lamindb 0.65.0__tar.gz → 0.65.1__tar.gz

{lamindb-0.65.0 → lamindb-0.65.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: lamindb
-Version: 0.65.0
+Version: 0.65.1
 Summary: A data framework for biology.
 Author-email: Lamin Labs <open-source@lamin.ai>
 Requires-Python: >=3.8
@@ -10,7 +10,7 @@ Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Requires-Dist: lnschema_core==0.60.0
-Requires-Dist: lamindb_setup==0.63.0
+Requires-Dist: lamindb_setup==0.63.1
 Requires-Dist: lamin_utils==0.13.0
 Requires-Dist: lamin_cli==0.5.0
 Requires-Dist: rapidfuzz

{lamindb-0.65.0 → lamindb-0.65.1}/docs/changelog.md

@@ -1,20 +1,27 @@
 # Changelog
 
-
 ```{eval-rst}
 .. role:: small
 ```
 
-### 0.65
+## 0.65
 
-### 0.65.0 {small}`2024-01-01`
+### 0.65.1 {small}`2024-01-05`
+
+- 🩹 Prepare a potential migration of the hub to Django [PR1385](https://github.com/laminlabs/lamindb/pull/1385) [@falexwolf](https://github.com/falexwolf)
+- 🚸 Various improvements [PR1384](https://github.com/laminlabs/lamindb/pull/1384) [@falexwolf](https://github.com/falexwolf)
+- 🩹 Track suffix of requirements.txt [PR1383](https://github.com/laminlabs/lamindb/pull/1383) [@falexwolf](https://github.com/falexwolf)
+- ✨ Add outer join and categories caching to Collection.mapped [PR1380](https://github.com/laminlabs/lamindb/pull/1380) [@Koncopd](https://github.com/Koncopd)
+- ♻️ Except memory error [PR1382](https://github.com/laminlabs/lamindb/pull/1382) [@falexwolf](https://github.com/falexwolf)
+
+### 0.65.0 {small}`2024-01-02`
 
-- 🚸 Speed up transfer and enable transfer parents [PR1371](https://github.com/laminlabs/lamindb/pull/1371) [@sunnyosun](https://github.com/sunnyosun)
 - 🚚 Rename `Dataset` to `Collection` [PR1377](https://github.com/laminlabs/lamindb/pull/1377) [@falexwolf](https://github.com/falexwolf)
-- 🎨 Version based on `stem_uid` instead of `initial_version_id` and replace `__lamindb_uid_prefix__` with `__transform_stem_uid__` [PR1375](https://github.com/laminlabs/lamindb/pull/1375) [PR1373](https://github.com/laminlabs/lamindb/pull/1373) [@falexwolf](https://github.com/falexwolf)
-- 🎨 Name `.lndb` files by instance id [PR1372](https://github.com/laminlabs/lamindb/pull/1372) [@falexwolf](https://github.com/falexwolf)
 - ✨ Track run environment [PR1368](https://github.com/laminlabs/lamindb/pull/1368) [@falexwolf](https://github.com/falexwolf)
 - ✨ Allow transfer from private instances [PR1370](https://github.com/laminlabs/lamindb/pull/1370) [@falexwolf](https://github.com/falexwolf)
+- 🚸 Speed up transfer and enable transfer parents [PR1371](https://github.com/laminlabs/lamindb/pull/1371) [@sunnyosun](https://github.com/sunnyosun)
+- 🎨 Version based on `stem_uid` instead of `initial_version_id` and replace `__lamindb_uid_prefix__` with `__transform_stem_uid__` [PR1369](https://github.com/laminlabs/lamindb/pull/1369) [PR1375](https://github.com/laminlabs/lamindb/pull/1375) [PR1373](https://github.com/laminlabs/lamindb/pull/1373) [@bpenteado](https://github.com/bpenteado) [@falexwolf](https://github.com/falexwolf)
+- 🎨 Name `.lndb` files by instance id [PR1372](https://github.com/laminlabs/lamindb/pull/1372) [@falexwolf](https://github.com/falexwolf)
 
 ## 0.64
 

{lamindb-0.65.0 → lamindb-0.65.1}/lamindb/__init__.py

@@ -54,7 +54,7 @@ Modules & settings:
 
 """
 
