dayhoff-tools 1.1.41__py3-none-any.whl → 1.1.43__py3-none-any.whl

dayhoff_tools/deployment/deploy_aws.py

@@ -374,6 +374,11 @@ def create_or_update_job_definition(
         "timeout": {"attemptDurationSeconds": aws_config.get("timeout_seconds", 86400)},
     }
 
+    # Add tags if specified in config
+    if "tags" in aws_config:
+        job_definition_args["tags"] = aws_config["tags"]
+        print(f"Adding tags to job definition: {aws_config['tags']}")
+
     # Register new revision using the session client
     response = batch.register_job_definition(**job_definition_args)
 
@@ -472,6 +477,11 @@ def submit_aws_batch_job(
     print(f"Setting retry attempts to {retry_attempts}")
     job_submit_args["retryStrategy"] = {"attempts": retry_attempts}
 
+    # Add tags if specified in config
+    if "tags" in aws_config:
+        job_submit_args["tags"] = aws_config["tags"]
+        print(f"Adding tags to batch job: {aws_config['tags']}")
+
     # Submit the job using the session client
     response = batch.submit_job(**job_submit_args)
 

dayhoff_tools/deployment/processors.py

@@ -1,9 +1,5 @@
-import csv
 import json
 import logging
-import os
-import shlex
-import shutil
 import subprocess
 from abc import ABC, abstractmethod
 from pathlib import Path
@@ -295,108 +291,3 @@ class InterProScanProcessor(Processor):
                 cleaned_input_file_path.unlink()
 
         return str(chunk_output_dir)
-
-
-class BoltzPredictor(Processor):
-    """Processor for running Boltz docking predictions.
-
-    This class wraps the Boltz docking tool to predict protein structures
-    from sequence data.
-    """
-
-    def __init__(self, num_workers: int, boltz_options: str | None = None):
-        """Initialize the BoltzPredictor.
-
-        Args:
-            num_workers: Number of worker threads to use as a default.
-                         This can be overridden if --num_workers is present
-                         in boltz_options.
-            boltz_options: A string containing additional command-line options
-                           to pass to the Boltz predictor. Options should be
-                           space-separated (e.g., "--option1 value1 --option2").
-        """
-        self.num_workers = num_workers
-        self.boltz_options = boltz_options
-
-    def run(self, input_file: str) -> str:
-        """Run Boltz prediction on the input file.
-
-        Constructs the command using the input file, default number of workers,
-        and any additional options provided via `boltz_options`. If `--num_workers`
-        is specified in `boltz_options`, it overrides the default `num_workers`.
-
-        Args:
-            input_file: Path to the input file containing sequences
-
-        Returns:
-            Path to the output directory created by Boltz
-
-        Raises:
-            subprocess.CalledProcessError: If Boltz prediction fails
-        """
-        # Determine expected output directory name
-        input_base = os.path.splitext(os.path.basename(input_file))[0]
-        expected_output_dir = f"boltz_results_{input_base}"
-        logger.info(f"Expected output directory: {expected_output_dir}")
-
-        # Start building the command
-        cmd = ["boltz", "predict", input_file]
-
-        # Parse additional options if provided
-        additional_args = []
-        num_workers_in_opts = False
-        if self.boltz_options:
-            try:
-                parsed_opts = shlex.split(self.boltz_options)
-                additional_args.extend(parsed_opts)
-                if "--num_workers" in parsed_opts:
-                    num_workers_in_opts = True
-                    logger.info(
-                        f"Using --num_workers from BOLTZ_OPTIONS: {self.boltz_options}"
-                    )
-            except ValueError as e:
-                logger.error(f"Error parsing BOLTZ_OPTIONS '{self.boltz_options}': {e}")
-                # Decide if we should raise an error or proceed without options
-                # For now, proceed without the additional options
-                additional_args = []  # Clear potentially partially parsed args
-
-        # Add num_workers if not specified in options
-        if not num_workers_in_opts:
-            logger.info(f"Using default num_workers: {self.num_workers}")
-            cmd.extend(["--num_workers", str(self.num_workers)])
-
-        # Add the parsed additional arguments
-        cmd.extend(additional_args)
-
-        # Log the final command
-        # Use shlex.join for safer command logging, especially if paths/args have spaces
-        try:
-            safe_cmd_str = shlex.join(cmd)
-            logger.info(f"Running command: {safe_cmd_str}")
-        except AttributeError:  # shlex.join is Python 3.8+
-            logger.info(f"Running command: {' '.join(cmd)}")
-
-        # Stream output in real-time
-        process = subprocess.Popen(
-            cmd,
-            stdout=subprocess.PIPE,
-            stderr=subprocess.STDOUT,
-            text=True,
-            bufsize=1,
-        )
-
-        stdout = process.stdout
-        if stdout:
-            for line in iter(stdout.readline, ""):
-                logger.info(f"BOLTZ: {line.rstrip()}")
-
-        # Wait for process to complete
-        return_code = process.wait()
-        if return_code != 0:
-            logger.error(f"Boltz prediction failed with exit code {return_code}")
-            raise subprocess.CalledProcessError(return_code, cmd)
-
-        logger.info(
-            f"Boltz prediction completed successfully. Output in {expected_output_dir}"
-        )
-        return expected_output_dir

