Layer class methods to list attachments up and tested

to-do: clean code

src/APIHandler.py

@@ -1,4 +1,5 @@
 import os, requests
+from getpass import getpass
 import elabapi_python as elabapi
 
 
@@ -80,16 +81,17 @@ class APIHandler:
         entry_data = response.json()
         return entry_data
 
-    def download_attachments_data(self, elabid, entryType="experiments"):
+    def download_attachment_data(self, elabid, upload_id, 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.
+        Downloads a specific attachment of a certain eLabFTW experiment (default) or item.
+        Only returns its binary data. Use method download_attachment_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.
+            * The key is the attachment's filename;
+            * The value is the attachment's binary data.
 
         Args:
             elabid:     eLabFTW internal ID of the selected resource.
+            upload_id:  eLabFTW internal ID of the selected upload.
             entryType:  Resource type. Anything other than "experiments" or "items" WILL raise an error.
         """
         if entryType not in ["experiments", "items"]:
@@ -108,29 +110,33 @@ class APIHandler:
         )
         uploads_api = elabapi.UploadsApi(api_client)
 
-        # Actual uploads (dictionary):
-        uploads = {
+        # Scans through the attachments and selects the one with corresponing ID.
+        attachment = {
             upload.real_name: uploads_api.read_upload(
-                entryType, elabid, upload.id, format="binary", _preload_content=False
+                entryType, elabid, upload_id, format="binary", _preload_content=False
             ).data
             for upload in uploads_api.read_uploads(entryType, elabid)
+            if upload.id == upload_id
         }
 
-        return uploads
+        return attachment
 
-    def download_attachments_to_disk(
+    def download_attachment_to_disk(
         self,
         elabid,
+        upload_id,
         entryType="experiments",
         dump_dir="output/attachments",
         # persistent=True,
     ):
         """
-        Downloads attachments of a certain eLabFTW experiment (default) or item.
+        Downloads a specific attachment of a certain eLabFTW experiment (default) or item.
         Downloads their binary data through method download_attachments_data and dumps it to dump_dir.
+        Returns nothing on success.
 
         Args:
             elabid:     eLabFTW internal ID of the selected resource.
+            upload_id:  eLabFTW internal ID of the selected upload.
             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.
@@ -142,9 +148,16 @@ class APIHandler:
                 "You can only download attachments from experiments or items."
             )
 
-        uploads = download_attachments_data(elabid, entryType=entryType)
+        uploads = self.download_attachment_data(elabid, upload_id, entryType=entryType)
         for file in uploads:
-            raw_data = uploads["file"]
+            raw_data = uploads[file]
             with open(os.path.join(dump_dir, f"exp{elabid}-{file}"), "wb") as f:
                 f.write(raw_data)
         return
+
+
+# Testing methods
+if __name__ == "__main__":
+    api_key = getpass("Paste API key here [no echo]: ")
+    handler = APIHandler(api_key=api_key)
+    handler.download_attachment_to_disk(elabid=58, upload_id=81)

src/classes.py

@@ -1,4 +1,5 @@
 import os, json, requests
+from getpass import getpass
 from APIHandler import APIHandler
 
 
@@ -149,16 +150,21 @@ class Layer:
     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).
+            * Each key is the attachment's progressive ID (0, 1...);
+            * Each value is a dictionary containing the attachment's elabid, 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.
+
+        Exception: returns {} (empty dictionary) if no uploads/attachments on Layer.
         """
         # Remember: Layers are experiments, so we only need to look for attachments in the experiment endpoint.
+        if self.uploads == []:
+            return {}
         attachments = {
-            attachment["id"]: {
+            self.uploads.index(attachment): {
+                "id": attachment["id"],
                 "filename": attachment["real_name"],
                 "hashname": attachment["long_name"],
                 "related_experiment": attachment["item_id"],
@@ -180,7 +186,7 @@ class Layer:
         textual_uploads = {
             attachment: attachments[attachment]
             for attachment in attachments
-            if attachments[attachments]["filename"][-4:] in (".txt", ".csv", ".tsv")
+            if attachments[attachment]["filename"][-4:] in (".txt", ".csv", ".tsv")
         }
         return textual_uploads
 
@@ -195,12 +201,12 @@ class Layer:
         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 = {
+        images = {
             attachment: attachments[attachment]
             for attachment in attachments
-            if attachments[attachments]["filename"][-4:] in (".png", ".bmp")
+            if attachments[attachment]["filename"][-4:] in (".png", ".bmp")
         }
-        return pictures
+        return images
 
 
 class Entrypoint:
@@ -325,6 +331,13 @@ class Target(Material):
 
