Mpes elab metadata

.cspell/custom-dictionary.txt

@@ -98,13 +98,17 @@ dtype
 dtypes
 easimon
 ecalibdict
+elab
+elabapi
+elabid
 electronanalyser
 Elektronen
 endstation
 energycal
 energycalfolder
 ENERGYDISPERSION
 ENOSPC
+entityid
 equiscale
 Eref
 errorbar
@@ -265,6 +269,7 @@ ontop
 OPCPA
 openmp
 OPTICALDELAY
+orcid
 otherax
 packetcoders
 Pandoc
@@ -339,6 +344,7 @@ sdir
 segs
 setp
 sfile
+sharelink
 shutil
 Sixten
 sohail
@@ -384,6 +390,7 @@ ufunc
 unbinned
 uncategorised
 undoc
+userid
 utime
 varnames
 venv

docs/sed/loader.rst

@@ -29,6 +29,10 @@ MpesLoader
    :members:
    :undoc-members:
 
+.. automodule:: sed.loader.mpes.metadata
+   :members:
+   :undoc-members:
+
 FlashLoader
 ###################################################
 .. automodule:: sed.loader.flash.loader

pyproject.toml

@@ -31,6 +31,7 @@ classifiers = [
 dependencies = [
     "bokeh>=2.4.2",
     "dask>=2021.12.0,<2024.8",
+    "elabapi-python>=5.0",
     "fastdtw>=0.3.4",
     "h5py>=3.6.0",
     "ipympl>=0.9.1",
@@ -95,8 +96,8 @@ all = [
 
 [tool.coverage.report]
 omit = [
-    "config.py",
-    "config-3.py",
+    "./config.py",
+    "./config-3.py",
 ]
 
 [tool.ruff]

src/sed/core/config_model.py

@@ -321,6 +321,7 @@ class OffsetColumn(BaseModel):
 class MetadataModel(BaseModel):
     model_config = ConfigDict(extra="forbid")
 
+    elab_url: Optional[HttpUrl] = None
     archiver_url: Optional[HttpUrl] = None
     epics_pvs: Optional[Sequence[str]] = None
     fa_in_channel: Optional[str] = None

src/sed/core/processor.py

@@ -38,8 +38,8 @@
 from sed.io import to_tiff
 from sed.loader import CopyTool
 from sed.loader import get_loader
-from sed.loader.mpes.loader import get_archiver_data
 from sed.loader.mpes.loader import MpesLoader
+from sed.loader.mpes.metadata import get_archiver_data
 
 N_CPU = psutil.cpu_count()
 

src/sed/loader/mpes/loader.py

@@ -8,13 +8,9 @@
 import datetime
 import glob
 import io
-import json
 import os
 from collections.abc import Sequence
 from typing import Any
-from urllib.error import HTTPError
-from urllib.error import URLError
-from urllib.request import urlopen
 
 import dask
 import dask.array as da
@@ -27,6 +23,7 @@
 from sed.core.logging import set_verbosity
 from sed.core.logging import setup_logging
 from sed.loader.base.loader import BaseLoader
+from sed.loader.mpes.metadata import MetadataRetriever
 
