}: 'lsst.afw.geom.coordinates.coordinates.Extent2I' object has no attribute 'getWidth' {code}
| 1 |
2,413 |
DM-11921
|
09/14/2017 18:53:01
|
Change match method in DIAObjectCollection to return DIAObject ids
|
The match method of DIAObject Collection currently returns the indices of the DIAObjects in the collection that are updated rather than their unique ids. This ticket is to change this return from the former to the latter.
| 2 |
2,414 |
DM-11931
|
09/15/2017 16:23:29
|
Fix jointcal exit status and doRaise handling
|
jointcal is giving these warnings since DM-4141 was implemented: {code} Unable to retrieve exit status ('Struct' object has no attribute 'exitStatus') {code} To fix this, I think I need to do the following: * add this to the start of {{run}}: {{exitStatus = 0 # exit status for shell}} * add {{exitStatus}} to the returned Struct * wrap most of {{run}} in {{try: except:}} * set {{exitStatus=1}} (or some other number) if {{doRaise}} is False and something went wrong. Anything else I'm missing?
| 2 |
2,415 |
DM-11942
|
09/19/2017 07:27:34
|
Create new JIRA teams
|
Per RFC-385, please create the following new JIRA teams: - System Management - DM Science - Data Facility
| 1 |
2,416 |
DM-11943
|
09/19/2017 07:33:00
|
Map tickets from old teams to new teams
|
Per RFC-385, please ensure that all tickets currently assigned to "old" teams are assigned to one of the "new" teams. The following direct mappings apply: - Process Middleware → Data Facility - Site Infrastructure → Data Facility The following will need manual intervention to map tickets to whichever team is paying for the work: - DMLT / Management - Project Science - T/CAMs
| 2 |
2,417 |
DM-11944
|
09/19/2017 08:11:59
|
Delete old JIRA teams
|
Please verify that there are no tickets currently assigned to the following teams (this should be the case after DM-11943 is completed), then delete them: - DMLT / Management - T/CAMs - Project Science - Process Middleware - Site Infrastructure
| 1 |
2,418 |
DM-11945
|
09/19/2017 08:19:40
|
Document usage of groups and labels on JIRA
|
Per RFC-385, we have defined a number of teams and standard labels for use on JIRA. Please ensure these are described in the Developer Guide.
| 1 |
2,419 |
DM-11948
|
09/19/2017 09:53:44
|
Add option to force detections in coaddDriver
|
If detections exist somewhere in the butler hierarchy the detection task is not being run. This is problematic if the input repository is a reprocessing of data. Add option to force rerunning detections.
| 1 |
2,420 |
DM-11949
|
09/19/2017 11:35:36
|
newinstall.sh should warn about env vars known to cause problems
|
https://pipelines.lsst.io/v/14-0/install/newinstall.html#running-newinstall-sh-in-an-already-set-up-shell
| 1 |
2,421 |
DM-11954
|
09/19/2017 12:36:46
|
add `w_latest` and `d_latest` docker image tags
|
We are publishing weekly and daily tags to the docker.io registry but it would be convent to also have {{7-stack-lsst_distrib-[dw]_latest}} tags.
| 0.5 |
2,422 |
DM-11957
|
09/19/2017 18:40:59
|
Cannot round-trip >7th degree Chebyshev photometry models
|
If I set {{photometryVisitDegree > 7}} in jointcal's {{ConstrainedPhotometryModel}}, I receive {{RuntimeError: Could not read an AST object from this channel}} when trying to unpersist the {{PhotoCalib}} object containing the {{TransformPoint2ToGeneric}} that I had written via {{butler.put('photoCalib', photoCalib)}}. I will attach example files produced with degree=7 and 11, to help with debugging.
| 3 |
2,423 |
DM-11966
|
09/20/2017 18:06:58
|
Update DCR and simulation tools to Python 3
|
Fix python 3 compatibility issues in the prototype DCR code and StarFast simulator before stackification.
| 1 |
2,424 |
DM-11971
|
09/21/2017 10:33:44
|
memory leak in astshim isSeries function
|
Running the attached script results in the failure {code} ERROR: RuntimeError: AST: Error at line 51 in file src/detail/utils.cc.AssocId(WinMap): There are too many AST Objects in use at once. [__main__] Traceback (most recent call last): File "cause_failure_afwCameraGeom.py", line 13, in <module> pt = getCenterPixel('R:2,2 S:1,1') File "cause_failure_afwCameraGeom.py", line 7, in getCenterPixel centerPoint = camera[name].getCenter(FOCAL_PLANE) RuntimeError: AST: Error at line 51 in file src/detail/utils.cc.AssocId(WinMap): There are too many AST Objects in use at once. {code} It is possible that this is expected/desirable behavior, given that I am needlessly calling {{makeCameraSys}} over and over again on the same inputs. I was under the impression, however, that these would get deleted upon exiting the {{getCenterPixel}} method defined in the script.
| 3 |
2,425 |
DM-11973
|
09/21/2017 11:01:21
|
run-rebuild job does not artifact build logs
|
The jenkins {{run-rebuild}} job does to artifact the build logs, other than the {{manifest.txt}} as is done by {{stack-os-matrix}}. This makes debugging build problems difficult.
| 0.5 |
2,426 |
DM-11974
|
09/21/2017 12:12:04
|
Add calibUsed-only qa plots to CoaddAnalysis
|
Add plots similar to the visit-level *calibUsed*-only versions added in DM-11322 but at the coadd level. This cannot be done until the flags are propagated to the coadd catalogs when DM-11866 lands.