-__version__ = "0.65.0"  # denote a release candidate for 0.1.0 with 0.1rc1
+__version__ = "0.65.1"  # denote a release candidate for 0.1.0 with 0.1rc1
 
 import os as _os
 

{lamindb-0.65.0 → lamindb-0.65.1}/lamindb/_collection.py

@@ -314,7 +314,9 @@ def mapped(
     label_keys: Optional[Union[str, List[str]]] = None,
     join_vars: Optional[Literal["auto", "inner"]] = "auto",
     encode_labels: bool = True,
+    cache_categories: bool = True,
     parallel: bool = False,
+    dtype: Optional[str] = None,
     stream: bool = False,
     is_run_input: Optional[bool] = None,
 ) -> "MappedCollection":
@@ -328,7 +330,15 @@ def mapped(
             path_list.append(artifact.stage())
         else:
             path_list.append(artifact.path)
-    return MappedCollection(path_list, label_keys, join_vars, encode_labels, parallel)
+    return MappedCollection(
+        path_list,
+        label_keys,
+        join_vars,
+        encode_labels,
+        cache_categories,
+        parallel,
+        dtype,
+    )
 
 
 # docstring handled through attach_func_to_class_method

{lamindb-0.65.0 → lamindb-0.65.1}/lamindb/_parents.py

@@ -317,13 +317,13 @@ def _get_all_parent_runs(data: Union[Artifact, Collection]) -> List:
             inputs_run = (
                 r.__getattribute__(f"input_{name}s").all().filter(visibility=1).list()
             )
-            if name == "file":
+            if name == "artifact":
                 inputs_run += r.input_collections.all().filter(visibility=1).list()
             run_inputs_outputs += [(inputs_run, r)]
             outputs_run = (
                 r.__getattribute__(f"output_{name}s").all().filter(visibility=1).list()
             )
-            if name == "file":
+            if name == "artifact":
                 outputs_run += r.output_collections.all().filter(visibility=1).list()
             run_inputs_outputs += [(r, outputs_run)]
             inputs += inputs_run
@@ -337,8 +337,11 @@ def _get_all_child_runs(data: Union[Artifact, Collection]) -> List:
     all_runs: Set[Run] = set()
     run_inputs_outputs = []
 
-    runs = {f.run for f in data.run.__getattribute__(f"output_{name}s").all()}
-    if name == "file":
+    if data.run is not None:
+        runs = {f.run for f in data.run.__getattribute__(f"output_{name}s").all()}
+    else:
+        runs = set()
+    if name == "artifact" and data.run is not None:
         runs.update(
             {
                 f.run
@@ -352,13 +355,13 @@ def _get_all_child_runs(data: Union[Artifact, Collection]) -> List:
             inputs_run = (
                 r.__getattribute__(f"input_{name}s").all().filter(visibility=1).list()
             )
-            if name == "file":
+            if name == "artifact":
                 inputs_run += r.input_collections.all().filter(visibility=1).list()
             run_inputs_outputs += [(inputs_run, r)]
             outputs_run = (
                 r.__getattribute__(f"output_{name}s").all().filter(visibility=1).list()
             )
-            if name == "file":
+            if name == "artifact":
                 outputs_run += r.output_collections.all().filter(visibility=1).list()
             run_inputs_outputs += [(r, outputs_run)]
             child_runs.update(
@@ -366,7 +369,7 @@ def _get_all_child_runs(data: Union[Artifact, Collection]) -> List:
                     **{f"input_{name}s__id__in": [i.id for i in outputs_run]}
                 ).list()
             )
-            if name == "file":
+            if name == "artifact":
                 child_runs.update(
                     Run.filter(
                         input_collections__id__in=[i.id for i in outputs_run]

{lamindb-0.65.0 → lamindb-0.65.1}/lamindb/_registry.py

@@ -469,7 +469,11 @@ def save(self, *args, **kwargs) -> None:
     if result is not None:
         init_self_from_db(self, result)
     else:
-        super(Registry, self).save(*args, **kwargs)
+        # here, we can't use the parents argument
+        save_kwargs = kwargs.copy()
+        if "parents" in save_kwargs:
+            save_kwargs.pop("parents")
+        super(Registry, self).save(*args, **save_kwargs)
     if db is not None and db != "default":
         if hasattr(self, "labels"):
             from copy import copy

