minimal modifications

one method fetches all
one filters textual uploads
one filters png and bmp images

.gitignore

@@ -1,6 +1,11 @@
-# ignora log di h5tojson e jsontoh5
+# ignores logs of h5tojson, jsontoh5
 *.log
 
+# ignores any output of main.py
+output/*.json
+output/*.h5
+output/*.nxs
+
 # ---> Python
 # Byte-compiled / optimized / DLL files
 __pycache__/

requirements.txt

@@ -1,2 +1,5 @@
 requests
 asyncio
+h5py
+pillow
+elabapi_python

src/APIHandler.py

@@ -1,39 +1,150 @@
-import requests
+import os, requests
+import elabapi_python as elabapi
+
 
 class APIHandler:
-    '''
-    Class to standardize the format of the headers of our http requests.
-    '''
+    """
+    Class which handles all interactions with the eLabFTW API.
+    It provides methods to retrieve data from the API and download attachments.
+    It relies minimally on the elabapi-python library, which is used only for downloading attachments
+    (since the API doesn't support downloading attachments AFAIK).
+
+    Args:
+        api_key:          A valid API key for the eLabFTW instance where the data is stored, with permissions to access the relevant entries.
+                          eLabFTW's API keys are well documented here: https://doc.elabftw.net/docs/usage/api/.
+                          If you don't have an API key and are uncapable of creating one, contact your eLabFTW administrator.
+                          Or RTFM and create one yourself, it's not that hard.
+        ELABFTW_API_URL:  Complete URL of the eLabFTW instance's root for the API endpoints.
+                          In full caps because it won't (shouldn't) be changed much.
+    """
+
     # TO-DO: remove static url.
-    def __init__(self, apikey="", ELABFTW_API_URL="https://elabftw.fisica.unina.it/api/v2"):
-        '''Init method, apikey suggested but not required (empty by default).'''
-        self.auth = {"Authorization" : apikey}
-        self.content = {"Content-Type" : "application/json"}
+    def __init__(
+        self, api_key="", ELABFTW_API_URL="https://elabftw.fisica.unina.it/api/v2"
+    ):
+        """Init method, apikey suggested but not required (empty by default)."""
+        self.api_key = api_key
+        self.auth = {"Authorization": api_key}
+        self.content = {"Content-Type": "application/json"}
         self.header = {**self.auth, **self.content}
         self.elaburl = ELABFTW_API_URL
-    def get_entry_from_elabid(self, elabid, entryType="items"):
-        '''
-        Method which returns a resource's raw data (as dictionary) from its elabid and entry type.
 
-        Entry type can be either "experiments" or "items".
-        '''
-        # TO-DO: validation and error handling on entryType value.
+    def get_entry_from_elabid(self, elabid, entryType="items"):
+        """
+        Returns raw data (as dictionary) from its elabid and entry type.
+
+        args:
+            elabid:     elabftw internal id of the selected resource.
+            entryType:  Resource type. Anything other than "experiments" or "items" WILL raise an error.
+        """
+        if entryType not in ["experiments", "items"]:
+            raise Exception(
+                "You can only download attachments from experiments or items."
+            )
+
         header = self.header
         response = requests.get(
-            headers = header,
-            url = f"{self.elaburl}/{entryType}/{elabid}",
-            verify=True
+            headers=header, url=f"{self.elaburl}/{entryType}/{elabid}", verify=True
         )
-        if response.status_code // 100 in [1,2,3]:
+
+        # Response is 5xx = server error:
+        if response.status_code // 100 == 5:
+            raise ConnectionError(
+                f"There's a problem on the server. Status code: {response.status_code}."
+            )
+
+        # Response is 4xx = client error:
+        if response.status_code // 100 == 4:
+            match response.status_code:
+                case 401 | 403:
+                    # Forbidden or unauthorized:
+                    raise ConnectionError(
+                        f"Invalid API key, authentication method or elabid. Check if an item with ID = {elabid} actually exists."
+                    )
+                case 404:
+                    # Lapalissian:
+                    raise ConnectionError(
+                        f"404: Not Found. This means there's no resource with this elabid (wrong elabid?) on your eLabFTW (wrong endpoint?)."
+                    )
+                case 400:
+                    # I genuinely have no idea:
+                    raise ConnectionError(
+                        f"400: Bad Request. This means the API endpoint you tried to reach is invalid. Did you tamper with the source code? If not, contact the developer."
+                    )
+                case _:
+                    # For some fucking reason, this is the only error I actually get from the API...
+                    raise ConnectionError(
+                        f"HTTP request failed with status code: {response.status_code} (NOTE: 4xx means user's fault)."
+                    )
+
         entry_data = response.json()
         return entry_data
-        elif response.status_code // 100 == 4:
-            match response.status_code:
-                case 401|403:
-                    raise ConnectionError(f"Invalid API key or authentication method.")
-                case 404:
-                    raise ConnectionError(f"404: Not Found. This means there's no resource with this elabid (wrong elabid?) on your eLabFTW (wrong endpoint?).")
-                case _:
-                    raise ConnectionError(f"HTTP request failed with status code: {response.status_code} (NOTE: 4xx means user's fault).")
-        else:
-            raise ConnectionError(f"There's a problem on the server. Status code: {response.status_code}.")
+
+    def download_attachments_data(self, elabid, entryType="experiments"):
+        """
+        Downloads attachments of a certain eLabFTW experiment (default) or item.
+        Only returns their binary data. Use method download_attachments_to_disk to save to file.
+        NOTE: Output is a dictionary where:
+            * The keys are the attachments' filenames;
+            * The values are the binary data for those attachments.
+
+        Args:
+            elabid:     eLabFTW internal ID of the selected resource.
+            entryType:  Resource type. Anything other than "experiments" or "items" WILL raise an error.
+        """
+        if entryType not in ["experiments", "items"]:
+            raise Exception(
+                "You can only download attachments from experiments or items."
+            )
+
+        config = elabapi.Configuration()
+        config.api_key["api_key"] = api_key
+        config.api_key_prefix["api_key"] = "Authorization"
+        config.host = self.elaburl
+        config.debug = False
+        api_client = elabapi.ApiClient(config)
+        api_client.set_default_header(
+            header_name="Authorization", header_value=self.api_key
+        )
+        uploads_api = elabapi.UploadsApi(api_client)
+
+        # Actual uploads (dictionary):
+        uploads = {
+            upload.real_name: uploads_api.read_upload(
+                entryType, elabid, upload.id, format="binary", _preload_content=False
+            ).data
+            for upload in uploads_api.read_uploads(entryType, elabid)
+        }
+
+        return uploads
+
+    def download_attachments_to_disk(
+        self,
+        elabid,
+        entryType="experiments",
+        dump_dir="output/attachments",
+        # persistent=True,
+    ):
+        """
+        Downloads attachments of a certain eLabFTW experiment (default) or item.
+        Downloads their binary data through method download_attachments_data and dumps it to dump_dir.
+
+        Args:
+            elabid:     eLabFTW internal ID of the selected resource.
+            entryType:  Resource type. Anything other than "experiments" or "items" WILL raise an error.
+            dump_dir:   Directory to which to save the attachments. Default is "output/attachments".
+            persistent: [Unused] Decides if the files will stay on disk after all operations are completed.
+                        If set to False, deletes the file upon exiting.
+        """
+
+        if entryType not in ["experiments", "items"]:
+            raise Exception(
+                "You can only download attachments from experiments or items."
+            )
+
+        uploads = download_attachments_data(elabid, entryType=entryType)
+        for file in uploads:
+            raw_data = uploads["file"]
+            with open(os.path.join(dump_dir, f"exp{elabid}-{file}"), "wb") as f:
+                f.write(raw_data)
+        return

src/classes.py

@@ -1,8 +1,9 @@
 import os, json, requests
 from APIHandler import APIHandler
 
+
 class Layer:
-    '''
+    """
     Layer(layer_data) - where layer_data is a Python dictionary.
 
