Merge pull request #338 : Adds mymdc techniques annotation

Adds mymdc techniques annotation

damnit/ctxsupport/ctxrunner.py

@@ -17,8 +17,10 @@
 import traceback
 from datetime import timezone
 from enum import Enum
+from functools import wraps
 from graphlib import CycleError, TopologicalSorter
 from pathlib import Path
+from typing import Any
 from unittest.mock import MagicMock
 
 import extra_data
@@ -70,45 +72,55 @@ def __init__(self, proposal, timeout=10, init_server="https://exfldadev01.desy.d
 
  self._headers = { "X-API-key": self.token }
 
- def _run_info(self, run):
- key = (run, "run_info")
- if key not in self._cache:
- response = requests.get(f"{self.server}/api/mymdc/proposals/by_number/{self.proposal}/runs/{run}",
- headers=self._headers, timeout=self.timeout)
- response.raise_for_status()
- json = response.json()
- if len(json["runs"]) == 0:
- raise RuntimeError(f"Couldn't get run information from mymdc for p{self.proposal}, r{run}")
-
- self._cache[key] = json["runs"][0]
-
- return self._cache[key]
-
- def sample_name(self, run):
- key = (run, "sample_name")
- if key not in self._cache:
- run_info = self._run_info(run)
- sample_id = run_info["sample_id"]
- response = requests.get(f"{self.server}/api/mymdc/samples/{sample_id}",
- headers=self._headers, timeout=self.timeout)
- response.raise_for_status()
-
- self._cache[key] = response.json()["name"]
-
- return self._cache[key]
-
- def run_type(self, run):
- key = (run, "run_type")
- if key not in self._cache:
- run_info = self._run_info(run)
- experiment_id = run_info["experiment_id"]
- response = requests.get(f"{self.server}/api/mymdc/experiments/{experiment_id}",
- headers=self._headers, timeout=self.timeout)
- response.raise_for_status()
-
- self._cache[key] = response.json()["name"]
-
- return self._cache[key]
+ @staticmethod
+ def _cache(func):
+ @wraps(func)
+ def wrapper(self, run):
+ key = (run, func.__name__)
+ if key in self._cache:
+ return self._cache[key]
+ self._cache[key] = func(self, run)
+ return self._cache[key]
+ return wrapper
+
+ @_cache
+ def _run_info(self, run: int) -> dict[str, Any]:
+ response = requests.get(f"{self.server}/api/mymdc/proposals/by_number/{self.proposal}/runs/{run}",
+ headers=self._headers, timeout=self.timeout)
+ response.raise_for_status()
+ json = response.json()
+ if len(json["runs"]) == 0:
+ raise RuntimeError(f"Couldn't get run information from mymdc for p{self.proposal}, r{run}")
+
+ return json["runs"][0]
+
+ @_cache
+ def techniques(self, run: int) -> dict[str, Any]:
+ run_info = self._run_info(run)
+ response = requests.get(f'{self.server}/api/mymdc/runs/{run_info["id"]}',
+ headers=self._headers, timeout=self.timeout)
+ response.raise_for_status()
+ return response.json()['techniques']
+
+ @_cache
+ def sample_name(self, run: int) -> str:
+ run_info = self._run_info(run)
+ sample_id = run_info["sample_id"]
+ response = requests.get(f"{self.server}/api/mymdc/samples/{sample_id}",
+ headers=self._headers, timeout=self.timeout)
+ response.raise_for_status()
+
+ return response.json()["name"]
+
+ @_cache
+ def run_type(self, run: int) -> str:
+ run_info = self._run_info(run)
+ experiment_id = run_info["experiment_id"]
+ response = requests.get(f"{self.server}/api/mymdc/experiments/{experiment_id}",
+ headers=self._headers, timeout=self.timeout)
+ response.raise_for_status()
+
+ return response.json()["name"]
 
 
 class ContextFileErrors(RuntimeError):
@@ -173,7 +185,7 @@ def check(self):
  for name, var in self.vars.items():
  mymdc_args = var.arg_dependencies("mymdc#")
  for arg_name, annotation in mymdc_args.items():
- if annotation not in ["sample_name", "run_type"]:
+ if annotation not in ["sample_name", "run_type", "techniques"]:
  problems.append(f"Argument '{arg_name}' of variable '{name}' has an invalid MyMdC dependency: '{annotation}'")
 