{lamindb-0.65.0 → lamindb-0.65.1}/lamindb/dev/__init__.py

@@ -24,6 +24,7 @@ Functionality of data registries:
    FeatureManager
    LabelManager
    IsTree
+   IsVersioned
 
 Functionality of metadata registries:
 
@@ -51,7 +52,14 @@ Auxiliary tools:
 """
 
 from lamin_utils._inspect import InspectResult
-from lnschema_core.models import CanValidate, Data, HasParents, IsTree, Registry
+from lnschema_core.models import (
+    CanValidate,
+    Data,
+    HasParents,
+    IsTree,
+    IsVersioned,
+    Registry,
+)
 
 from lamindb._query_manager import QueryManager
 from lamindb._query_set import QuerySet

{lamindb-0.65.0 → lamindb-0.65.1}/lamindb/dev/_data.py

@@ -113,8 +113,7 @@ def describe(self: Data):
         "created_by": "👤",
         "transform": _transform_emoji(self.transform),
         "run": "👣",
-        "initial_version": "🔖",
-        "file": "📄",
+        "artifact": "📄",
     }
     if len(foreign_key_fields) > 0:  # always True for Artifact and Collection
         record_msg = f"{colors.green(model_name)}{__repr__(self, include_foreign_keys=False).lstrip(model_name)}"
@@ -209,7 +208,7 @@ def add_labels(
 ) -> None:
     """{}."""
     if self._state.adding:
-        raise ValueError("Please save the file/collection before adding a label!")
+        raise ValueError("Please save the artifact/collection before adding a label!")
 
     if isinstance(records, (QuerySet, QuerySet.__base__)):  # need to have both
         records = records.list()
@@ -331,7 +330,7 @@ def add_labels(
                             id=old_feature_set_link.feature_set_id
                         ).one()
                         logger.info(
-                            "no file links to it anymore, deleting feature set"
+                            "nothing links to it anymore, deleting feature set"
                             f" {old_feature_set}"
                         )
                         old_feature_set.delete()
@@ -368,7 +367,7 @@ def _track_run_input(
         if run is None:
             if settings.track_run_inputs:
                 logger.hint(
-                    "you can auto-track this file as a run input by calling"
+                    "you can auto-track these data as a run input by calling"
                     " `ln.track()`"
                 )
         # assume we have a run record
@@ -390,7 +389,7 @@ def _track_run_input(
                     track_run_input = True
                 else:
                     logger.hint(
-                        "track this file as a run input by passing `is_run_input=True`"
+                        "track these data as a run input by passing `is_run_input=True`"
                     )
     else:
         track_run_input = is_run_input

{lamindb-0.65.0 → lamindb-0.65.1}/lamindb/dev/_label_manager.py

@@ -23,11 +23,12 @@ def get_labels_as_dict(self: Data):
     ).items():
         if related_name in {
             "feature_sets",
-            "files",
+            "artifacts",
             "input_of",
             "collections",
             "source_of",
             "report_of",
+            "environment_of",
         }:
             continue
         if self.id is not None:

{lamindb-0.65.0 → lamindb-0.65.1}/lamindb/dev/_mapped_collection.py

@@ -57,10 +57,14 @@ class MappedCollection:
         self,
         path_list: List[Union[str, PathLike]],
         label_keys: Optional[Union[str, List[str]]] = None,
-        join_vars: Optional[Literal["auto", "inner"]] = "auto",
+        join_vars: Optional[Literal["auto", "inner", "outer"]] = "auto",
         encode_labels: bool = True,
+        cache_categories: bool = True,
         parallel: bool = False,
+        dtype: Optional[str] = None,
     ):
+        assert join_vars in {None, "auto", "inner", "outer"}
+
         self.storages = []  # type: ignore
         self.conns = []  # type: ignore
         self.parallel = parallel
@@ -86,8 +90,15 @@ class MappedCollection:
 
         self.encode_labels = encode_labels
         self.label_keys = [label_keys] if isinstance(label_keys, str) else label_keys
-        if self.label_keys is not None and self.encode_labels:
-            self._make_encoders(self.label_keys)
+        if self.label_keys is not None:
+            if cache_categories:
+                self._cache_categories(self.label_keys)
+            else:
+                self._cache_cats: dict = {}
+            if self.encode_labels:
+                self._make_encoders(self.label_keys)
+
+        self._dtype = dtype
 
         self._closed = False
 
@@ -104,6 +115,18 @@ class MappedCollection:
             self.conns.append(conn)
             self.storages.append(storage)
 