 
 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.")
+    # 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.")
+    api_key = getpass("Paste API key here [no echo]: ")
+    handler = APIHandler(api_key=api_key)
+    exp58 = handler.get_entry_from_elabid(elabid=58, entryType="experiments")
+    layer58 = Layer(exp58)
+    print(layer58.list_attachments())
+    print(layer58.fetch_textual_uploads())
+    print(layer58.fetch_images())

src/main.py

@@ -204,6 +204,84 @@ def deduplicate_instruments_from_layers(layers):
     # }
 
 
+def select_rheed_data(layer):
+    """
+    Takes a Layer-class object and selects the attachments to use to create the RHEED dataset for the NeXus file.
+
+    There are two categories of attachments considered: text-files and pictures.
+    The only accepted formats are ".txt", ".csv" and ".tsv" for the first ones, and ".png" or ".bmp" for the others.
+    The function is extension-sensitive, and only one attachment for each category will be downloaded.
+
+    If there are more than one attachment for each category, the user is prompted to select one of them from a list.
+    If there are no attachments for a category the function will return {} (empty dictionary) for that category.
+
+    Returns the set: (rheed_data_file, rheed_image_file). Both variables are dictionaries in the following format:
+        {
+            "fullname": real_name (with extension),
+            "hashname": long_name (with extension),
+            "related_experiment": elabid
+        }
+    """
+    n = layer.layer_number
+    textual_uploads = layer.fetch_textual_uploads()
+    images = layer.fetch_images()
+
+    if len(textual_uploads) == 0:
+        rheed_data_file = {}
+    elif len(textual_uploads) > 1:
+        # prompt user to select from list
+        print(f"Attention: Layer {n} contains multiple TEXTUAL attachments.\n")
+        print("These are used to populate the 'RHEED intensities' dataset.")
+        print("=== USER INTERVENTION REQUIRED ===")
+        for id in textual_uploads:
+            print(f"{id} - {textual_uploads[id]}")
+        ans = None
+        while not type(ans) == int or not ans in range(0, len(textual_uploads)):
+            ans = (
+                input(
+                    "Select one of the attachments from the list (0, 1, ...) [default = 0]: "
+                )
+                or 0
+            )
+            if ans.isdigit():
+                ans = int(ans)
+            continue
+        rheed_data_file = textual_uploads[ans]  # still a dictionary
+    else:
+        rheed_data_file = textual_uploads[0]
+
+    if len(images) == 0:
+        rheed_image_file = {}
+    elif len(images) > 1:
+        # prompt user to select from list
+        print(f"Attention: Layer {n} contains multiple PNG/BMP attachments.\n")
+        print("These are used to create the RHEED heatmap.")
+        print("=== USER INTERVENTION REQUIRED ===")
+        for id in images:
+            print(f"{id} - {images[id]}")
+        ans = None
+        while not type(ans) == int or not ans in range(0, len(images)):
+            ans = (
+                input(
+                    "Select one of the attachments from the list (0, 1, ...) [default = 0]: "
+                )
+                or 0
+            )
+            if ans.isdigit():
+                ans = int(ans)
+            continue
+        rheed_image_file = images[ans]  # still a dictionary
+    else:
+        rheed_image_file = images[0]
+
+    return (rheed_data_file, rheed_image_file)
+
+
+def download_rheed_data():
+
+    return
+
+
 def analyse_rheed_data(data):
     """
     Takes the content of a tsv file and returns a dictionary with timestamps and intensities.
@@ -236,7 +314,7 @@ def analyse_rheed_data(data):
         raise ValueError(
             f"Insufficient number of columns: expected 4, got n_cols = {n_cols}."
         )
-    n_time_points = data.shape[0]
+    # n_time_points = data.shape[0]
 
     # Get time (all rows of col 0) as Float64:
     time = data[:, 0].astype(
@@ -254,7 +332,11 @@ def analyse_rheed_data(data):
 
 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.
+    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 = {
@@ -280,8 +362,10 @@ def make_nexus_schema_dictionary(substrate_object, layers):
             "multilayer": {},
         },
         "instruments_used": instruments["multilayer"],
+        "rheed_data": {},
     }
     multilayer = pld_fabrication["sample"]["multilayer"]
+    rheed_data = pld_fabrication["rheed_data"]
     for layer in layers:
         name = "layer_" + layer.layer_number
         target_object = chain_layer_to_target(layer)
@@ -381,6 +465,7 @@ def make_nexus_schema_dictionary(substrate_object, layers):
             },
             "instruments_used": instruments[name],
         }
+        rheed_data[name] = {}
     return pld_fabrication
 
 
@@ -650,6 +735,24 @@ def build_nexus_file(pld_fabrication, output_path, rheed_osc=None, heatmap_matri
         nx_rheed = nx_pld_entry.create_group("rheed_data")
         nx_rheed.attrs["NX_class"] = "NXdata"
 
+        # here's what we gon do: (to be read with the voice of Mike from Breaking Bad)
+        # 1. rheed_osc and heatmap_matrix are NOT given in input to the function so no need for checking that
+        # 2. loop through the layers, each with its elabid and metadata
+        # 2a. read said metadata for each layer, print list of txt and png files (dedicated Layer class methods)
+        # 2b. prompt the user for file choice (1 text file per layer - in tsv format, 1 picture file - either png [default] or bmp)
+        # 2c. download the chosen file
+        # 2d. with chosen file do analysis as before
+        # 3. the schema should be:
+        #   * /rheed_data
+        #       * /layer_n
+        #           * time (rheed_osc)
+        #           * intensity (rheed_osc)
+        #           * diffraction_image (heatmap_matrix)
+        # first problem is probably finding out how to recover the following meta from the original Layer object:
+        #   * Layer.elabid - integer
+        #   * Layer.fetch_textual_uploads() - dictionary
+        #   * Layer.fetch_images() - dictionary
+
         if rheed_osc is not None:
             # Asse temporale
             t_ds = nx_rheed.create_dataset("time", data=rheed_osc["time"])