  if problems:
@@ -301,12 +313,8 @@ def execute(self, run_data, run_number, proposal, input_vars) -> 'Results':
  elif annotation.startswith("mymdc#"):
  if mymdc is None:
  mymdc = MyMetadataClient(proposal)
-
- mymdc_field = annotation.removeprefix("mymdc#")
- if mymdc_field == "sample_name":
- kwargs[arg_name] = mymdc.sample_name(run_number)
- elif mymdc_field == "run_type":
- kwargs[arg_name] = mymdc.run_type(run_number)
+ metadata = annotation.removeprefix('mymdc#')
+ kwargs[arg_name] = getattr(mymdc, metadata)(run_number)
 
  elif annotation == "meta#run_number":
  kwargs[arg_name] = run_number

docs/backend.md

@@ -76,16 +76,26 @@ Variable functions can return any of:
 
 The functions must always take in one argument, `run`, to which is passed a
 [`DataCollection`](https://extra-data.readthedocs.io/en/latest/reading_files.html#data-structure)
-of the data in the run. In addition, a function can take some other special
-arguments if they have the right _annotations_:
+of the data in the run. In addition, a function can take some other special
+arguments if they have the right _annotations_, currently. 
+`meta` accesses internal arguments:
 
 - `meta#run_number`: The number of the current run being processed.
 - `meta#proposal`: The number of the current proposal.
 - `meta#proposal_dir`: The root
  [Path](https://docs.python.org/3/library/pathlib.html) to the current
  proposal.
-- `mymdc#sample_name`: The sample name from myMdC.
+
+`mymdc` requests information from the EuXFEL data management portal
+[MyMDC](https://in.xfel.eu/metadata/):
+
 - `mymdc#run_type`: The run type from myMdC.
+- `mymdc#sample_name`: The sample name from myMdC.
+- `mymdc#techniques`: list of
+ [technique](https://expands-eu.github.io/ExPaNDS-experimental-techniques-ontology/index-en.html)
+ associated with the run. Each technique listed is a `dict` containing the
+ following keys: `description`, `flg_available`, `id`, `identifier`, `name`,
+ `runs_techniques_id`, `url`.
 
 You can also use annotations to express a dependency between `Variable`'s using
 the `var#<name>` annotation:

tests/test_backend.py

@@ -398,6 +398,10 @@ def sample(run, x: "mymdc#sample_name"):
  @Variable(title="Run type")
  def run_type(run, x: "mymdc#run_type"):
  return x
+
+ @Variable(title="Run Techniques")
+ def techniques(run, x: "mymdc#techniques"):
+ return ', '.join(t['name'] for t in x)
  """
  mymdc_ctx = mkcontext(mymdc_code)
 
@@ -416,11 +420,16 @@ def mock_get(url, headers, timeout):
  assert headers["X-API-key"] == "foo"
 
  if "proposals/by_number" in url:
- result = dict(runs=[dict(sample_id=1, experiment_id=1)])
+ result = dict(runs=[dict(id=1, sample_id=1, experiment_id=1)])
  elif "samples" in url:
  result = dict(name="mithril")
  elif "experiments" in url:
  result = dict(name="alchemy")
+ elif "/runs/" in url:
+ result = {'techniques': [
+ {'identifier': 'PaNET01168', 'name': 'SFX'},
+ {'identifier': 'PaNET01188', 'name': 'SAXS'},
+ ]}
 
  response = MagicMock()
  response.json.return_value = result
@@ -430,8 +439,10 @@ def mock_get(url, headers, timeout):
  with patch.object(requests, "get", side_effect=mock_get), \
  patch.object(ed.read_machinery, "find_proposal", return_value=tmp_path):
  results = results_create(mymdc_ctx)
+
  assert results.cells["sample"].data == "mithril"
  assert results.cells["run_type"].data == "alchemy"
+ assert results.cells["techniques"].data == "SFX, SAXS"
 
 
 def test_return_bool(mock_run, tmp_path):