+    def _cache_categories(self, label_keys: list):
+        self._cache_cats = {}
+        decode = np.frompyfunc(lambda x: x.decode("utf-8"), 1, 1)
+        for label in label_keys:
+            self._cache_cats[label] = []
+            for storage in self.storages:
+                with _Connect(storage) as store:
+                    cats = self.get_categories(store, label)
+                    if cats is not None:
+                        cats = decode(cats) if isinstance(cats[0], bytes) else cats[...]
+                    self._cache_cats[label].append(cats)
+
     def _make_encoders(self, label_keys: list):
         self.encoders = []
         for label in label_keys:
@@ -115,20 +138,31 @@ class MappedCollection:
         for storage in self.storages:
             with _Connect(storage) as store:
                 var_list.append(_safer_read_index(store["var"]))
+
+        self.var_joint = None
         if self.join_vars == "auto":
             vars_eq = all(var_list[0].equals(vrs) for vrs in var_list[1:])
             if vars_eq:
                 self.join_vars = None
                 return
             else:
-                self.join_vars = "inner"
+                self.var_joint = reduce(pd.Index.intersection, var_list)
+                if len(self.var_joint) > 0:
+                    self.join_vars = "inner"
+                else:
+                    self.join_vars = "outer"
+
         if self.join_vars == "inner":
-            self.var_joint = reduce(pd.Index.intersection, var_list)
-            if len(self.var_joint) == 0:
-                raise ValueError(
-                    "The provided AnnData objects don't have shared varibales."
-                )
+            if self.var_joint is None:
+                self.var_joint = reduce(pd.Index.intersection, var_list)
+                if len(self.var_joint) == 0:
+                    raise ValueError(
+                        "The provided AnnData objects don't have shared varibales."
+                    )
             self.var_indices = [vrs.get_indexer(self.var_joint) for vrs in var_list]
+        elif self.join_vars == "outer":
+            self.var_joint = reduce(pd.Index.union, var_list)
+            self.var_indices = [self.var_joint.get_indexer(vrs) for vrs in var_list]
 
     def __len__(self):
         return self.n_obs
@@ -137,15 +171,21 @@ class MappedCollection:
         obs_idx = self.indices[idx]
         storage_idx = self.storage_idx[idx]
         if self.var_indices is not None:
-            var_idxs = self.var_indices[storage_idx]
+            var_idxs_join = self.var_indices[storage_idx]
         else:
-            var_idxs = None
+            var_idxs_join = None
 
         with _Connect(self.storages[storage_idx]) as store:
-            out = [self.get_data_idx(store, obs_idx, var_idxs)]
+            out = [self.get_data_idx(store, obs_idx, var_idxs_join)]
             if self.label_keys is not None:
                 for i, label in enumerate(self.label_keys):
-                    label_idx = self.get_label_idx(store, obs_idx, label)
+                    if label in self._cache_cats:
+                        cats = self._cache_cats[label][storage_idx]
+                        if cats is None:
+                            cats = []
+                    else:
+                        cats = None
+                    label_idx = self.get_label_idx(store, obs_idx, label, cats)
                     if self.encode_labels:
                         label_idx = self.encoders[i][label_idx]
                     out.append(label_idx)
@@ -155,26 +195,50 @@ class MappedCollection:
         self,
         storage: StorageType,  # type: ignore
         idx: int,
-        var_idxs: Optional[list] = None,
+        var_idxs_join: Optional[list] = None,
         layer_key: Optional[str] = None,
     ):
         """Get the index for the data."""
         layer = storage["X"] if layer_key is None else storage["layers"][layer_key]  # type: ignore
         if isinstance(layer, ArrayTypes):  # type: ignore
-            # todo: better way to select variables
-            return layer[idx] if var_idxs is None else layer[idx][var_idxs]
+            layer_idx = layer[idx]
+            if self.join_vars is None:
+                result = layer_idx
+                if self._dtype is not None:
+                    result = result.astype(self._dtype, copy=False)
+            elif self.join_vars == "outer":
+                dtype = layer_idx.dtype if self._dtype is None else self._dtype
+                result = np.zeros(len(self.var_joint), dtype=dtype)
+                result[var_idxs_join] = layer_idx
+            else:  # inner join
+                result = layer_idx[var_idxs_join]
+                if self._dtype is not None:
+                    result = result.astype(self._dtype, copy=False)
+            return result
         else:  # assume csr_matrix here
             data = layer["data"]
             indices = layer["indices"]
             indptr = layer["indptr"]
             s = slice(*(indptr[idx : idx + 2]))
-            # this requires more memory than csr_matrix when var_idxs is not None
-            # but it is faster
-            layer_idx = np.zeros(layer.attrs["shape"][1])
-            layer_idx[indices[s]] = data[s]
-            return layer_idx if var_idxs is None else layer_idx[var_idxs]
+            data_s = data[s]
+            dtype = data_s.dtype if self._dtype is None else self._dtype
+            if self.join_vars == "outer":
+                layer_idx = np.zeros(len(self.var_joint), dtype=dtype)
+                layer_idx[var_idxs_join[indices[s]]] = data_s
+            else:
+                layer_idx = np.zeros(layer.attrs["shape"][1], dtype=dtype)
+                layer_idx[indices[s]] = data_s
+                if self.join_vars == "inner":
+                    layer_idx = layer_idx[var_idxs_join]
+            return layer_idx
 