     Meant to be used for eLabFTW Experiments of the "PLD Deposition" category.
@@ -10,21 +11,29 @@ class Layer:
     eLabFTW experiments contain most of the data required by the NeXus file - although every layer is on a different eLab entry;
     unfortunately, some data like the target's chemical formula must be retrieved through additional HTTP requests.
     Attributes 'target_elabid', 'rheed_system_elabid' and 'laser_system_elabid' contain elabid's for these resources, which are all items.
-    '''
+    """
+
     def __init__(self, layer_data):
         try:
+            self.elabid = layer_data["id"]
+            self.operator = layer_data["fullname"]
             self.extra = layer_data["metadata_decoded"]["extra_fields"]
-            self.layer_number = self.extra["Layer Progressive Number"]["value"] # integer
+            self.uploads = layer_data["uploads"]  # dict
+            self.layer_number = self.extra["Layer Progressive Number"][
+                "value"
+            ]  # integer
             self.target_elabid = self.extra["Target"]["value"]  # elabid
-            self.rheed_system_elabid = self.extra["RHEED System"]["value"] # elabid
             self.laser_system_elabid = self.extra["Laser System"]["value"]  # elabid
-            self.start_time = layer_data.get("created_at")
-            self.operator = layer_data.get("fullname")
-            self.description = layer_data.get("body")
+            self.chamber_elabid = self.extra["Chamber"]["value"]  # elabid
+            self.rheed_system_elabid = self.extra["RHEED System"]["value"]  # elabid
             self.deposition_time = self.extra["Duration"]["value"]
+            self.deposition_time_unit = self.extra["Duration"]["unit"]
             self.repetition_rate = self.extra["Repetition rate"]["value"]
+            self.repetition_rate_unit = self.extra["Repetition rate"]["unit"]
             try:
-                self.number_of_pulses = (float(self.deposition_time) * float(self.repetition_rate)).__floor__()
+                self.number_of_pulses = (
+                    float(self.deposition_time) * float(self.repetition_rate)
+                ).__floor__()
             except ValueError:
                 # Since number_of_pulses is required, if it can't be calculated raise error:
                 raise ValueError("""
@@ -32,16 +41,33 @@ class Layer:
     This has to be an error, since these fields are required by the NeXus standard.
     Please edit your eLabFTW entry and retry.
                 """)
-            self.temperature = self.extra["Heater temperature"]["value"] # Note: this field used to have a trailing space in its name
-            self.process_pressure = self.extra["Process pressure"]["value"] # Note: this field used to have a trailing space in its name
+            self.temperature = self.extra["Heater temperature"][
+                "value"
+            ]  # Note: this field used to have a trailing space in its name
+            self.temperature_unit = self.extra["Heater temperature"]["unit"]
+            self.process_pressure = self.extra["Process pressure"][
+                "value"
+            ]  # Note: this field used to have a trailing space in its name
+            self.process_pressure_unit = self.extra["Process pressure"]["unit"]
             self.heating_method = self.extra["Heating Method"]["value"]
             self.layer_thickness = self.extra["Thickness"]["value"]
+            self.layer_thickness_unit = self.extra["Thickness"]["unit"]
             self.buffer_gas = self.extra["Buffer gas"]["value"]
             self.heater_target_distance = self.extra["Heater-target distance"]["value"]
-            self.laser_fluence = self.extra["Laser Intensity"]["value"] # here fluence = intensity
+            self.heater_target_distance_unit = self.extra["Heater-target distance"][
+                "unit"
+            ]
+            self.laser_fluence = self.extra["Laser Intensity"][
+                "value"
+            ]  # here fluence = intensity
+            self.laser_fluence_unit = "J/(s cm^2)"
             self.laser_spot_area = self.extra["Spot Area"]["value"]
+            self.laser_spot_area_unit = "mm^2"
             try:
-                self.laser_energy = (float(self.laser_fluence) * float(self.laser_spot_area)).__round__(3)
+                self.laser_energy = (
+                    float(self.laser_fluence) * float(self.laser_spot_area) / 100
+                ).__round__(3)
+                self.laser_energy_unit = "J/s"
             except ValueError:
                 # Since laser_energy is NOT required, if it can't be calculated warn user but allow the software to continue execution:
                 print("""
@@ -50,53 +76,166 @@ class Layer:
     Setting Laser Energy to NoneType.
                 """)
                 # Placeholder
-                self.laser_energy = None
+                self.laser_energy = "N/A"
+                self.laser_energy_unit = "J/s"
             # Laser rasternig section
-            self.laser_rastering_geometry = self.extra["Laser Rastering Geometry"]["value"]
-            self.laser_rastering_positions = self.extra["Laser Rastering Position"]["value"]
-            self.laser_rastering_velocities = self.extra["Laser Rastering Speed"]["value"]
+            self.laser_rastering_geometry = self.extra["Laser Rastering Geometry"][
+                "value"
+            ]
+            self.laser_rastering_positions = self.extra["Laser Rastering Position"][
+                "value"
+            ]
+            self.laser_rastering_velocities = self.extra["Laser Rastering Speed"][
+                "value"
+            ]
             # Pre annealing section
             self.pre_annealing_ambient_gas = self.extra["Buffer gas Pre"]["value"]
             self.pre_annealing_pressure = self.extra["Process pressure Pre"]["value"]
-            self.pre_annealing_temperature = self.extra["Heater temperature Pre"]["value"]
+            self.pre_annealing_temperature = self.extra["Heater temperature Pre"][
+                "value"
+            ]
             self.pre_annealing_duration = self.extra["Duration Pre"]["value"]
+            self.pre_annealing_pressure_unit = self.extra["Process pressure Pre"][
+                "unit"
+            ]
+            self.pre_annealing_temperature_unit = self.extra["Heater temperature Pre"][
+                "unit"
+            ]
+            self.pre_annealing_duration_unit = self.extra["Duration Pre"]["unit"]
             # Post annealing section
             self.post_annealing_ambient_gas = self.extra["Buffer gas PA"]["value"]
             self.post_annealing_pressure = self.extra["Process pressure PA"]["value"]
-            self.post_annealing_temperature = self.extra["Heater temperature PA"]["value"]
+            self.post_annealing_temperature = self.extra["Heater temperature PA"][
+                "value"
+            ]
             self.post_annealing_duration = self.extra["Duration PA"]["value"]
+            self.post_annealing_pressure_unit = self.extra["Process pressure PA"][
+                "unit"
+            ]
+            self.post_annealing_temperature_unit = self.extra["Heater temperature PA"][
+                "unit"
+            ]
+            self.post_annealing_duration_unit = self.extra["Duration PA"]["unit"]
+
             # Rejected but suggested by the NeXus standard:
-            #self.laser_rastering_coefficients = None
+            # self.laser_rastering_coefficients = None
         except KeyError as k:
             # Some keys are not required and can be called through the .get() method - which is permissive and allows null values;
             # Other keys are required so if they can't be called (invalid or null) raise error and stop execution of the program:
-            raise KeyError(f"The provided dictionary lacks a \"{k}\" key. Check the deposition layer entry on eLabFTW and make sure you used the correct Experiment template.")
+            raise KeyError(
+                f'The provided dictionary lacks a "{k}" key. Check the deposition layer entry on eLabFTW and make sure you used the correct Experiment template.'
+            )
+        # Optional
+        self.start_time = layer_data.get("created_at") or None
+        self.description = layer_data.get("body") or None
+
+    def get_instruments(self, api_key):
+        raw_lasersys_data = APIHandler(api_key).get_entry_from_elabid(
+            self.laser_system_elabid, entryType="items"
+        )
+        raw_chamber_data = APIHandler(api_key).get_entry_from_elabid(
+            self.chamber_elabid, entryType="items"
+        )
+        raw_rheedsys_data = APIHandler(api_key).get_entry_from_elabid(
+            self.rheed_system_elabid, entryType="items"
+        )
+        instruments_used = {
+            "laser_system": raw_lasersys_data.get("title") or None,
+            "deposition_chamber": raw_chamber_data.get("title") or None,
+            "rheed_system": raw_rheedsys_data.get("title") or None,
+        }
+        return instruments_used
+
+    def list_attachments(self):
+        """
+        Returns a dictionary of all the attachments linked to the layer, where:
+            * Each key is the attachment's elabid;
+            * Each value is a dictionary containing the attachment's filename, hashname and related experiment elabid (= self.elabid).
+
+        Data is already in layer_data, so the API key is unrequired. Same goes for:
+            * fetch_textual_uploads() - no arguments;
+            * fetch_images() - no arguments.
+        """
+        # Remember: Layers are experiments, so we only need to look for attachments in the experiment endpoint.
+        attachments = {
+            attachment["id"]: {
+                "filename": attachment["real_name"],
+                "hashname": attachment["long_name"],
+                "related_experiment": attachment["item_id"],
+            }
+            for attachment in self.uploads
+        }
+        return attachments
+
+    def fetch_textual_uploads(self):
+        """
+        Starting from the list of attachments, filters out and returns a list of the textual uploads linked to the layer, which can be either plain text, csv, tsv etc.
+        Returns only their names, so that the user may select which one to import into the NeXus file as a dataset.
+
+        It only looks for .txt, .csv and .tsv files, although it could be easily modified to include other formats.
+        It is also file extension-sensitive, so anything not ending with .txt, .csv or .tsv won't be retrieved.
+        That's because the API (v5.3.11) doesn't provide MIME Type or similar metadata on the attachments, so the only way to know if an attachment is an image or not is through its filename.
+        """
+        attachments = self.list_attachments()
+        textual_uploads = {
+            attachment: attachments[attachment]
+            for attachment in attachments
+            if attachments[attachments]["filename"][-4:] in (".txt", ".csv", ".tsv")
+        }
+        return textual_uploads
+
+    def fetch_images(self):
+        """
+        Starting from the list of attachments, filters out and returns a Starting from the list of attachments, filters out and returns a list of all the (PNG or BMP) images attached to the layer.
+        Hopefully one of them is a RHEED pattern.
+        Returns only their names, so that the user may select which one to import into the NeXus file as a RHEED acquisition.
+
+        It only looks for .png and .bmp files, although it could be easily modified to include other formats.
+        It is also file extension-sensitive, so anything not ending with .png or .bmp won't be retrieved, even if it's an actual image.
+        That's because the API (v5.3.11) doesn't provide MIME Type or similar metadata on the attachments, so the only way to know if an attachment is an image or not is through its filename.
+        """
+        attachments = self.list_attachments()
+        pictures = {
+            attachment: attachments[attachment]
+            for attachment in attachments
+            if attachments[attachments]["filename"][-4:] in (".png", ".bmp")
+        }
+        return pictures
+
 