| 1 |
2,427 |
DM-11987
|
09/21/2017 18:34:07
|
ap_verify should allow output to non-empty repos
|
At present, {{ap_verify}} will not run if the destination (output repo) has anything in it. Ideally, it would not care and use the checks in {{ap_pipe}} to be able to pick up where it left off. For example, if it previously completed successfully through raw image ingestion only, I don't want to have to blow that away in order to resume running the next step (calib ingestion).
| 1 |
2,428 |
DM-12003
|
09/22/2017 15:07:14
|
Add default slot for PSF shape
|
{{validate_drp}} needs to access various columns from source catalogs to do its work. Not all of these have slots defined. I am going to add the slot that's missing so the validation scripts are more future proof.
| 2 |
2,429 |
DM-12006
|
09/23/2017 11:07:36
|
Improve testing of copyPolymorphic
|
While working on DM-11971 I found that {{CmpMap}} was missing an override of {{copyPolymorphic}}. Improve the unit tests so this would be caught. Consider removing the definition of {{copyPolymporphic}} and {{copy}} from {{Mapping}}, since that class is abstract. This may make it easier to detect omission in subclasses of {{Mapping}}, since such code will not compile. However, that still leaves deeper classes that could forget to override it, so unit testing is more important. Finally, remove the {{virtual}} since it is implied by {{override}}. Consider adding explicit @copydoc to {{copyPolymorphic}}, though Doxygen already shows the documentation.
| 0.5 |
2,430 |
DM-12007
|
09/25/2017 10:26:12
|
Add tests to display_firefly
|
Following discussions at the 2017 Project & Community Workshop, add a minimal test suite to the display_firefly package. Since a Firefly server is needed to use this package, a minimal test suite can implement an import test. Interactive user tests are already included in the afw.display package.
| 2 |
2,431 |
DM-12008
|
09/25/2017 10:28:23
|
Add a test suite to firefly_client
|
Following discussion at the 2017 Project & Community Workshop, implement a minimal test suite for firefly_client. A minimum automated test is to import the package.
| 2 |
2,432 |
DM-12015
|
09/25/2017 17:04:08
|
Derive SQuaRE, CI and commissioning butler requirements
|
Go through CI, commissioning, and SQuaRE use cases and derive requirements for them.
| 2 |
2,433 |
DM-12019
|
09/26/2017 09:11:08
|
Document use of k8s setup
|
Explain to [~jgates] how to use k8s on ccqserv100 to ccqserv124 and improve related documentation.
| 8 |
2,434 |
DM-12028
|
09/26/2017 13:58:16
|
newinstall.sh should not set PATH unless miniconda is bootstraped
|
{code:java} Craig Loomis [10:40 AM] A fresh `https://raw.githubusercontent.com/lsst/lsst/master/scripts/newinstall.sh`, run without installing a miniconda, generates a loadLSST.bash which contains a troublesome `export PATH="/bin:${PATH}"`. Not entirely sure what is supposed to be happening with the `miniconda_path`, etc. variables in the use-user-anaconda case. MacOS 10.12, anaconda 4.4.0. Tim Jenness [10:42 AM] @josh I think you need to protect `export PATH="${miniconda_path}/bin:${PATH}"` so that it only happens if we installed miniconda. (edited) [10:43] actually it’s this line sorry: ```local cmd_setup_miniconda="export PATH=\"${miniconda_path}/bin:\${PATH}\""``` {code}
| 0.5 |
2,435 |
DM-12029
|
09/26/2017 15:15:16
|
recompress jointcal's testdata zeroed images with fpack
|
Now that the stack supports tile compression of FITS files, and a number of bugs related to handling of FITS image metadata were fixed (DM-11332), I can recompress the images (really just holders of the metadata) I zeroed out in {{testdata_jointcal}} so that they aren't just gzipped files. Once done I can use the new butler calls (DM-9060, DM-9153) to directly access that metadata. This should drastically speed up file ingestion for jointcal. The compression step is probably best done via a short bash script that runs gunzip and then fpack, with some carefully chosen parameters (the files are zeroed, so it probably doesn't matter much which compression scheme is selected). If this works, we can close DM-6911 as well, as that probably got fixed "for free" as part of DM-11332.
| 2 |
2,436 |
DM-12030
|
09/27/2017 09:44:44
|
Persist parquet tables from pipe_analysis scripts
|
In order to make interactive QA plots such as DM-11682, we need to persist parquet files containing the compiled data that are used to make the static matplotlib plots. Admittedly, this should be a temporary step until all the source info is easily loadable from a column-store database, but for now it is necessary. At the visit level it might be less so, but I'll do it anyway. This is probably technically part of DM-10859, but I thought it might be useful to get it in its own ticket so it can be merged quickly.
| 1 |
2,437 |
DM-12031
|
09/27/2017 10:39:16
|
Write presentation for DESC Blending Task Force meeting
|
I've been asked to outline the DRP pipelines to kick off DESC's Blending Task Force on Monday, October 2 at 12pm. This will take some preparation.
| 2 |
2,438 |
DM-12038
|
09/27/2017 14:41:05
|
Update pipe_analysis to provide inputs for Bokeh based QA system
|
Together with [~lauren], ensure that the pipe_analysis scripts efficiently & effectively dump all the information that Tim's plotting scripts need to Parquet files without wasting time generating redundant plots or stripping out useful data.
| 2 |
2,439 |
DM-12040
|
09/27/2017 17:33:47
|
Errors in test_transformFactory.py
|
test_transformFactory.py has a few errors that flake8 catches in the method {{testLinearize}} including: - rng is undefined - nDelta is undefined The reason the test passes is that the code in question never runs: {{invertible}} is False for the one time the code in question runs. This code looks like a potentially useful check for several different transforms (e.g. a radial polynomial and a normal polynomial). Perhaps it should be moved into a new "check<something>" method and called in more than one place?