 
 # Configure logging
@@ -580,34 +577,6 @@ def get_elapsed_time(
     return secs
 
 
-def get_archiver_data(
-    archiver_url: str,
-    archiver_channel: str,
-    ts_from: float,
-    ts_to: float,
-) -> tuple[np.ndarray, np.ndarray]:
-    """Extract time stamps and corresponding data from and EPICS archiver instance
-
-    Args:
-        archiver_url (str): URL of the archiver data extraction interface
-        archiver_channel (str): EPICS channel to extract data for
-        ts_from (float): starting time stamp of the range of interest
-        ts_to (float): ending time stamp of the range of interest
-
-    Returns:
-        tuple[np.ndarray, np.ndarray]: The extracted time stamps and corresponding data
-    """
-    iso_from = datetime.datetime.utcfromtimestamp(ts_from).isoformat()
-    iso_to = datetime.datetime.utcfromtimestamp(ts_to).isoformat()
-    req_str = archiver_url + archiver_channel + "&from=" + iso_from + "Z&to=" + iso_to + "Z"
-    with urlopen(req_str) as req:
-        data = json.load(req)
-        secs = [x["secs"] + x["nanos"] * 1e-9 for x in data[0]["data"]]
-        vals = [x["val"] for x in data[0]["data"]]
-
-    return (np.asarray(secs), np.asarray(vals))
-
-
 class MpesLoader(BaseLoader):
     """Mpes implementation of the Loader. Reads from h5 files or folders of the
     SPECS Metis 1000 (FHI Berlin)
@@ -729,6 +698,7 @@ def read_dataframe(
                 metadata=metadata,
             )
 
+        token = kwds.pop("token", None)
         channels = kwds.pop(
             "channels",
             self._config.get("dataframe", {}).get("channels", None),
@@ -777,6 +747,7 @@ def read_dataframe(
             metadata = self.gather_metadata(
                 files=self.files,
                 metadata=self.metadata,
+                token=token,
             )
         else:
             metadata = self.metadata
@@ -877,13 +848,15 @@ def gather_metadata(
         self,
         files: Sequence[str],
         metadata: dict = None,
+        token: str = None,
     ) -> dict:
         """Collect meta data from files
 
         Args:
             files (Sequence[str]): List of files loaded
             metadata (dict, optional): Manual meta data dictionary. Auto-generated
                 meta data are added to it. Defaults to None.
+            token (str, optional):: The elabFTW api token to use for fetching metadata
 
         Returns:
             dict: The completed metadata dictionary.
@@ -921,140 +894,21 @@ def gather_metadata(
             os.path.realpath(files[0]),
         )
 
-        logger.info("Collecting data from the EPICS archive...")
-        # Get metadata from Epics archive if not present already
-        epics_channels = self._config["metadata"]["epics_pvs"]
-
-        start = datetime.datetime.utcfromtimestamp(ts_from)
+        metadata_retriever = MetadataRetriever(self._config["metadata"], token)
 
-        channels_missing = set(epics_channels) - set(
-            metadata["file"].keys(),
+        metadata = metadata_retriever.fetch_epics_metadata(
+            ts_from=ts_from,
+            ts_to=ts_to,
+            metadata=metadata,
         )
-        for channel in channels_missing:
-            try:
-                _, vals = get_archiver_data(
-                    archiver_url=str(self._config["metadata"].get("archiver_url")),
-                    archiver_channel=channel,
-                    ts_from=ts_from,
-                    ts_to=ts_to,
-                )
-                metadata["file"][f"{channel}"] = np.mean(vals)
-
-            except IndexError:
-                metadata["file"][f"{channel}"] = np.nan
-                logger.info(
-                    f"Data for channel {channel} doesn't exist for time {start}",
-                )
-            except HTTPError as exc:
-                logger.warning(
-                    f"Incorrect URL for the archive channel {channel}. "
-                    "Make sure that the channel name and file start and end times are "
-                    "correct.",
-                )
-                logger.warning(f"Error code: {exc}")
-            except URLError as exc:
-                logger.warning(
-                    f"Cannot access the archive URL for channel {channel}. "
-                    f"Make sure that you are within the FHI network."
-                    f"Skipping over channels {channels_missing}.",
-                )
-                logger.warning(f"Error code: {exc}")
-                break
 