 class Entrypoint:
-    '''
+    """
     Entrypoint(sample_data) - where sample_data is a Python dictionary.
 
     Meant to be used for eLabFTW Resources of the "Sample" category.
 
     The entrypoint is the starting point of the process of resolving the data chain.
     The entrypoint must be a dictionary containing the data of a sample, created directly from the JSON of the item endpoint on eLabFTW - which can be done through the function get_entry_from_elabid.
-    '''
+    """
+
     def __init__(self, sample_data):
         try:
             self.extra = sample_data["metadata_decoded"]["extra_fields"]
             self.linked_items = sample_data["items_links"]  # dict
             self.batch_elabid = self.extra["Substrate batch"]["value"]  # elabid
             self.linked_experiments = sample_data["related_experiments_links"]  # dict
-            self.linked_experiments_elabid = [ i["entityid"] for i in self.linked_experiments ] # list of elabid
+            self.linked_experiments_elabid = [
+                i["entityid"] for i in self.linked_experiments
+            ]  # list of elabid
         except KeyError as k:
             # Some keys are not required and can be called through the .get() method - which is permissive and allows null values;
             # Other keys are required so if they can't be called (invalid or null) raise error and stop execution of the program:
-            raise KeyError(f"The provided dictionary lacks a \"{k}\" key. Check the sample entry on eLabFTW and make sure you used the correct Resource template.")
+            raise KeyError(
+                f'The provided dictionary lacks a "{k}" key. Check the sample entry on eLabFTW and make sure you used the correct Resource template.'
+            )
         # Non-required attributes:
-        self.name = sample_data.get("title") or None # error prevention is more important than preventing empty fields here
+        self.name = (
+            sample_data.get("title") or None
+        )  # error prevention is more important than preventing empty fields here
+
 
 class Material:
-    '''
+    """
     Material(material_data) - where material_data is a Python dictionary.
 
     Meant to be used for eLabFTW Resources of either the "PLD Target" or the "Substrate" categories.
@@ -105,62 +244,87 @@ class Material:
         * Name and formula;
         * Shape and dimensions;
         * Misc.
-    '''
+    """
+
     def __init__(self, material_data):
         try:
             self.name = material_data["title"]  # required
             self.extra = material_data["metadata_decoded"]["extra_fields"]
             self.compound_elabid = self.extra["Compound"]["value"]
-            self.dimensions = self.extra["Size"]["value"]
+            self.dimensions = str(
+                self.extra["Size"]["value"]
+            )  # strings have a .count() method
+            if self.dimensions.count("mm") == 2:
+                self.dimensions_unit = "mm x mm"
+            elif self.dimensions[-1] == '"':
+                self.dimensions_unit = "inches"
+            else:
+                self.dimensions_unit = None
         except KeyError as k:
             # Some keys are not required and can be called through the .get() method - which is permissive and allows null values;
             # Other keys are required so if they can't be called (invalid or null) raise error and stop execution of the program:
-            raise KeyError(f"The provided dictionary lacks a \"{k}\" key. Check the target/substrate entry on eLabFTW and make sure you used the correct Resource template.")
+            raise KeyError(
+                f'The provided dictionary lacks a "{k}" key. Check the target/substrate entry on eLabFTW and make sure you used the correct Resource template.'
+            )
+
     def get_compound_data(self, apikey):
-        raw_compound_data = APIHandler(apikey).get_entry_from_elabid(self.compound_elabid, entryType="items")
+        raw_compound_data = APIHandler(apikey).get_entry_from_elabid(
+            self.compound_elabid, entryType="items"
+        )
         name = raw_compound_data["title"]
         extra = raw_compound_data["metadata_decoded"]["extra_fields"]
         formula = extra.get("Chemical formula")
-        cas = extra.get("CAS number ") or { "value": None }
+        cas = extra.get("CAS number ") or {"value": None}
         compound_data = {
-            "name" : name,
-            "chemical_formula" : formula.get("value"),
-            "cas_number" : cas.get("value")
+            "name": name,
+            "chemical_formula": formula.get("value"),
+            "cas_number": cas.get("value"),
         }
         return compound_data
+
     def get_compound_formula(self, apikey):
         formula = self.get_compound_data(apikey).get("chemical_formula")
         return formula
 