| 2 |
2,440 |
DM-12052
|
09/28/2017 15:24:23
|
Upgrade ws4py to 0.4.2
|
A {{pip install firefly_client}} pulls in the latest released version of ws4py for web sockets, as version 0.4.2, released on 2017 March 29. This package is included in the LSST stack as a TaP package, version 0.3.5 released on 2014 April 1. This ticket is for upgrading the TaP package to version 0.4.2.
| 2 |
2,441 |
DM-12053
|
09/28/2017 15:29:20
|
Package v1.2 of firefly_client for PyPI and sync the LSST fork
|
Recent improvements to the {{firefly_client}} package need to be gathered into a v1.2 release due to some API changes. At an appropriate point, tag this version, publish to PyPI (for {{pip install firefly_client}}, and synchronize the fork in the lsst Github org.
| 3 |
2,442 |
DM-12061
|
09/29/2017 12:59:42
|
Eliminate test warnings in test_methods.py
|
test_methods.py issues {{DeprecationWarnings}} while testing deprecated assert methods. As of Python 3.2 unittest offers {{assertWarns}} which would allow us to make sure the deprecated methods actually do issue the appropriate warning, and stop pytest from complaining. The trick will be making sure the tests run under older versions of Python (even if pytest warns in that situation).
| 1 |
2,443 |
DM-12068
|
10/02/2017 06:45:29
|
Create Presentation for STSci PSF Photometry Workshop
|
Create a 20 minute presentation on the deblender and the LSST's current plan for photometry in crowded fields.
| 2 |
2,444 |
DM-12069
|
10/02/2017 09:54:59
|
In imageREST_v1 _file_response(): investigate writing lsst.afw.Image straight to data vs temp FITS file first
|
There is existing comment inside imageREST_v1.py mentioning the desirability of writing Image object to data object in memory instead of to a temporary FITS file, then reading it into memory, to improve image response performance and avoidance of using temp files in the local file system. Since afw.Image does not provide an existing/ready method to do so, some investigation of alternative ways of doing this is warranted.
| 5 |
2,445 |
DM-12070
|
10/02/2017 10:17:38
|
Include obs_ctio0m9 and obs_comCam in lsst_distrib
|
Implementation ticket for RFC-391. Get {{obs_ctio0m9}} and {{obs_comCam}} included in {{lsst_obs}} and thus {{lsst_distrib}}
| 1 |
2,446 |
DM-12080
|
10/02/2017 13:28:45
|
Update lsst_repos to include obs_comCam
|
Add {{obs_comCam}} to https://github.com/lsst/repos
| 1 |
2,447 |
DM-12085
|
10/02/2017 14:06:21
|
Camera geometry incorrect and outdated in obs_test
|
The optical distortion model is backwards.
| 1 |
2,448 |
DM-12089
|
10/03/2017 12:03:02
|
Plotly chart with 1M points
|
While it's generally a bad idea to plot more than 100,000 points, Firefly should be still producing the requested plot, however slow. Several issues are covered by this Story, the attached test comes from David Shupe. 1. Firefly produces “Maximum call stack size exceeded”, when mapping large table columns into points to plot. 2. hoverinfo = 'skip' is not honored. When no hover layer is present, tooltips are not displayed and point click and select are not supported. The chart should be displayed with no highlight or select layer and no point select option. Also as a part of this ticket remove fireflyScatter type. Regular scatter plot should behave as fireflyScatter (with the improved tooltips and axis labels).
| 5 |
2,449 |
DM-12102
|
10/04/2017 10:46:19
|
add input validation for SpherePoint(double, double, AngleUnits) constructor
|
Add input validation for the new {{SpherePoint(double, double, AngleUnits)}} constructor
| 1 |
2,450 |
DM-12104
|
10/04/2017 11:57:41
|
Add JSON Schema validation package for ImageServ
|
The specification for the data model of imageREST_v1 is based on the standard JSON schema specification: http://json-schema.org. The schema validator package for JSON-based queries is available on Github: https://github.com/Julian/jsonschema.
