Repository navigation
test: hermetic unit-test layer, batches 21-30 - #1084
Open
ilayfalach wants to merge 59 commits into
Open
ilayfalach wants to merge 59 commits into
ilayfalach wants to merge 59 commits into
Conversation
…eded) processorList, read_points_file, getMeshExtent, getTimeList. Records B83: getTimeList's single-processor branch checks os.path.isdir(x) on the bare name instead of os.path.join(case, x), so it always resolves against the cwd instead of the case directory -- the method returns an empty list for the common (non-parallel) case. Local-only, not pushed. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…nsion Records three NameError bugs in absractEulerianSolver_toolkitExtension (B84-B86): flowType references an undefined SIMULATIONTYPE_COMPRESSIBLE; blockMesh_setBoundFromFile checks an undefined eulerianWF instead of its own eulerianWorkFlow parameter; blockMesh_setDomainHeight is an unadapted copy-paste of the same method, referencing fileName/dx/dy that aren't its parameters, with its real input (Z) never used. Local-only, not pushed. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…concentration sourcesTypeList, the six makeSource_* geometry generators, writeParticlePositionFile, and analysis.calcConcentrationPointWise. No new defects -- all verified correct against their documented geometry/math. Local-only, not pushed. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
time, dict, csv_pandas, JSON_pandas, netcdf_xarray, geopandas, image. Records B87: DataHandler_JSON_geopandas.saveData calls resource.to_json(fileName, **kwargs), but geopandas.GeoDataFrame.to_json's first positional parameter is `na` (not a path) and the method returns a JSON string rather than writing a file -- every call raises immediately. getData is independent and works correctly, verified against a hand-written file. Local-only, not pushed. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…ct.py) Adds unit coverage for FilesDirectory/filesDirectory, updateProjectNameOnDoc, the saveMeasurementData/saveCacheData/saveSimulationData family, the get/delete *DocumentsAsDict trio across all three collections, getMetadata, getAllDocuments, getDocumentByID, setCounter/getCounterAndAdd, and both getProjectList entry points. Finds B88: Project.getProjectList is decorated @staticmethod but declares an unused leading `cls` parameter, so the intended call Project.getProjectList() raises TypeError. Pinned via xfail(strict=True). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…dule) Adds unit coverage for Parser_CampbellBinary/CampbellBinaryInterface and Parser_OldStyleMetaDataParquet in the (currently unimported) experiment parsers module. Finds B89: Parser_CampbellBinary.getData carries the exact same off-by-two columnsIndexes bug already pinned as B82 in the meteorology highfreqdata CampbellBinary parser -- an independent copy of the same broken code. Finds B90: _getLists builds stationsData with the default (wrong) DataFrame.from_dict orientation, putting attribute names in the row index instead of station names -- breaks parse() for any station with more than one location attribute. Both pinned via xfail(strict=True). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds unit coverage for ToolkitRepository (register/lookup/table, backed by the session mongomock reroute) and the pure dict/JSON repository-export helpers (documentContentHash, documentToRepositoryItem, mergeDocumentsIntoRepository, deduplicateRepository). Finds B91: mergeDocumentsIntoRepository records report['added']/ ['overwritten'] with the pre-disambiguation item_name, before _uniqueItemName renames the key actually written -- on a name collision between distinct documents, the report points at the wrong document. Pinned via xfail(strict=True). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds unit coverage for presenation/Plots wiring and the pure dict/colormap builders (_getCountourDict, _getContourfDict, _getcmap). No bugs found; the heavier SeasonalPlots/DailyPlots contour-plotting methods need realistically-shaped wind data and are left uncovered. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…esTable Adds unit coverage for LSMToolkit's constructor/property validation, the internal analysis wiring, and loadData's file/DB paths (NOSAVE, FILEANDDB, duplicate-registration rejection). Finds B92: loadData reads the template JSON's own desc into a local variable used only to extract the name -- it's never passed to addDataSource, so registered datasource documents never carry the template's actual content (e.g. params). getTemplatesTable unconditionally pops 'params' from every document's desc, so it crashes for every template ever registered via the default (FILEANDDB) loadData path. Pinned via xfail(strict=True). getSimulations/getSimulationsList/prepareSlurmLSMExecution left uncovered -- need real SingleSimulation documents or a Slurm environment. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds unit coverage for interp/checkInterpulation (distance-weighted station interpolation, both scalar and vector-valued). Finds three bugs: - B93: windprofile is defined inside the class body with no `self` parameter -- calling it on an instance (the normal way) shifts every argument and crashes; only works called directly on the class. - B94: interpPandas writes results via a chained assignment that pandas' copy-on-write silently drops -- the interpulation column always stays None, no exception raised. - B95: interpArray passes topography=... to interpPandas, which has no such parameter -- every call raises TypeError immediately. All three pinned via xfail(strict=True). canopyWindProfile.py's single function left uncovered (needs full geopandas geometry + 3D grid setup). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…n constructor Adds unit coverage for the property setters/getters, particle mass derivation, all four drag-coefficient formulas, and the hit_ground ODE event -- all verified via __new__ since the real constructor cannot run. Finds B96: __init__ calls MeteorologyFactory().getMeteorology(...), a method that does not exist (only getMeteorologyFromU10/ getMeteorologyFromURefHeight are defined, with different parameters). Every construction of FallingNonEvaporatingDroplets raises AttributeError -- the class is entirely unusable through its public API. Pinned via xfail(strict=True). getTerminalVelocity/_VTFunc/correctionCloud_Plume/_Puff/solveToTime/ _fallingParticle left uncovered -- all need self.meteorology, unreachable while B96 stands. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds unit coverage for _parse_query_value (query-string type coercion) and load_project_name (reading projectName from caseConfiguration.json). No bugs found. The ~26 argparse command handlers all touch a real Project/toolkit via the module's lazy _setup() and are left uncovered. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
… tests Discovered while validating against a venv pinned to requirements.txt with the hermes.workflow stub fix installed: - B84/B85/B86 (flowType/blockMesh_setBoundFromFile/blockMesh_setDomainHeight): were real, accurate findings at the time; commit 276d9a9 ("fix: repair OpenFOAM Eulerian solver name errors and method mismatches") fixed all three independently of this test-expansion effort. Rewrote test_openfoam_eulerian_solver.py to check the real (working) behavior: flowType for both incompressible values, and both the non-eulerian-workflow rejection path and the real-workflow success path for the two blockMesh_* setters (via a _FakeWorkflow subclass of the now properly-stubbed workflow_Eulerian). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Discovered while validating against a venv pinned to requirements.txt: B94 was verified against a locally-drifted pandas (3.0.2) where copy-on-write is mandatory. Under the pinned pandas (2.2.3), copy-on-write is opt-in -- the chained assignment in interpPandas still works, with only a FutureWarning/SettingWithCopyWarning. Replaced the xfail characterization with a single passing test; noted in both the module docstring and the findings log that this would resurface if pandas is upgraded past 2.x with copy-on-write made the default. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…lass) Adds unit coverage for guessHandler (module-level wrapper), DataHandler_tif (real GeoTIFF via rasterio, which is installed under the pinned requirements.txt unlike osgeo/tables/zarr), DataHandler_numpy_dict_array (.npz round trip), and DataHandler_Class (dynamic class loading via both the sys.path and resource-fallback code paths, the parameters/kwargs merge rule, instantiate=False, and the two documented error cases). No bugs found. Verified directly against the pinned venv, not the drifted shared heraenv. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…kup) Adds unit coverage for SingleSimulation's construction (Dataset directly, and a document with getData()), params/version properties, getDosage (both the datetime64 and numeric-datetime branches, Q scaling, unit attrs), and getConcentration. Finds B98: getConcentrationAtPoint, called exactly as documented with a single scalar x/y/datetime, always raises IndexError -- .interp() on a fully-specified point returns a 0-dimensional DataArray, and .values[0] on a 0-d array is invalid. Pinned via xfail(strict=True). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…okups) Adds unit coverage for LSMTemplate's properties (Toolkit, doctype_simulation, dirPath, params, version, templateName, modelFolder), prepareParams (unit conversion, duration-as-minutes, TopoXn/TopoYn int casting), and getSimulationByName/getSimulations query dispatch (via a monkeypatched SingleSimulation, since a real one needs actual netcdf files on disk). The document is built by hand rather than through LSMToolkit.loadData(), to isolate this file's own logic from the already-documented B92 (loadData never persists the template JSON's own desc). Finds B99: getSimulationByID calls self.getDocumentByID(id), a Project/toolkit method LSMTemplate does not have on itself -- every call raises AttributeError. Pinned via xfail(strict=True). run()/_toNetcdf()/_getSimulationsList/getSimulationsTable left uncovered -- need a real LSM binary or heavier table-building setup. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…MaxConcentration No bugs found. presentation/classLoggerName/getToolkitDocument on abstractToolkit; Agent.tenbergeCoefficient getter/setter; and CalculatorMaxConcentration.calculate (pandas path) + toJSON. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…che) No bugs found. Serialization helpers, _get_full_func_name, and cacheFunction end-to-end (caching, cache-clear, returnDoc, postProcessFunction). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Finds B100: FixedPointClippedDropletCloud.__init__ reads self.clippedDiameter before setting it (via super().__init__() polymorphic dispatch), crashing before it can even reach the already-documented B96. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
This was referenced Sep 20, 2026
Fix: Five plotting defects that drew the wrong thing or nothing (B143, B144, B224, B225, B322)
#1090
Open
Open
Open
Open
Open
Open
Open
lior-antonov
added this pull request to stack #1117
September 23, 2026 10:19
This was referenced Oct 15, 2026
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fourth in the stack: #1067 is 1-10, #1083 is 11-20, this is 21-30, and one more follows for 31-33.
Base is
tests/batch20-gis-buildings, so this diff is batches 21-30 alone (59 commits). Merge #1083 first.These are the batches that take the unit layer from partial coverage to roughly 60%, and they are where most of the defect log after B200 comes from.
Production code is touched only where the suite could not otherwise run. Every defect is pinned by an
xfail(strict=True)test rather than fixed here.🤖 Generated with Claude Code