+
 class Substrate(Material):
     def __init__(self, material_data):
         super().__init__(material_data)
         try:
             self.orientation = self.extra["Orientation"]["value"]
             self.miscut_angle = self.extra["Miscut Angle"]["value"]
+            self.miscut_angle_unit = self.extra["Miscut Angle"]["unit"]
             self.miscut_direction = self.extra["Miscut Direction"]["value"]
             # Not present (yet) on eLabFTW for Substrates:
-            self.thickness = None #self.extra["Thickness"]["value"]
+            self.thickness = ""  # self.extra["Thickness"]["value"]
+            self.thickness_unit = "μm"  # self.extra["Thickness"]["unit"]
             self.surface_treatment = self.extra["Surface treatment"]["value"]
             self.manufacturer = self.extra["Supplier"]["value"]
             self.batch_id = self.extra["Batch ID"]["value"]
         except KeyError as k:
-            raise KeyError(f"The provided dictionary lacks a \"{k}\" key - which is specific for substrates. Check the {self.name} substrate entry on eLabFTW and make sure you used the correct Resource template.")
+            raise KeyError(
+                f'The provided dictionary lacks a "{k}" key - which is specific for substrates. Check the {self.name} substrate entry on eLabFTW and make sure you used the correct Resource template.'
+            )
+
 
 class Target(Material):
     def __init__(self, material_data):
         super().__init__(material_data)
         try:
             self.thickness = self.extra["Thickness"]["value"]
+            self.thickness_unit = self.extra["Thickness"]["unit"]
             self.shape = self.extra["shape"]["value"]
             self.solid_form = self.extra["Solid form"]["value"]
             self.manufacturer = self.extra["Supplier"]["value"]
         except KeyError as k:
-            raise KeyError(f"The provided dictionary lacks a \"{k}\" key - which is specific for PLD targets. Check the {self.name} target entry on eLabFTW and make sure you used the correct Resource template.")
+            raise KeyError(
+                f'The provided dictionary lacks a "{k}" key - which is specific for PLD targets. Check the {self.name} target entry on eLabFTW and make sure you used the correct Resource template.'
+            )
+        # Non-required attributes:
+        self.description = material_data.get("body") or ""
 
 
-
-if __name__=="__main__":
-    head = Header("MyApiKey-123456789abcdef")
+if __name__ == "__main__":
+    head = APIHandler("MyApiKey-123456789abcdef")
     print(f"Example header:\n\t{head.header}\n")
     print("Warning: you're not supposed to be running this as the main program.")

src/functions.py

@@ -1,62 +0,0 @@
-"""
-Currently unused!
-"""
-import json, requests
-from APIHandler import APIHandler
-
-def get_entry_from_elabid(elabid, entryType="items"):
-    '''
-    Function which returns entrypoint data (as dictionary) from its elabid.
-    '''
-    header = APIHandler(apikey).dump
-    response = requests.get(
-        headers = header,
-        url = f"{ELABFTW_API_URL}/{entryType}/{elabid}",
-        verify=True
-    )
-    if response.status_code // 100 in [2,3]:
-        entry_data = response.json()
-        return entry_data
-    else:
-        raise ConnectionError(f"HTTP request failed with status code: {response.status_code}.")
-
-def get_sample_layers_data(elabid):
-    '''
-    Return the following data from every eLabFTW experiment linked
-    to a certain sample, identified by elabid.
-
-    - Title of the experiment
-    - Category (should check it's "PLD Deposition")
-    - Layer number - if present (PLD depositions)
-    - Deposition time - returns error if not present
-    - Repetition rate - returns error if not present
-    '''
-    # header = {
-    #     "Authorization": apikey,
-    #     "Content-Type": "application/json"
-    # }
-    sample_data = requests.get(
-        headers = header,
-        url = f"https://elabftw.fisica.unina.it/api/v2/items/{elabid}",
-        verify=True
-    ).json()
-    related_experiments = sample_data["related_experiments_links"]
-    result = []
-    for exp in related_experiments:
-        experiment_data = requests.get(
-            headers = header,
-            url = f"https://elabftw.fisica.unina.it/api/v2/experiments/{exp.get("entityid")}",
-            verify=True
-        ).json()
-        extra = experiment_data["metadata_decoded"]["extra_fields"]
-        result.append(
-            {"title": exp.get("title"),
-             "layer_number": extra.get("Layer Progressive Number").get("value"),
-             "category": exp.get("category_title"),
-             "deposition_time": extra.get("Duration").get("value"),
-             "repetition_rate": extra.get("Repetition rate").get("value")}
-        )
-    return result
-
-if __name__=="__main__":
-    print("Warning: you're not supposed to be running this as the main program.")

src/main.py

@@ -1,38 +1,51 @@
-import os, json, requests
+#!/usr/bin/env python3
+import os, json, requests, h5py
+import numpy as np
 from getpass import getpass
 from APIHandler import APIHandler
 from classes import *
+from PIL import Image
+# from schema import pld_deposition
 
 
 def call_entrypoint_from_elabid(elabid):
-    '''
+    """
     Calls an entrypoint sample from eLabFTW using its elabid, then returns an object of the Entrypoint class.
 
     If the entry is not a sample (category_title not matching exactly "Sample") returns ValueError.
-    '''
+    """
     try:
-        sample_data = APIHandler(apikey).get_entry_from_elabid(elabid, entryType="items")
+        sample_data = APIHandler(apikey).get_entry_from_elabid(
+            elabid, entryType="items"
+        )
         if not sample_data.get("category_title") == "Sample":
-            raise ValueError("The resource you selected is not a sample, therefore it can't be used as an entrypoint.")
+            raise ValueError(
+                "The resource you selected is not a sample, therefore it can't be used as an entrypoint."
+            )
         sample_object = Entrypoint(sample_data)
     except ConnectionError as e:
         raise ConnectionError(e)
     return sample_object  # Entrypoint-class object
 
+
 def call_material_from_elabid(elabid):
-    '''
+    """
     Calls a material from eLabFTW using its elabid, then returns an object of the Material class.
 
     If the entry is neither a PLD Target or a Substrate batch returns ValueError. Such entries always have a category_title key with its value matching exactly "PLD Target" or "Substrate".
     Because of an old typo, the value "Subtrate" (second 's' is missing) is also accepted.
-    '''
+    """
     try:
-        material_data = APIHandler(apikey).get_entry_from_elabid(elabid, entryType="items")
+        material_data = APIHandler(apikey).get_entry_from_elabid(
+            elabid, entryType="items"
+        )
         material_category = material_data.get("category_title")
         # TO-DO: correct this typo on elabftw: Subtrate → Substrate.
         if not material_category in ["PLD Target", "Substrate", "Subtrate"]:
             print(f"Category of the resource: {material_category}.")
-            raise ValueError(f"The referenced resource (elabid = {elabid}) is not a material.")
+            raise ValueError(
+                f"The referenced resource (elabid = {elabid}) is not a material."
+            )
         elif material_category == "PLD Target":
             material_object = Target(material_data)
         else:
@@ -41,128 +54,692 @@ def call_material_from_elabid(elabid):
         raise ConnectionError(e)
     return material_object  # Material-class object
 
+
 def call_layers_from_list(elabid_list):
-    '''
+    """
     Calls a list of (PLD deposition) experiments from eLabFTW using their elabid - which means the input must be a list of integers instead of a single one - then returns a list of Layer-class objects.
 
     If one of the entries is not related to a deposition layer (category_title not matching exactly "PLD Deposition") that entry is skipped, with no error raised.
-    '''
+    """
     list_of_layers = []
     for elabid in elabid_list:
         try:
-            layer_data = APIHandler(apikey).get_entry_from_elabid(elabid, entryType="experiments")
+            layer_data = APIHandler(apikey).get_entry_from_elabid(
+                elabid, entryType="experiments"
+            )
             if not layer_data.get("category_title") == "PLD Deposition":
                 continue
             layer_object = Layer(layer_data)
             list_of_layers.append(layer_object)
         except ConnectionError as e:
-            nums = [ layer.layer_number for layer in list_of_layers ]
+            nums = [layer.layer_number for layer in list_of_layers]
             nums.sort()
             print(f"LIST OF THE LAYERS PROCESSED (unordered):\n" + str(nums))
-            raise ConnectionError(f"An error occurred while fetching the experiment with elabid = {elabid}:\n" +
-                str(e) + f"\nPlease solve the problem before retrying." + "\n\n" +
-                f"Last resource attempted to call: {ELABFTW_API_URL}/experiments/{elabid}"
+            raise ConnectionError(
+                f"An error occurred while fetching the experiment with elabid = {elabid}:\n"
+                + str(e)
+                + f"\nPlease solve the problem before retrying."
+                + "\n\n"
+                + f"Last resource attempted to call: {ELABFTW_API_URL}/experiments/{elabid}"
             )
     return list_of_layers  # list of Layer-class objects
 
+
 def chain_entrypoint_to_batch(sample_object):
-    '''
+    """
     Takes an Entrypoint-class object, looks at its .batch_elabid attribute and returns a Material-class object containing data on the substrate batch associated to the starting sample.
 