| 5 |
2,451 |
DM-12107
|
10/04/2017 12:49:42
|
add CI (jenkins) to developers docs
|
Create dev/user documentation for the DM Jenkins instance. Presumably, to supplant or be linked to from: https://developer.lsst.io/processes/workflow.html?highlight=jenkins#testing-with-jenkins
| 0.5 |
2,452 |
DM-12108
|
10/04/2017 12:55:54
|
Add fake sources after wcs update
|
As it stands now, fake sources that are added according to wcs positions are using the wcs provided by the telescope which can be off. If adding fake sources this should happen after wcs has been updated. This also means that detection and measurements will need to be re-run such that the fake sources are inserted,
| 1 |
2,453 |
DM-12117
|
10/05/2017 12:37:56
|
repositoryCfg.yaml input root not backwards compatible
|
The following butler command fails on a dataset on lsst-dev that was written with {{w_2017_17}}, when using a more recent stack (in this case, {{w_2017_39}}). It also fails if {{root=PATH}} is used instead of just passing the path as the first arg. See discussion beginning here on slack: https://lsstc.slack.com/archives/C3UCAEW3D/p1507226925000049 {code} lsst.daf.persistence.Butler("/datasets/hsc/repo/rerun/DM-10404/SFM/") {code} On the other hand, if the input is specified as "inputs=\{'root':PATH\}" it succeeds. Either {{RepositoryCfg_v1}} should have its version number bumped, or the newer Butler should transparently read the older .yaml. Full stack trace of the failure: {code} In [7]: lsst.daf.persistence.Butler(root="/datasets/hsc/repo/rerun/DM-10404/SFM/ ...: ") --------------------------------------------------------------------------- AttributeError Traceback (most recent call last) <ipython-input-7-04365c25b492> in <module>() ----> 1 lsst.daf.persistence.Butler(root="/datasets/hsc/repo/rerun/DM-10404/SFM/") /software/lsstsw/stack/Linux64/daf_persistence/13.0-40-g460649b+5/python/lsst/daf/persistence/butler.pyc in __init__(self, root, mapper, inputs, outputs, **mapperArgs) 523 self._addParents(repoDataList) 524 --> 525 self._setAndVerifyParentsLists(repoDataList) 526 527 self._setDefaultMapper(repoDataList) /software/lsstsw/stack/Linux64/daf_persistence/13.0-40-g460649b+5/python/lsst/daf/persistence/butler.pyc in _setAndVerifyParentsLists(self, repoDataList) 900 repoData.cfg.addParents(parents) 901 elif repoData.cfgOrigin in ('existing', 'nested'): --> 902 if repoData.cfg.parents != parents: 903 try: 904 repoData.cfg.extendParents(parents) /software/lsstsw/stack/Linux64/daf_persistence/13.0-40-g460649b+5/python/lsst/daf/persistence/repositoryCfg.pyc in __eq__(self, other) 95 if not other: 96 return False ---> 97 return self.root == other.root and \ 98 self.mapper == other.mapper and \ 99 self.mapperArgs == other.mapperArgs and \ AttributeError: 'str' object has no attribute 'root' {code}
| 3 |
2,454 |
DM-12121
|
10/05/2017 14:38:38
|
build-jupyterlabdemo failing
| ERROR: type should be string, got "https://ci.lsst.codes/job/sqre/job/infrastructure/job/build-jupyterlabdemo/67/consoleFull {code:java} > npm pack /usr/share/git/jupyterlab-hub npm WARN lifecycle @jupyterlab/[email protected]~prepublish: cannot run in wd %s %s (wd=%s) @jupyterlab/[email protected] npm run build /usr/share/git/jupyterlab-hub jupyterlab-hub-extension-0.4.1.tgz > node node-version-check.js Uninstalling @jupyterlab/hub-extension from /usr/share/jupyter/lab/extensions Traceback (most recent call last): File \"/usr/bin/jupyter-labextension\", line 11, in <module> load_entry_point('jupyterlab==0.28.0.dev0', 'console_scripts', 'jupyter-labextension')() File \"/usr/lib/python3.4/site-packages/jupyter_core/application.py\", line 267, in launch_instance return super(JupyterApp, cls).launch_instance(argv=argv, **kwargs) File \"/usr/lib/python3.4/site-packages/traitlets/config/application.py\", line 658, in launch_instance app.start() File \"/usr/lib/python3.4/site-packages/jupyterlab/labextensions.py\", line 168, in start super(LabExtensionApp, self).start() File \"/usr/lib/python3.4/site-packages/jupyter_core/application.py\", line 256, in start self.subapp.start() File \"/usr/lib/python3.4/site-packages/jupyterlab/labextensions.py\", line 90, in start raise e File \"/usr/lib/python3.4/site-packages/jupyterlab/labextensions.py\", line 86, in start build(self.app_dir, logger=self.log) File \"/usr/lib/python3.4/site-packages/jupyterlab/commands.py\", line 525, in build return IOLoop.instance().run_sync(func) File \"/usr/lib64/python3.4/site-packages/tornado/ioloop.py\", line 458, in run_sync return future_cell[0].result() File \"/usr/lib64/python3.4/site-packages/tornado/concurrent.py\", line 238, in result raise_exc_info(self._exc_info) File \"<string>\", line 4, in raise_exc_info File \"/usr/lib64/python3.4/site-packages/tornado/gen.py\", line 1069, in run yielded = self.gen.send(value) File \"/usr/lib/python3.4/site-packages/jupyterlab/commands.py\", line 541, in build_async _ensure_package(app_dir, name=name, version=version, logger=logger) File \"/usr/lib/python3.4/site-packages/jupyterlab/commands.py\", line 823, in _ensure_package shutil.copy(src, dest) File \"/usr/lib64/python3.4/shutil.py\", line 229, in copy copyfile(src, dst, follow_symlinks=follow_symlinks) File \"/usr/lib64/python3.4/shutil.py\", line 108, in copyfile with open(src, 'rb') as fsrc: FileNotFoundError: [Errno 2] No such file or directory: '/usr/lib/python3.4/site-packages/packages/shortcuts-extension/schema/plugin.json' /bin/sh: line 0: cd: jupyterlab-savequit: No such file or directory The command '/bin/sh -c mkdir -p /usr/share/git && cd /usr/share/git && npm install -g webpack && git clone https://github.com/jupyterhub/jupyterlab-hub.git && git clone https://github.com/lsst-sqre/jupyterlab-savequit && for i in hub savequit; do cd jupyterlab-${i} && npm install --unsafe-perm && python3 /usr/bin/jupyter labextension link . && npm run build && cd .. ; done' returned a non-zero code: 1 Docker build failed. {code} " | 3 |
2,455 |
DM-12128
|
10/05/2017 17:11:15
|
Fix y-band for HSC in validate_drp
|
The filter maps are not quite right in validate_drp for HSC any more. Fix up the filtermaps so that y-band data can run to completion.