-        # Determine the correct aperture_config
-        stamps = sorted(
-            list(self._config["metadata"]["aperture_config"].keys()) + [start],
-        )
-        current_index = stamps.index(start)
-        timestamp = stamps[current_index - 1]  # pick last configuration before file date
-
-        # Aperture metadata
-        if "instrument" not in metadata.keys():
-            metadata["instrument"] = {"analyzer": {}}
-        metadata["instrument"]["analyzer"]["fa_shape"] = "circle"
-        metadata["instrument"]["analyzer"]["ca_shape"] = "circle"
-        metadata["instrument"]["analyzer"]["fa_size"] = np.nan
-        metadata["instrument"]["analyzer"]["ca_size"] = np.nan
-        # get field aperture shape and size
-        if {
-            self._config["metadata"]["fa_in_channel"],
-            self._config["metadata"]["fa_hor_channel"],
-        }.issubset(set(metadata["file"].keys())):
-            fa_in = metadata["file"][self._config["metadata"]["fa_in_channel"]]
-            fa_hor = metadata["file"][self._config["metadata"]["fa_hor_channel"]]
-            for key, value in self._config["metadata"]["aperture_config"][timestamp][
-                "fa_size"
-            ].items():
-                if value[0][0] < fa_in < value[0][1] and value[1][0] < fa_hor < value[1][1]:
-                    try:
-                        k_float = float(key)
-                        metadata["instrument"]["analyzer"]["fa_size"] = k_float
-                    except ValueError:  # store string if numeric interpretation fails
-                        metadata["instrument"]["analyzer"]["fa_shape"] = key
-                    break
-            else:
-                logger.warning("Field aperture size not found.")
-
-        # get contrast aperture shape and size
-        if self._config["metadata"]["ca_in_channel"] in metadata["file"]:
-            ca_in = metadata["file"][self._config["metadata"]["ca_in_channel"]]
-            for key, value in self._config["metadata"]["aperture_config"][timestamp][
-                "ca_size"
-            ].items():
-                if value[0] < ca_in < value[1]:
-                    try:
-                        k_float = float(key)
-                        metadata["instrument"]["analyzer"]["ca_size"] = k_float
-                    except ValueError:  # store string if numeric interpretation fails
-                        metadata["instrument"]["analyzer"]["ca_shape"] = key
-                    break
-            else:
-                logger.warning("Contrast aperture size not found.")
-
-        # Storing the lens modes corresponding to lens voltages.
-        # Use lens voltages present in first lens_mode entry.
-        lens_list = self._config["metadata"]["lens_mode_config"][
-            next(iter(self._config["metadata"]["lens_mode_config"]))
-        ].keys()
-
-        lens_volts = np.array(
-            [metadata["file"].get(f"KTOF:Lens:{lens}:V", np.nan) for lens in lens_list],
-        )
-        for mode, value in self._config["metadata"]["lens_mode_config"].items():
-            lens_volts_config = np.array([value[k] for k in lens_list])
-            if np.allclose(
-                lens_volts,
-                lens_volts_config,
-                rtol=0.005,
-            ):  # Equal upto 0.5% tolerance
-                metadata["instrument"]["analyzer"]["lens_mode"] = mode
-                break
-        else:
-            logger.warning(
-                "Lens mode for given lens voltages not found. "
-                "Storing lens mode from the user, if provided.",
-            )
-
-        # Determining projection from the lens mode
-        try:
-            lens_mode = metadata["instrument"]["analyzer"]["lens_mode"]
-            if "spatial" in lens_mode.split("_")[1]:
-                metadata["instrument"]["analyzer"]["projection"] = "real"
-                metadata["instrument"]["analyzer"]["scheme"] = "momentum dispersive"
-            else:
-                metadata["instrument"]["analyzer"]["projection"] = "reciprocal"
-                metadata["instrument"]["analyzer"]["scheme"] = "spatial dispersive"
-        except IndexError:
-            logger.warning(
-                "Lens mode must have the form, '6kV_kmodem4.0_20VTOF_v3.sav'. "
-                "Can't determine projection. "
-                "Storing projection from the user, if provided.",
-            )
-        except KeyError:
-            logger.warning(
-                "Lens mode not found. Can't determine projection. "
-                "Storing projection from the user, if provided.",
+        if self.runs:
+            metadata = metadata_retriever.fetch_elab_metadata(
+                runs=self.runs,
+                metadata=metadata,
             )
+        else:
+            logger.warning('Fetching elabFTW metadata only supported for loading from "runs"')
 
         return metadata