     Dependency: call_material_from_elabid.
-    '''
+    """
     material_elabid = sample_object.batch_elabid
     material_object = call_material_from_elabid(material_elabid)
     return material_object
 
+
 def chain_entrypoint_to_layers(sample_object):
-    '''
+    """
     Takes an Entrypoint-class object, looks at its .linked_experiments_elabid attribute (list) and returns a list of Layer-class objects containing data on the deposition layers associated to the starting sample - using the function call_layers_from_list.
 
     The list is sorted by progressive layer number (layer_number attribute).
 
     Dependency: call_layers_from_list.
-    '''
-    linked_experiments_elabid = sample_object.linked_experiments_elabid # list of elabid
+    """
+    linked_experiments_elabid = (
+        sample_object.linked_experiments_elabid
+    )  # list of elabid
     layer_object_list = call_layers_from_list(linked_experiments_elabid)
     layer_object_list.sort(key=lambda x: x.layer_number)
     return layer_object_list
 
+
 def chain_layer_to_target(layer_object):
-    '''
+    """
     Takes a Layer-class object, looks at its .target_elabid attribute and returns a Material-class object containing data on the PLD target used in the deposition of said layer.
 
     Dependency: call_material_from_elabid.
-    '''
+    """
     target_elabid = layer_object.target_elabid
     material_object = call_material_from_elabid(target_elabid)
     return material_object
 
-#sample_object = call_entrypoint_from_elabid(elabid)
-#from_entrypoint_to_material(sample_object)
+
+def deduplicate_instruments_from_layers(layers):
+    """
+    Takes a list of Layer-class objects and for each layer gets the instruments used (laser, depo chamber and RHEED), returns dictionary with one item per category. This means that if more layers share the same instruments it returns a dictionary with just their names as strings (no lists or sub-dictionaries).
+
+    If different layers have different instruments (e.g. laser systems) the user is prompted to only select one.
+    """
+    lasers = []
+    chambers = []
+    rheeds = []
+    elegant_dict = {}
+    for lyr in layers:
+        instruments = lyr.get_instruments(apikey)
+        lasers.append(instruments["laser_system"])
+        chambers.append(instruments["deposition_chamber"])
+        rheeds.append(instruments["rheed_system"])
+        elegant_dict[f"layer_{lyr.layer_number}"] = {
+            "laser_system": instruments["laser_system"],
+            "deposition_chamber": instruments["deposition_chamber"],
+            "rheed_system": instruments["rheed_system"],
+        }
+    ded_lasers = list(set(lasers))
+    ded_chambers = list(set(chambers))
+    ded_rheeds = list(set(rheeds))
+    elegant_dict["multilayer"] = {
+        # Keep key names human readable since they're used in the messages of the following errors
+        "laser_system": ", ".join(ded_lasers),
+        "deposition_chamber": ", ".join(ded_chambers),
+        "rheed_system": ", ".join(ded_rheeds),
+    }  # dictionary's name is a joke
+    # updated_dict = {} # use this for containing the final dataset
+    # for ded in elegant_dict:
+    #     if len(elegant_dict[ded]) == 0:
+    #         # if len of list is 0 - empty list - raise error
+    #         raise IndexError(f"Missing data: no Laser System, Chamber and/or RHEED System is specified in any of the Deposition-type experiments related to this sample. Fix this on eLabFTW before retrying. Affected list: {ded}.")
+    #     elif len(elegant_dict[ded]) > 1:
+    #         # if len of list is > 1 - too many values - allow the user to pick one
+    #         print("Warning: different instruments have been used for different layers - which is currently not allowed.")
+    #         # there's a better way to do this but I can't remember now for the life of me...
+    #         i = 0
+    #         while i < len(elegant_dict[ded]):
+    #             print(f"{i} - {elegant_dict[ded][i]}")
+    #             i += 1
+    #         ans = None
+    #         while not type(ans) == int or not ans in range(0, len(elegant_dict[ded])):
+    #             ans = input("Please pick one of the previous (0, 1, ...) [default = 0]: ") or "0"
+    #             if ans.isdigit():
+    #                 ans = int(ans)
+    #             continue # unnecessary?
+    #         updated_dict[ded] = elegant_dict[ded][ans]
+    #     elif elegant_dict[ded][0] in ["", 0, None]:
+    #         # if len is 1 BUT value is "", 0 or None raise error
+    #         raise ValueError(f"Missing data: a Laser System, Chamber and/or RHEED System which is specified across all the Deposition-type experiments related to this sample is either empty or invalid. Fix this on eLabFTW before retrying. Affected list: {ded}.")
+    #     else:
+    #         # if none of the previous (only 1 value), that single value is used
+    #         updated_dict[ded] = elegant_dict[ded][0]
+    # instruments_used_dict = {
+    #     "laser_system": updated_dict["Laser Systems"],
+    #     "deposition_chamber": updated_dict["Deposition Chamber"],
+    #     "rheed_system": updated_dict["RHEED Systems"],
+    # }
+    return elegant_dict
+
+    ### OLD CODE
+    # if 0 in [ len(i) for i in elegant_list ]:
+    #     # i.e. if length of one of the lists in elegant_list is zero (missing data):
+    #     raise IndexError("Missing data: no Laser System, Chamber and/or RHEED System is specified in any of the Deposition-type experiments related to this sample.")
+    # if not all([ len(i) == 1 for i in elegant_list ]):
+    #     print("Warning: different instruments have been used for different layers - which is currently not allowed.")
+    #     # for every element in elegant list check if len > 1 and if it is
+    #     print("Selecting the first occurence for every category...")
+    ###
+    # lasers = { f"layer_{lyr.layer_number}": lyr.laser_system for lyr in layers }
+    # chambers = { f"layer_{lyr.layer_number}": lyr.deposition_chamber for lyr in layers }
+    # rheeds = { f"layer_{lyr.layer_number}": lyr.rheed_system for lyr in layers }
+    # instruments_used_dict = {
+    #     "laser_system": lasers,
+    #     "deposition_chamber": chambers,
+    #     "rheed_system": rheeds,
+    # }
 