| 2 |
2,456 |
DM-12132
|
10/06/2017 08:18:27
|
Finish high-level Butler design (J)
|
Turn DMTN-056 into a polished, internally-consistent proposal for a new Butler design. This should include: - Python and SQL interfaces at and above the Registry and Datastore level - walkthroughs of how the design supports various use cases - a preliminary list of the concrete Registry and Datastore implementations needed. This is the ticket for [~jbosch]'s piece of the work; a similar ticket will follow for [~pschella].
| 8 |
2,457 |
DM-12135
|
10/06/2017 08:19:45
|
Finish high-level Butler design (P)
|
Turn DMTN-056 into a polished, internally-consistent proposal for a new Butler design. This should include: - Python and SQL interfaces at and above the Registry and Datastore level - walkthroughs of how the design supports various use cases - a preliminary list of the concrete Registry and Datastore implementations needed. This is the ticket for [~pschella]'s piece of the work; DM-12132 captures [~jbosch]'s.
| 8 |
2,458 |
DM-12153
|
10/06/2017 13:55:57
|
Replication master/worker comms improvements
|
Refined and improved communication model providing master-side connection pull and logical connection multiplexing
| 8 |
2,459 |
DM-12155
|
10/06/2017 13:57:27
|
Replication state schema design
|
Database schema to support persistent states for requests, complex replication operations, replicas statuses (including file sizes, checksums), and persistent logging.
| 8 |
2,460 |
DM-12159
|
10/06/2017 14:06:57
|
Replication DB implementation
|
Implementing database support for requests and jobs
| 8 |
2,461 |
DM-12170
|
10/06/2017 16:24:46
|
Create LDM for Butler Use Cases
|
Convert the Butler Use Cases developed by the Working Group into an LDM document ready for DM internal review.
| 8 |
2,462 |
DM-12174
|
10/08/2017 16:09:51
|
community.lsst.org backups not being sent to S3
|
The daily backups on community.lsst.org _should_ be uploaded to S3 (builtin Discourse functionality), but they aren't. We are doing weekly DigitalOcean snapshots that mitigate some of the data loss risk, but it would be better if we copied those daily backups. Check if it's a permissions issue that can be fixed. Alternatively, set up a {{cron}} service that {{scp}}s the backup to S3.
| 1 |
2,463 |
DM-12184
|
10/09/2017 16:57:07
|
Coaddition Tasks cannot assume that N masks can fit in memory
|
[~pgee] and [~tjohnson] have recently reported high memory usage by {{SafeClipAssembleCoadd}} for examples with either large N (number of warps) or large patch sizes. For simulated data processed by DESC, [~tjohnson] reported 70GB processes coadding 6kX6k patches and unknown N. (Fill this in if you remember) For DES data, [~pgee] reported memory usage >32GB with 10kX10k patches and unknown N. (Note: patches this large are not recommended). Masks were still uint16 at the time. Profiling reveals that this memory usage was likely due to holding N masks in memory. In this example, patches were 4200X4200, mask pixels = int32, and N=33. 4200*4200*4*33/2**20 = 2220MiB. Full profiling print out attached. {code} Line # Mem usage Increment Line Contents ================================================ 1294 556.6 MiB 0.0 MiB @profile 1295 def detectClip(self, exp, tempExpRefList): 1296 """! 1297 \brief Detect clipped regions on an exposure and set the mask on the individual tempExp masks ... 1326 586.2 MiB 0.0 MiB clipFootprints = [] 1327 586.2 MiB 0.0 MiB clipIndices = [] 1328 1329 # build a list with a mask for each visit which can be modified with clipping information 1330 586.2 MiB 0.0 MiB tempExpClipList = [tmpExpRef.get(self.getTempExpDatasetName(self.warpType), 1331 immediate=True).getMaskedImage().getMask() for 1332 2806.8 MiB 2220.7 MiB tmpExpRef in tempExpRefList] 1333 1334 2813.1 MiB 6.3 MiB for footprint in footprints.getFootprints(): 1335 2813.1 MiB 0.0 MiB nPixel = footprint.getArea() {code} For the simulated data example, N=~1000, patch size = 6KX6K, mask pixel size = uint16 (May 2016) , would have yielded a theoretical usage of 6e3*6e3*2*1000/2**30 = 67GiB. For the DES, example with the 10kX10k patches any more than 150 visits would have not fit on the 32G system. Now that we've switched to int32 masks this is even more important, as the memory usage doubles for examples with large N or large patches. Even with 4k X 4k patches, it only takes N=180 to get memory usage > 10GB. This ticket will remove the assumption that N masks can fit in memory from the implementation of {{CompareWarpAssembleCoadd}}. Because the solution will likely be the same for SafeClipAssemble too it will likely be implemented at the same time.
| 8 |
2,464 |
DM-12189
|
10/10/2017 12:19:48
|
Firefly should not longer use path segment parameters, it should use query string
|
Firefly has been using path segment parameters to specify meta parameters {{;}} such as channel and action type. This does not work exactly has expected on all servers. Therefore we are going to go back to using classic query string {{?}}. {{wsch}} (channel) parameter will now be {{__wsch}}, {{a}} (action) will now be {{__a}} Fix should be put into rc.