dayhoff_tools/deployment/swarm.py

@@ -128,23 +128,58 @@ def publish_cards(
     names: List[str],
     firestore_collection: str,
 ):
-    """Publish cards to Firebase. Expects a list of filenames (not full paths),
-    which will each be published as a new document in the collection."""
+    """Publish cards to Firebase using batch writes for optimal performance.
+
+    Expects a list of filenames (not full paths), which will each be published
+    as a new document in the collection. Uses Firestore batch writes to minimize
+    network round-trips and improve performance.
+
+    Args:
+        names: List of packet filenames to publish as cards
+        firestore_collection: Name of the Firestore collection to write to
+    """
+    if not names:
+        print("No cards to upload.")
+        return
 
     initialize_firebase()
-    collection = firestore.client().collection(firestore_collection)
+    db = firestore.client()
+    collection = db.collection(firestore_collection)
+
+    # Firestore batch limit is 500 operations
+    BATCH_SIZE = 500
+    total_cards = len(names)
+    cards_processed = 0
+
+    # Process names in batches of up to 500
+    for i in range(0, total_cards, BATCH_SIZE):
+        batch = db.batch()
+        batch_names = names[i : i + BATCH_SIZE]
+
+        # Add all operations for this batch
+        for name in batch_names:
+            doc_ref = collection.document()  # Auto-generate document ID
+            batch.set(
+                doc_ref,
+                {
+                    "status": "available",
+                    "packet_filename": name,
+                    "created": datetime.now(ZoneInfo("America/Los_Angeles")),
+                },
+            )
 
-    for name in names:
-        collection.document().set(
-            {
-                "status": "available",
-                "packet_filename": name,
-                "created": datetime.now(ZoneInfo("America/Los_Angeles")),
-            }
+        # Commit the entire batch atomically
+        batch.commit()
+        cards_processed += len(batch_names)
+
+        print(
+            f"Batch {i // BATCH_SIZE + 1}: Created {len(batch_names)} cards "
+            f"({cards_processed}/{total_cards} total)"
         )
-        print(f"Creating card {name}")
 
-    print(f"Uploaded {len(names)} cards.")
+    print(
+        f"Successfully uploaded {total_cards} cards in {(total_cards + BATCH_SIZE - 1) // BATCH_SIZE} batch(es)."
+    )
 
 
 @transactional

{dayhoff_tools-1.1.41.dist-info → dayhoff_tools-1.1.43.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: dayhoff-tools
-Version: 1.1.41
+Version: 1.1.43
 Summary: Common tools for all the repos at Dayhoff Labs
 Author: Daniel Martin-Alarcon
 Author-email: dma@dayhofflabs.com