 
-if __name__=="__main__":
-    print(f"=======================\n===== DEBUG MODE! =====\n=======================\n")
+def analyse_rheed_data(data):
+    """
+    Takes the content of a tsv file and returns a dictionary with timestamps and intensities.
+    The file should contain a 2D array composed of 4 columns - where the first column is a timestamp and the other three are RHEED intensities - and an unspecified number of rows.
+
+    -----
+    Time    Layer1_Int1     Layer1_Int2     Layer1_Int3
+    -----
+
+    Distinct ValueErrors are raised if:
+    * The array is not 2-dimensional;
+    * The total number of columns does not equate exactly 1+3 (= 4).
+
+    Time is expressed in seconds, intensities are adimensional on 8 bits (min. 0, max. 255).
+
+    # TO-DO: complete this description...
+    Written with help from DeepSeek.
+    """
+    # Verifying the format of the input file:
+    if data.ndim != 2:
+        raise ValueError(
+            f"Unexpected trace format: expected 2D array, got ndim = {data.ndim}."
+        )
+    n_cols = data.shape[1]  # 0 = rows, 1 = columns
+    if n_cols > 4:
+        print(
+            f"Warning! The input file (for Realtime Window Analysis) has {n_cols - 4} more than needed.\nOnly 4 columns will be considered - with the first representing time and the others representing RHEED intensities."
+        )
+    if n_cols < 4:
+        raise ValueError(
+            f"Insufficient number of columns: expected 4, got n_cols = {n_cols}."
+        )
+    n_time_points = data.shape[0]
+
+    # Get time (all rows of col 0) as Float64:
+    time = data[:, 0].astype(
+        np.float64, copy=False
+    )  # copy=False suggested by LLM for mem. eff.
+
+    # Get intensities (all rows of cols 1,2,3) as Float32:
+    intensities = data[:, 1:4].astype(np.float32, copy=False)
+
+    return {
+        "time": np.transpose(time),
+        "intensity": np.transpose(intensities),
+    }
+
+
+def make_nexus_schema_dictionary(substrate_object, layers):
+    """
+    Main function, takes all the other functions to reconstruct the full dataset. Takes a Substrate-class object (output of the chain_entrypoint_to_batch() function) and a list of Layer-class objects (output of the chain_entrypoint_to_layers() function), returns dictionary with the same schema as the NeXus standard for PLD fabrications.
+    """
+    instruments = deduplicate_instruments_from_layers(layers)
+    pld_fabrication = {
+        "sample": {
+            "substrate": {
+                "name": substrate_object.name,
+                "chemical_formula": substrate_object.get_compound_formula(apikey),
+                "orientation": substrate_object.orientation,
+                "miscut_angle": {
+                    "value": substrate_object.miscut_angle,
+                    "units": substrate_object.miscut_angle_unit,
+                },
+                "miscut_direction": substrate_object.miscut_direction,
+                "thickness": {
+                    "value": substrate_object.thickness,
+                    "units": substrate_object.thickness_unit,
+                },
+                "dimensions": substrate_object.dimensions,
+                "surface_treatment": substrate_object.surface_treatment,
+                "manufacturer": substrate_object.manufacturer,
+                "batch_id": substrate_object.batch_id,
+            },
+            "multilayer": {},
+        },
+        "instruments_used": instruments["multilayer"],
+    }
+    multilayer = pld_fabrication["sample"]["multilayer"]
+    for layer in layers:
+        name = "layer_" + layer.layer_number
+        target_object = chain_layer_to_target(layer)
+        target_dict = {
+            "name": target_object.name,
+            "chemical_formula": target_object.get_compound_formula(apikey),
+            "description": target_object.description,
+            "shape": target_object.shape,
+            "dimensions": target_object.dimensions,
+            "thickness": {
+                "value": target_object.thickness,
+                "units": target_object.thickness_unit,
+            },
+            "solid_form": target_object.solid_form,
+            "manufacturer": target_object.manufacturer,
+            "batch_id": target_object.name,
+            # TO-DO: currently not available:
+        }
+        multilayer[name] = {
+            "target": target_dict,
+            "start_time": layer.start_time,
+            "operator": layer.operator,
+            "description": layer.description,
+            "number_of_pulses": layer.number_of_pulses,
+            "deposition_time": {
+                "value": layer.deposition_time,
+                "units": layer.deposition_time_unit,
+            },
+            "temperature": {
+                "value": layer.temperature,
+                "units": layer.temperature_unit,
+            },
+            "heating_method": layer.heating_method,
+            "layer_thickness": {
+                "value": layer.layer_thickness,
+                "units": layer.layer_thickness_unit,
+            },
+            "buffer_gas": layer.buffer_gas,
+            "process_pressure": {
+                "value": layer.process_pressure,
+                "units": layer.process_pressure_unit,
+            },
+            "heater_target_distance": {
+                "value": layer.heater_target_distance,
+                "units": layer.heater_target_distance_unit,
+            },
+            "repetition_rate": {
+                "value": layer.repetition_rate,
+                "units": layer.repetition_rate_unit,
+            },
+            "laser_fluence": {
+                "value": layer.laser_fluence,
+                "units": layer.laser_fluence_unit,
+            },
+            "laser_spot_area": {
+                "value": layer.laser_spot_area,
+                "units": layer.laser_spot_area_unit,
+            },
+            "laser_energy": {
+                "value": layer.laser_energy,
+                "units": layer.laser_energy_unit,
+            },
+            "laser_rastering": {
+                "geometry": layer.laser_rastering_geometry,
+                "positions": layer.laser_rastering_positions,
+                "velocities": layer.laser_rastering_velocities,
+            },
+            "pre_annealing": {
+                "ambient_gas": layer.pre_annealing_ambient_gas,
+                "pressure": {
+                    "value": layer.pre_annealing_pressure,
+                    "units": layer.pre_annealing_pressure_unit,
+                },
+                "temperature": {
+                    "value": layer.pre_annealing_temperature,
+                    "units": layer.pre_annealing_temperature_unit,
+                },
+                "duration": {
+                    "value": layer.pre_annealing_duration,
+                    "units": layer.pre_annealing_duration_unit,
+                },
+            },
+            "post_annealing": {
+                "ambient_gas": layer.post_annealing_ambient_gas,
+                "pressure": {
+                    "value": layer.post_annealing_pressure,
+                    "units": layer.post_annealing_pressure_unit,
+                },
+                "temperature": {
+                    "value": layer.post_annealing_temperature,
+                    "units": layer.post_annealing_temperature_unit,
+                },
+                "duration": {
+                    "value": layer.post_annealing_duration,
+                    "units": layer.post_annealing_duration_unit,
+                },
+            },
+            "instruments_used": instruments[name],
+        }
+    return pld_fabrication
+
+
+def build_nexus_file(pld_fabrication, output_path, rheed_osc=None, heatmap_matrix=None):
+    # NOTE: look at the mail attachment from Emiliano...
+    with h5py.File(output_path, "w") as f:
+        nx_pld_entry = f.create_group("pld_fabrication")
+        nx_pld_entry.attrs["NX_class"] = "NXentry"
+
+        # Sample section
+        nx_sample = nx_pld_entry.create_group("sample")
+        nx_sample.attrs["NX_class"] = "NXsample"
+        sample_dict = pld_fabrication["sample"]
+
+        # Substrate sub-section
+        nx_substrate = nx_sample.create_group("substrate")
+        nx_substrate.attrs["NX_class"] = "NXsubentry"
+        substrate_dict = sample_dict["substrate"]
+        try:
+            # Substrate fields (datasets)
+            nx_substrate.create_dataset("name", data=substrate_dict["name"])
+            nx_substrate.create_dataset(
+                "chemical_formula", data=substrate_dict["chemical_formula"]
+            )
+            nx_substrate.create_dataset(
+                "orientation", data=substrate_dict["orientation"]
+            )
+            nx_substrate.create_dataset(
+                "miscut_angle", data=substrate_dict["miscut_angle"]["value"]
+            )  # float
+            nx_substrate["miscut_angle"].attrs["units"] = substrate_dict[
+                "miscut_angle"
+            ]["units"]
+            nx_substrate.create_dataset(
+                "miscut_direction", data=substrate_dict["miscut_direction"]
+            )
+            nx_substrate.create_dataset(
+                "thickness", data=substrate_dict["thickness"]["value"]
+            )  # float/int
+            nx_substrate["thickness"].attrs["units"] = substrate_dict["thickness"][
+                "units"
+            ]
+            nx_substrate.create_dataset("dimensions", data=substrate_dict["dimensions"])
+            nx_substrate.create_dataset(
+                "surface_treatment", data=substrate_dict["surface_treatment"]
+            )
+            nx_substrate.create_dataset(
+                "manufacturer", data=substrate_dict["manufacturer"]
+            )
+            nx_substrate.create_dataset("batch_id", data=substrate_dict["batch_id"])
+        except TypeError as te:
+            # sooner or later I'll handle this too - not today tho
+            raise TypeError(te)
+
+        # Multilayer sub-section
+        nx_multilayer = nx_sample.create_group("multilayer")
+        nx_multilayer.attrs["NX_class"] = "NXsubentry"
+        multilayer_dict = sample_dict["multilayer"]
+        # Repeat FOR EACH LAYER:
+        for layer in multilayer_dict:
+            nx_layer = nx_multilayer.create_group(layer)
+            nx_layer.attrs["NX_class"] = "NXsubentry"
+            layer_dict = multilayer_dict[layer]
+            # Sub-groups of a layer
+            ## Target
+            nx_target = nx_layer.create_group("target")
+            nx_target.attrs["NX_class"] = "NXsample"
+            target_dict = layer_dict["target"]
+            ## Rastering and Annealing
+            nx_laser_rastering = nx_layer.create_group("laser_rastering")
+            nx_laser_rastering.attrs["NX_class"] = "NXprocess"
+            rastering_dict = layer_dict["laser_rastering"]
+            nx_pre_annealing = nx_layer.create_group("pre_annealing")
+            nx_pre_annealing.attrs["NX_class"] = "NXprocess"
+            pre_ann_dict = layer_dict["pre_annealing"]
+            nx_post_annealing = nx_layer.create_group("post_annealing")
+            nx_post_annealing.attrs["NX_class"] = "NXprocess"
+            post_ann_dict = layer_dict["post_annealing"]
+            nx_layer_instruments = nx_layer.create_group("instruments_used")
+            nx_layer_instruments.attrs["NX_class"] = "NXinstrument"
+            layer_instruments_dict = layer_dict["instruments_used"]
+
+            ## Target metadata
+            try:
+                nx_target.create_dataset("name", data=target_dict["name"])
+                nx_target.create_dataset(
+                    "chemical_formula", data=target_dict["chemical_formula"]
+                )
+                nx_target.create_dataset("description", data=target_dict["description"])
+                nx_target.create_dataset("shape", data=target_dict["shape"])
+                nx_target.create_dataset("dimensions", data=target_dict["dimensions"])