| 2 |
2,465 |
DM-12200
|
10/11/2017 11:48:09
|
jointcal tests fail when optional testdata_jointcal is not present
|
The most recently merged PR is https://github.com/lsst/jointcal/pull/45 {{jointcal}} from master is not buiding via {{eups distrib install}}. I suspect this is because unit tests we introduced that are dependent on the {{setupOptiona}} {{testdata_jointcall}} product. Test that rely on optional packages should bail out gracefully without failing. {code:java} [build] lsst.afw.image.utils.resetFilters() # Load or fake the necessary metadata for each CcdImage (using ccd=12) > data_dir = lsst.utils.getPackageDir('testdata_jointcal') E lsst.pex.exceptions.wrappers.NotFoundError: E File "src/Utils.cc", line 42, in std::string lsst::utils::getPackageDir(const string&) E Package testdata_jointcal not found {0} E lsst::pex::exceptions::NotFoundError: 'Package testdata_jointcal not found' tests/test_photometryModel.py:26: NotFoundError =============== 4 failed, 37 passed, 19 skipped in 6.84 seconds ================ Global pytest run: failed {code}
| 0.5 |
2,466 |
DM-12208
|
10/12/2017 10:42:52
|
Firefly viewer loading race condition - should wait for document load
|
Firefly sprint can attempt to access the DOM before it is ready. It needs to check {{window.document.readyState}} and wait for onload if necessary.
| 2 |
2,467 |
DM-12215
|
10/12/2017 15:55:47
|
matchedVisitMetricsTask outputPrefix results in hidden files
|
{{matchedVisitMetricsTask.py}} has a default {{outputPrefix="."}}, which results in files like {{.\_PA1.png}} and {{..json}} in the directory it was run from. The default should probably be {{""}}, or {{"./"}}, as the plotting code uses {{%s_%s}} with the first substitution being outputPrefix. An alternate solution would be to use {{--output}} for this purpose, since this is a {{Task}}, and does appear to require an explicit output. Edit to add: the {{outputPrefix}} that gets passed on to {{validate}} should maybe be {{output/outputPrefix}}, though that still doesn't work with the substitution that's used when making the plot files, if outputPrefix is {{""}}
| 0.5 |
2,468 |
DM-12216
|
10/12/2017 19:10:48
|
Not all plot files append outputPrefix with an underscore
|
Some of validate_drp's plot filenames are created as {{outputPrefix+"something.png"}} instead of {{"%s_%s"%(outputPrefix, "something.png")}}. This can make the output filenames less readble. As an example showing the differences: {code} matchedVisit_HSC-IAM1_D_5_arcmin_17.0_21.5_mag.png matchedVisit_HSC-IAM2_D_20_arcmin_17.0_21.5_mag.png matchedVisit_HSC-ITE1_D_1_arcmin.png matchedVisit_HSC-ITE2_D_5_arcmin.png matchedVisit_HSC-I_PA1.png matchedVisit_HSC-I_check_astrometry.png matchedVisit_HSC-Icheck_photometry.png {code} Should be an easy fix, unless there's other code that expects the files to be handled that way?
| 0.5 |
2,469 |
DM-12230
|
10/13/2017 17:10:53
|
Mapping.applyForward and applyInverse fail on empty arrays
|
Mapping.applyForward and Mapping.applyInverse fail on empty arrays: it triggers an AST error. This showed up as a mysterious error in DM-10765 in afw test_footprint1.py The fix is trivial: have Mapping._tran skip the call to astTranN if the input array has 0 points.
| 1 |
2,470 |
DM-12232
|
10/16/2017 03:11:26
|
Create presentation for DESC blending group
|
The LSST-DESC blending group has asked Peter and I to give a presentation on the new deblender. Most of the talk was already created for LSST-2017, however some work is needed to update it with the latest results using multiple components and the current status.
| 1 |
2,471 |
DM-12233
|
10/16/2017 07:24:57
|
Improve interface for multiple components
|
The current interface for the deblender takes constraints and components as arguments but doesn't utilize that information when building the constraints for each object (for example having default constraints for different component types). It also doesn't pass those parameters to the main NMF algorithm without being explicitly given as function arguments, so improving this interface would make constraints and components easier to use and more robust against inconsistencies (for example this weekend I had forgotten to pass the constraints to the function to build the monotonicity operator, causing half of the components to not have monotonicity applied).
| 3 |
2,472 |
DM-12254
|
10/17/2017 15:30:27
|
Switch jointcal to default to IndexedRefObj instead of a.net
|
Jointcal currently defaults to using {{LoadAstrometryNetObjectsTask}} for its {{refObjLoader}}. This can be dangerous in that if one doesn't retarget in the config, loading the skyCircle will fail and getting to that point can take a long time with lots of files. We also plan to phase out A.Net catalogs over time, so this is a good idea long term as well. It does require some reconfiguration of the unittests however.
| 1 |
2,473 |
DM-12256
|
10/17/2017 17:26:31
|
Incorporate AssociationTask into ap_pipe
|
Let {{ap_pipe}} call {{AssociationTask}} and verify that the code runs to completion on the standard {{ap_pipe}} dataset ({{verify_ap_hits2015}}, {{visit=410985 ccdnum=25}}).
| 2 |
2,474 |
DM-12257
|
10/17/2017 17:28:14
|
Implement association step in ap_verify
|
Let {{ap_verify}} call the association step in {{ap_pipe}} and verify that the code runs to completion on the standard {{ap_pipe}} dataset ({{verify_ap_hits2015}}, {{visit=410985 ccdnum=25}}).
| 1 |
2,475 |
DM-12270
|
10/19/2017 10:54:37
|
AST persistence is not exact
|
AST persistence by writing to a string is not exact because double precision values are not written to sufficient precision. The result is tiny differences in computations from FrameSets that have been persisted and unpersisted. David Berry is working on a fix. This ticket provides a place to record the issue and mark temporary workarounds in our tests.