{dayhoff_tools-1.1.41.dist-info → dayhoff_tools-1.1.43.dist-info}/RECORD

@@ -7,12 +7,12 @@ dayhoff_tools/cli/main.py,sha256=47EGb28ALaYFc7oAUGlY1D66AIDmc4RZiXxN-gPVrpQ,451
 dayhoff_tools/cli/swarm_commands.py,sha256=5EyKj8yietvT5lfoz8Zx0iQvVaNgc3SJX1z2zQR6o6M,5614
 dayhoff_tools/cli/utility_commands.py,sha256=ER4VrJt4hu904MwrcltUXjwBWT4uFrP-aPXjdXyT3F8,24685
 dayhoff_tools/deployment/base.py,sha256=8tXwsPYvRo-zV-aNhHw1c7Rji-KWg8S5xoCCznFnVVI,17412
-dayhoff_tools/deployment/deploy_aws.py,sha256=jQyQ0fbm2793jEHFO84lr5tNqiOpdBg6U0S5zCVJr1M,17884
+dayhoff_tools/deployment/deploy_aws.py,sha256=GvZpE2YIFA5Dl9rkAljFjtUypmPDNbWgw8NicHYTP24,18265
 dayhoff_tools/deployment/deploy_gcp.py,sha256=xgaOVsUDmP6wSEMYNkm1yRNcVskfdz80qJtCulkBIAM,8860
 dayhoff_tools/deployment/deploy_utils.py,sha256=StFwbqnr2_FWiKVg3xnJF4kagTHzndqqDkpaIOaAn_4,26027
 dayhoff_tools/deployment/job_runner.py,sha256=hljvFpH2Bw96uYyUup5Ths72PZRL_X27KxlYzBMgguo,5086
-dayhoff_tools/deployment/processors.py,sha256=f4L52ekx_zYirl8C4WfavxtOioyD-c34TdTJVDoLpWs,16572
-dayhoff_tools/deployment/swarm.py,sha256=MGcS2_x4RNFtnVjWlU_SwNfhICz8NlGYr9cYBK4ZKDA,21688
+dayhoff_tools/deployment/processors.py,sha256=LM0CQbr4XCb3AtLbrcuDQm4tYPXsoNqgVJ4WQYDjzJc,12406
+dayhoff_tools/deployment/swarm.py,sha256=YJfvVOcAS8cYcIj2fiN4qwC2leh0I9w5A4px8ZWSF6g,22833
 dayhoff_tools/embedders.py,sha256=fRkyWjHo8OmbNUBY_FwrgfvyiLqpmrpI57UAb1Szn1Y,36609
 dayhoff_tools/fasta.py,sha256=_kA2Cpiy7JAGbBqLrjElkzbcUD_p-nO2d5Aj1LVmOvc,50509
 dayhoff_tools/file_ops.py,sha256=JlGowvr-CUJFidV-4g_JmhUTN9bsYuaxtqKmnKomm-Q,8506
@@ -26,7 +26,7 @@ dayhoff_tools/intake/uniprot.py,sha256=BZYJQF63OtPcBBnQ7_P9gulxzJtqyorgyuDiPeOJq
 dayhoff_tools/logs.py,sha256=DKdeP0k0kliRcilwvX0mUB2eipO5BdWUeHwh-VnsICs,838
 dayhoff_tools/sqlite.py,sha256=jV55ikF8VpTfeQqqlHSbY8OgfyfHj8zgHNpZjBLos_E,18672
 dayhoff_tools/warehouse.py,sha256=TqV8nex1AluNaL4JuXH5zuu9P7qmE89lSo6f_oViy6U,14965
-dayhoff_tools-1.1.41.dist-info/METADATA,sha256=HgiBVffpoYUtLMGV4uAKXJCyiPVmo39ytRHQ41b6-hg,2843
-dayhoff_tools-1.1.41.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-dayhoff_tools-1.1.41.dist-info/entry_points.txt,sha256=iAf4jteNqW3cJm6CO6czLxjW3vxYKsyGLZ8WGmxamSc,49
-dayhoff_tools-1.1.41.dist-info/RECORD,,
+dayhoff_tools-1.1.43.dist-info/METADATA,sha256=S3WFgeHSXhXJUg6E8nHcSBmbGktzd9B-2A7LIdX1c9k,2843
+dayhoff_tools-1.1.43.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+dayhoff_tools-1.1.43.dist-info/entry_points.txt,sha256=iAf4jteNqW3cJm6CO6czLxjW3vxYKsyGLZ8WGmxamSc,49
+dayhoff_tools-1.1.43.dist-info/RECORD,,