+                nx_target.create_dataset(
+                    "thickness", data=target_dict["thickness"]["value"]
+                )  # float/int
+                nx_target["thickness"].attrs["units"] = target_dict["thickness"][
+                    "units"
+                ]
+                nx_target.create_dataset("solid_form", data=target_dict["solid_form"])
+                nx_target.create_dataset(
+                    "manufacturer", data=target_dict["manufacturer"]
+                )
+                nx_target.create_dataset("batch_id", data=target_dict["batch_id"])
+            except TypeError as te:
+                raise TypeError(te)
+            ## Other layer-specific metadata
+            try:
+                nx_layer.create_dataset("start_time", data=layer_dict["start_time"])
+                nx_layer.create_dataset("operator", data=layer_dict["operator"])
+                nx_layer.create_dataset(
+                    "number_of_pulses", data=layer_dict["number_of_pulses"]
+                )
+                nx_layer.create_dataset(
+                    "deposition_time", data=layer_dict["deposition_time"]["value"]
+                )
+                nx_layer["deposition_time"].attrs["units"] = layer_dict[
+                    "deposition_time"
+                ]["units"]
+                nx_layer.create_dataset(
+                    "repetition_rate", data=layer_dict["repetition_rate"]["value"]
+                )
+                nx_layer["repetition_rate"].attrs["units"] = layer_dict[
+                    "repetition_rate"
+                ]["units"]
+                nx_layer.create_dataset(
+                    "temperature", data=layer_dict["temperature"]["value"]
+                )
+                nx_layer["temperature"].attrs["units"] = layer_dict["temperature"][
+                    "units"
+                ]
+                nx_layer.create_dataset(
+                    "heating_method", data=layer_dict["heating_method"]
+                )
+                nx_layer.create_dataset(
+                    "layer_thickness", data=layer_dict["layer_thickness"]["value"]
+                )
+                nx_layer["layer_thickness"].attrs["units"] = layer_dict[
+                    "layer_thickness"
+                ]["units"]
+                nx_layer.create_dataset("buffer_gas", data=layer_dict["buffer_gas"])
+                nx_layer.create_dataset(
+                    "process_pressure", data=layer_dict["process_pressure"]["value"]
+                )
+                nx_layer["process_pressure"].attrs["units"] = layer_dict[
+                    "process_pressure"
+                ]["units"]
+                nx_layer.create_dataset(
+                    "heater_target_distance",
+                    data=layer_dict["heater_target_distance"]["value"],
+                )
+                nx_layer["heater_target_distance"].attrs["units"] = layer_dict[
+                    "heater_target_distance"
+                ]["units"]
+                nx_layer.create_dataset(
+                    "laser_fluence", data=layer_dict["laser_fluence"]["value"]
+                )
+                nx_layer["laser_fluence"].attrs["units"] = layer_dict["laser_fluence"][
+                    "units"
+                ]
+                nx_layer.create_dataset(
+                    "laser_spot_area", data=layer_dict["laser_spot_area"]["value"]
+                )
+                nx_layer["laser_spot_area"].attrs["units"] = layer_dict[
+                    "laser_spot_area"
+                ]["units"]
+                nx_layer.create_dataset(
+                    "laser_energy", data=layer_dict["laser_energy"]["value"]
+                )
+                nx_layer["laser_energy"].attrs["units"] = layer_dict["laser_energy"][
+                    "units"
+                ]
+            except TypeError as te:
+                raise TypeError(te)
+            ## Rastering metadata
+            try:
+                nx_laser_rastering.create_dataset(
+                    "geometry", data=rastering_dict["geometry"]
+                )
+                nx_laser_rastering.create_dataset(
+                    "positions", data=rastering_dict["positions"]
+                )
+                nx_laser_rastering.create_dataset(
+                    "velocities", data=rastering_dict["velocities"]
+                )
+            except TypeError as te:
+                raise TypeError(te)
+            ## Annealing metadata
+            try:
+                nx_pre_annealing.create_dataset(
+                    "ambient_gas", data=pre_ann_dict["ambient_gas"]
+                )
+                nx_pre_annealing.create_dataset(
+                    "pressure", data=pre_ann_dict["pressure"]["value"]
+                )
+                nx_pre_annealing["pressure"].attrs["units"] = pre_ann_dict["pressure"][
+                    "units"
+                ]
+                nx_pre_annealing.create_dataset(
+                    "temperature", data=pre_ann_dict["temperature"]["value"]
+                )
+                nx_pre_annealing["temperature"].attrs["units"] = pre_ann_dict[
+                    "temperature"
+                ]["units"]
+                nx_pre_annealing.create_dataset(
+                    "duration", data=pre_ann_dict["duration"]["value"]
+                )
+                nx_pre_annealing["duration"].attrs["units"] = pre_ann_dict["duration"][
+                    "units"
+                ]
+            except TypeError as te:
+                raise TypeError(te)
+            try:
+                nx_post_annealing.create_dataset(
+                    "ambient_gas", data=post_ann_dict["ambient_gas"]
+                )
+                nx_post_annealing.create_dataset(
+                    "pressure", data=post_ann_dict["pressure"]["value"]
+                )
+                nx_post_annealing["pressure"].attrs["units"] = post_ann_dict[
+                    "pressure"
+                ]["units"]
+                nx_post_annealing.create_dataset(
+                    "temperature", data=post_ann_dict["temperature"]["value"]
+                )
+                nx_post_annealing["temperature"].attrs["units"] = post_ann_dict[
+                    "temperature"
+                ]["units"]
+                nx_post_annealing.create_dataset(
+                    "duration", data=post_ann_dict["duration"]["value"]
+                )
+                nx_post_annealing["duration"].attrs["units"] = post_ann_dict[
+                    "duration"
+                ]["units"]
+            except TypeError as te:
+                raise TypeError(te)
+            try:
+                nx_layer_instruments.create_dataset(
+                    "laser_system", data=layer_instruments_dict["laser_system"]
+                )
+                nx_layer_instruments.create_dataset(
+                    "deposition_chamber",
+                    data=layer_instruments_dict["deposition_chamber"],
+                )
+                nx_layer_instruments.create_dataset(
+                    "rheed_system", data=layer_instruments_dict["rheed_system"]
+                )
+            except TypeError as te:
+                raise TypeError(te)
+
+        # Instruments used section
+        nx_instruments = nx_pld_entry.create_group("instruments_used")
+        nx_instruments.attrs["NX_class"] = "NXinstrument"
+        instruments_dict = pld_fabrication["instruments_used"]
+        try:
+            nx_instruments.create_dataset(
+                "laser_system", data=instruments_dict["laser_system"]
+            )
+            nx_instruments.create_dataset(
+                "deposition_chamber", data=instruments_dict["deposition_chamber"]
+            )
+            nx_instruments.create_dataset(
+                "rheed_system", data=instruments_dict["rheed_system"]
+            )
+        except TypeError as te:
+            raise TypeError(te)
+
+        # RHEED data section
+        nx_rheed = nx_pld_entry.create_group("rheed_data")
+        nx_rheed.attrs["NX_class"] = "NXdata"
+
+        if rheed_osc is not None:
+            # Asse temporale
+            t_ds = nx_rheed.create_dataset("time", data=rheed_osc["time"])
+            t_ds.attrs["units"] = "s"
+            t_ds.attrs["long_name"] = "Time"
+
+            # Intensità: shape (n_layers, n_timepoints, 3)
+            i_ds = nx_rheed.create_dataset("intensity", data=rheed_osc["intensity"])
+            i_ds.attrs["units"] = "a.u."
+            i_ds.attrs["long_name"] = "RHEED Intensity"
+
+            # Attributi NXdata — notazione NeXus 3.x corretta
+            nx_rheed.attrs["signal"] = "intensity"
+            nx_rheed.attrs["axes"] = [
+                ".",
+                "time",
+                ".",
+            ]  # solo l'asse 1 (time) è denominato
+            nx_rheed.attrs["time_indices"] = np.array([1], dtype=np.int32)
+            # ###########
+            # nx_rheed = nx_pld_entry.create_group("rheed_data")
+            # nx_rheed.attrs["NX_class"] = "NXdata"
+
+            # nx_rheed.create_dataset("time", data=rheed_osc["time"])
+            # nx_rheed["time"].attrs["units"] = "s"
+
+            # nx_rheed.create_dataset("intensity", data=rheed_osc["intensity"])
+            # #nx_rheed["intensity"].attrs["units"] = "counts"
+            # nx_rheed["intensity"].attrs["long_name"] = "RHEED intensity"
+            # nx_rheed.attrs["signal"] = "intensity"
+            # nx_rheed.attrs["axes"] = "layer:time:channel"
+            # nx_rheed.attrs["layer_indices"] = [0]  # asse layer
+            # nx_rheed.attrs["time_indices"] = [1]   # asse tempo
+            # nx_rheed.attrs["channel_indices"] = [2]
+        if heatmap_matrix is not None:
+            heatmap = nx_rheed.create_dataset("diffraction_image", data=heatmap_matrix)
+            heatmap.attrs["long_name"] = "Diffraction Image"
+            heatmap.attrs["units"] = "a.u."
+            # this is of my own initiative. good???
+            heatmap.attrs["interpretation"] = "spectrum"
+            # suggested by DeepSeek, useful? probably not.
+            # heatmap.attrs["suggested_colormap"] = "inferno"
+            # heatmap.attrs["scale_min"] = 0.0
+            # heatmap.attrs["scale_max"] = 1.0
+    return
+
+
+if __name__ == "__main__":
+    # TO-DO: place the API base URL somewhere else.
     ELABFTW_API_URL = "https://elabftw.fisica.unina.it/api/v2"
     apikey = getpass("Paste API key here: ")
-    elabid = input("Enter elabid of your starting sample [default= 1111]: ") or 1111
+    elabid = input("Enter elabid of your starting sample [default = 1111]: ") or 1111
     data = APIHandler(apikey).get_entry_from_elabid(elabid)
     sample = Entrypoint(data)
-    batch = chain_entrypoint_to_batch(sample) # Material-class object
-    bd = batch.__dict__
-    bd.pop("extra")
+    sample_name = sample.name.strip().replace(" ", "_")
+    substrate_object = chain_entrypoint_to_batch(sample)  # Substrate-class object
     layers = chain_entrypoint_to_layers(sample)  # list of Layer-class objects
-    print(f"Sample name:\n{sample.name}\n")
-    print(f"Substrate data:\n{bd}\n")
-    print(f"Layers data:")
-    for layer in layers:
-        ld = layer.__dict__
-        ld.pop("extra")
-        tgt = chain_layer_to_target(layer)
-        td = tgt.__dict__
-        td.pop("extra")
-        print(ld)
-        print(td)
-        print()
+    n_layers = len(layers)  # total number of layers on the sample
+    result = make_nexus_schema_dictionary(substrate_object, layers)
+    # print(make_nexus_schema_dictionary(substrate_object, layers)) # debug
+    with open(f"output/sample-{sample_name}.json", "w") as f:
+        json.dump(result, f, indent=3)
+    # TO-DO: remove the hard-coded path of the RWA file
+    # ideally the script should download a TXT/CSV file from each layer
+    # (IF PRESENT ←→ also handle missing file error)
+    # and merge all data in a single file to analyse it
+    # WARNING: fails if file is missing
 