| 1 |
2,476 |
DM-12272
|
10/19/2017 12:04:58
|
Fix bug in arrayFromVector
|
arrayFromVector has a bug: internally the strides vector is wrong, which can lead to memory issues. Fix this, wrap it for use in Python, and provide a unit test. Also, coefficient array arguments for {{ChebyMap}}, {{PolyMap}} and {{MatrixMap}} do not have const elements. Fix this.
| 0.5 |
2,477 |
DM-12305
|
10/19/2017 14:55:21
|
Research HiPS viewing in Firefly
|
Begin researching HiPS viewing in Firefly. Do the following: * Learn about Healpix * Learn about HiPS * Determine how a viewer would fit into Firefly * Look at Aladin Lite code and tool * Look at Aladin Lite API * If straightforward, add a proof of concept HiPS viewing into Firefly.
| 20 |
2,478 |
DM-12306
|
10/19/2017 15:08:00
|
Make datasets optional for ap_verify
|
According to the [developer guide|https://developer.lsst.io/build-ci/new_package.html#handling-git-lfs-backed-repos], git-lfs repos cannot be a required dependency of any Stack package. {{ap_verify}} currently requires {{ap_verify_hits2015}} for testing purposes. This behavior must be changed before {{ap_verify}} can be included in {{lsst_apps}}. Note that {{Dataset}} cannot be tested at all without a dataset package, and command-line parsing can only be partially tested.
| 2 |
2,479 |
DM-12307
|
10/19/2017 15:48:02
|
update blue-ocean to 1.3.0
|
Blueocean 1.3.0 has been released and the changelog lists a fix for non-working http links in console output. Perhaps this will also fix linking to triggered jobs? changelog: https://plugins.jenkins.io/blueocean
| 1 |
2,480 |
DM-12315
|
10/20/2017 12:15:21
|
Generalize ap_pipe to non-HiTS data
|
Currently, {{ap_pipe}} has some hardcoded assumptions that it is working on DECam data, and the ingestion step is specific to HiTS. Both of these assumptions should be lifted (likely delegating much of the work to a standards-compliant {{obs_*}} package) to allow {{ap_pipe}} to be run on arbitrary datasets.
| 8 |
2,481 |
DM-12359
|
10/23/2017 11:38:55
|
send Task log output to stdout
|
Send Task {{lsst.log}} output to stdout instead of stderr. This is the implementation ticket for RFC-402.
| 1 |
2,482 |
DM-12361
|
10/23/2017 13:11:39
|
Qserv needs more changes after xrootd shared object renaming
|
In DM-12308 we changed the names of some shared objects that come from xrootd. There are still references to other shared object in xrootd config file that needs similar change.
| 0.5 |
2,483 |
DM-12362
|
10/23/2017 14:54:45
|
SCons 3 rebuilds qserv binaries every time
|
For some reason new SCons 3 wants to rebuild all binaries (shared objects included) on every scons invocation.
| 0.5 |
2,484 |
DM-12367
|
10/24/2017 09:14:43
|
Add an abstract base class for spatially-varying transmission curves
|
Add a C++ abstract base class (with Python wrappers) for classes that represent wavelength-dependent transmission (e.g. filter curves) as a function of position. These should make it possible to use an implementation in one coordinate system (e.g. focal plane coordinates) to provide a transmission curve looked up via another set of coordinates (e.g. pixels) via a proxy subclass of the ABC. Design probably deserves an RFC. Many conversations on this the appropriate design for this functionality have been started but none have converged; some prior art exists in Sims. Should defer discussions about functionality beyond what's needed for HSC filter curve coaddition as much as possible. These objects will need to be persistable with afw::table::io.
| 2 |
2,485 |
DM-12368
|
10/24/2017 09:16:23
|
Add a trivial spatially-constant transmission curve implementation
|
Implement the interface of DM-12367 for non-spatially varying transmission curves.
| 0.5 |
2,486 |
DM-12369
|
10/24/2017 09:18:20
|
Add a coordinate-transform proxy transmission curve
|
Implement the interface of DM-12367 for a proxy that changes the coordinate system of a different transmission curve object, using the new Transform objects. Should support at least focal-plane to CCD pixels and coadd pixels to CCD pixels. Does not obviously need to support sky coordinates.
| 1 |
2,487 |
DM-12370
|
10/24/2017 09:19:49
|
Add a coadded transmission curve implementation
|
Implement the interface of DM-12367 with a proxy that combines other concrete transmission curves as a function of position.
| 2 |
2,488 |
DM-12372
|
10/24/2017 09:21:07
|
Add transmission curves for HSC filter traces
|
Implement the interface of DM-12367 using [~erykoff]'s approach to interpolating HSC filter traces.
| 2 |
2,489 |
DM-12373
|
10/24/2017 09:25:48
|
Add spatially-varying transmission curves to Exposure/ExposureRecord
|
Attach an instance of the ABC defined in DM-12367 to Exposure and ExposureRecord. While we could add these to Filter rather than add it to Exposure directly, my preliminary feeling is that it'll be better to let this represent the full throughput rather than associating it strictly with the filter. It may *also* make sense to make Filter hold one of these objects to represent its contribution only, but that isn't what's needed for DM-12366.
| 3 |
2,490 |
DM-12374
|
10/24/2017 09:26:49
|
Add transmission curve coaddition to coadd Task code
|
Add high-level code to combine spatially-varying transmission curves during coaddition.
| 2 |
2,491 |
DM-12375
|
10/24/2017 09:31:22
|
Attach transmission curves during ISR or std_raw
|
Make sure we have spatially-varying transmission curves attached to Exposures by the time we write the calexp dataset. Will need to decide how to store filter curves - best bet is probably as a butler dataset keyed only by filter for now.
| 2 |
2,492 |
DM-12384
|
10/24/2017 18:53:48
|
exit/raise when data is less than parameters
|
Jointcal should exit or raise when the number of data points is less than the number of fitting parameters (e.g., constrained 4th order polynomial) with only 3 stars). I thought there was once such a check for astrometry, but I can't find it now.
| 2 |
2,493 |
DM-12390
|
10/25/2017 11:34:44
|
Butler doesn't raise when failing to write data
|
If you supply insufficient data in the dataId for a butler to {{put()}} the data, it will just pretend to have done it (_i.e._ it doesn't raise or warn, it just silently returns as if all was fine). If one does {noformat} butler.put(img, 'postISRCCD', dataId={'visit':123,'filter':'r'}) {noformat} where the {{postISRCCD}} template looks like {{template: "postISRCCD/postISRCCD_v%(visit)d_f%(filter)s.fits"}} then this will work. However, if you provide an empty dataId dict, it will not complain, but nothing will get written (of course). This seems like a moderately serious problem, as it is easy to accidentally under-specify or mis-specify a dataId. (FWIW, if I change 'postISRCCD' to an undefined dataset type I _do_ get an error).
| 2 |
2,494 |
DM-12391
|
10/25/2017 13:22:12
|
Improve error message for case when warpCompare can't find any psf-Matched warps
|
This message means that no psfMatched warps were found, and a static sky model could not be built but a user might not deduce that from: ```Traceback (most recent call last): File "/software/lsstsw/stack/Linux64/ctrl_pool/13.0-6-gf96f8ec+54/python/lsst/ctrl/pool/parallel.py", line 496, in logOperation yield File "/software/lsstsw/stack/Linux64/pipe_drivers/13.0-21-g61c0bd4+9/python/lsst/pipe/drivers/coaddDriver.py", line 261, in coadd coaddResults = self.assembleCoadd.run(patchRef, selectDataList) File "/software/lsstsw/stack/Linux64/pipe_base/13.0-14-g8b3bf66+40/python/lsst/pipe/base/timer.py", line 150, in wrapper res = func(self, *args, **keyArgs) File "/home/yusra/lsst_devel/LSST/DMS/pipe_tasks/python/lsst/pipe/tasks/assembleCoadd.py", line 388, in run supplementaryData = self.makeSupplementaryData(dataRef, selectDataList) File "/home/yusra/lsst_devel/LSST/DMS/pipe_tasks/python/lsst/pipe/tasks/assembleCoadd.py", line 1629, in makeSupplementaryData templateCoadd = self.assembleStaticSkyModel.run(dataRef, selectDataList).coaddExposure AttributeError: 'NoneType' object has no attribute 'coaddExposure'```
| 1 |
2,495 |
DM-12417
|
10/26/2017 11:26:44
|
yaml dump prepends !!python/unicode to everything
|
Calling {{dumpToFile}} on a daf_persistence.policy causes !!python/unicode to get prepended to everything. As per https://stackoverflow.com/questions/1950306/pyyaml-dumping-without-tags it seems that changing {{yaml.dump()}} to {{yaml.safe_dump()}} should fix this.
| 0.5 |
2,496 |
DM-12432
|
10/26/2017 13:21:33
|
Fix timing measurement construction
|
At present, {{ap_verify}} produces few timing measurements because I was not aware of DM-11935 when I wrote the code (and the pipeline was not yet in a state where it could be tested). The code should be changed so that measurements are reliably generated, either by fixing DM-11935 or by working around it.
| 2 |
2,497 |
DM-12436
|
10/26/2017 14:52:53
|
remove jenkins py2 builds - post v15 release
|
Remove py2 support from: - jenkins ({{stack-os-matrix}} + assorted publishing/release related jobs) - {{newinstall.sh}} - {{lsstsw}}
| 1 |
2,498 |
DM-12445
|
10/26/2017 20:03:54
|
Set appropriate default configs for CompareWarp Coadds
|
Set default configs for makeCoaddTempExp, assembleCoadd, and coaddDriver in pipe_tasks and obs_subaru based on testing results. These include the usual obs_subaru defaults for AssembleTask and SafeClipAssembleTask and more. For example, these will probably include these configs: {code} config.makeCoaddTempExp.makePsfMatched=True config.makeCoaddTempExp.warpAndPsfMatch.psfMatch.kernel['AL'].kernelSize=29 config.makeCoaddTempExp.warpAndPsfMatch.psfMatch.kernel['AL'].alardSigGauss=[1.0, 2.0, 4.5] config.makeCoaddTempExp.modelPsf.defaultFwhm=6.1 from lsst.pipe.tasks.assembleCoadd import CompareWarpAssembleCoaddTask config.assembleCoadd.retarget(CompareWarpAssembleCoaddTask) config.assembleCoadd.badMaskPlanes = ("BAD", "EDGE", "SAT", "INTRP", "NO_DATA",) config.assembleCoadd.doSigmaClip = False config.assembleCoadd.subregionSize = (10000, 200) # 200 rows (since patch width is typically < 10k pixels) config.assembleCoadd.doMaskBrightObjects = True config.assembleCoadd.sigmaClip = 1.5 config.assembleCoadd.clipIter = 3 config.assembleCoadd.assembleStaticSkyModel.sigmaClip = 1.5 config.assembleCoadd.assembleStaticSkyModel.clipIter = 3 config.assembleCoadd.assembleStaticSkyModel.statistic = 'MEANCLIP' config.assembleCoadd.doNImage=True config.assembleCoadd.doWrite=True {code}
| 3 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.