-    # entryType = None
-    # while entryType not in ["items", "experiments"]:
-    #     eT = input("Enter a valid entry type [items, experiments]: ")
-    #     # This allows for a shortcut: instead of prompting the type before and the elabid after I can just prompt both at the same time - e.g. e51 is exp. 51, i1108 is item 1108...
-    #     if eT[0] in ["e", "i"] and eT[-1].isnumeric():
-    #         try:
-    #             elabid = int(eT[1:])
-    #             eT = eT[0]
-    #         except Exception:
-    #             print("Usage: i|item|items|i[ELABID] for items, e|experiment|experiments|e[ELABID] for experiments.")
-    #             continue
-    #     match eT:
-    #         case "items" | "i" | "item":
-    #             entryType = "items"
-    #         case "experiments" | "e" | "exp" | "experiment":
-    #             entryType = "experiments"
-    #         case _:
-    #             continue
-    # # This will probably be reworked in production
-    # try:
-    #     elabid = elabid
-    # except NameError:
-    #     elabid = input("Input elabid here [default = 1111]: ") or 1111
-    # data = APIHandler(apikey).get_entry_from_elabid(elabid, entryType)
-    # if entryType == "experiments":
-    #     layer = Layer(data)
-    #     result = layer.__dict__
-    #     result.pop("extra")
-    #     print(result)
-    # elif entryType == "items":
-    #     if data.get("category_title") == "Sample":
-    #         item = Entrypoint(data)
-    #     elif data.get("category_title") in ["PLD Target", "Substrate"]:
-    #         item = Material(data)
-    #         print(item.get_compound_formula(apikey))
-    #     else:
-    #         raise Exception("The selected item or experiment is not in one of the following categories: [Sample, PLD Target, Substrate, PLD Deposition].")
-    #     result = item.__dict__
-    #     result.pop("extra")
-    #     print(result)
+    with open("tests/Realtime_Window_Analysis.txt", "r") as o:
+        osc = np.loadtxt(o, delimiter="\t")
+    try:
+        rheed_osc = (
+            analyse_rheed_data(data=osc) or None
+        )  # analyze rheed data first, build the file later
+    except ValueError as ve:
+        raise ValueError(
+            f"Error with function analyse_rheed_data. {ve}\nPlease make sure the Realtime Window Analysis file is exactly 4 columns wide - where the first column represents time and the others are RHEED intensities."
+        )
+    # This one tries to open a png image.
+    # Emiliano said to keep it to one image per layer tops.
+    # In this test I will only consider one image.
+    # TO-DO: make it format-agnostic. If not possible, make it PNG-only.
+    if os.path.isfile("tests/LAO_16min50s_736C_STO.bmp"):  # if BMP
+        # if os.path.isfile("tests/LAO_16min50s_736C_STO.png"): # if PNG
+        img = Image.open("tests/LAO_16min50s_736C_STO.bmp").convert("L")
+        mx = np.array(img, dtype=np.uint8)
+        # mx = mx.astype(np.float32) / 255.0  # consider deleting???
+    build_nexus_file(
+        result,
+        output_path=f"output/sample-{sample_name}-nexus.h5",
+        rheed_osc=rheed_osc,
+        heatmap_matrix=mx,
+    )

src/schema/pld_deposition.py

@@ -0,0 +1,3 @@
+class Prova:
+    def __init__(self):
+        self.hello = "Hello world"