-    def get_label_idx(self, storage: StorageType, idx: int, label_key: str):  # type: ignore
+    def get_label_idx(
+        self,
+        storage: StorageType,
+        idx: int,
+        label_key: str,
+        categories: Optional[list] = None,
+    ):
         """Get the index for the label by key."""
         obs = storage["obs"]  # type: ignore
         # how backwards compatible do we want to be here actually?
@@ -186,9 +250,11 @@ class MappedCollection:
                 label = labels[idx]
             else:
                 label = labels["codes"][idx]
-
-        cats = self.get_categories(storage, label_key)
-        if cats is not None:
+        if categories is not None:
+            cats = categories
+        else:
+            cats = self.get_categories(storage, label_key)
+        if cats is not None and len(cats) > 0:
             label = cats[label]
         if isinstance(label, bytes):
             label = label.decode("utf-8")
@@ -215,11 +281,14 @@ class MappedCollection:
         """Get merged labels."""
         labels_merge = []
         decode = np.frompyfunc(lambda x: x.decode("utf-8"), 1, 1)
-        for storage in self.storages:
+        for i, storage in enumerate(self.storages):
             with _Connect(storage) as store:
                 codes = self.get_codes(store, label_key)
                 labels = decode(codes) if isinstance(codes[0], bytes) else codes
-                cats = self.get_categories(store, label_key)
+                if label_key in self._cache_cats:
+                    cats = self._cache_cats[label_key][i]
+                else:
+                    cats = self.get_categories(store, label_key)
                 if cats is not None:
                     cats = decode(cats) if isinstance(cats[0], bytes) else cats
                     labels = cats[labels]
@@ -230,9 +299,12 @@ class MappedCollection:
         """Get merged categories."""
         cats_merge = set()
         decode = np.frompyfunc(lambda x: x.decode("utf-8"), 1, 1)
-        for storage in self.storages:
+        for i, storage in enumerate(self.storages):
             with _Connect(storage) as store:
-                cats = self.get_categories(store, label_key)
+                if label_key in self._cache_cats:
+                    cats = self._cache_cats[label_key][i]
+                else:
+                    cats = self.get_categories(store, label_key)
                 if cats is not None:
                     cats = decode(cats) if isinstance(cats[0], bytes) else cats
                     cats_merge.update(cats)

{lamindb-0.65.0 → lamindb-0.65.1}/lamindb/dev/_run_context.py

@@ -33,7 +33,9 @@ msg_manual_init = (
 )
 
 
-class UpdateNbWithNonInteractiveEditorError(Exception):
+# we don't want a real error here, as this is so frequent
+# in VSCode
+class UpdateNbWithNonInteractiveEditor(SystemExit):
     pass
 
 
@@ -230,7 +232,7 @@ class run_context:
                             "it looks like you are running ln.track() from a "
                             "notebook!\nplease install nbproject: pip install nbproject"
                         )
-                    elif isinstance(e, UpdateNbWithNonInteractiveEditorError):
+                    elif isinstance(e, UpdateNbWithNonInteractiveEditor):
                         raise e
                     elif isinstance(e, (NotebookNotSavedError, NoTitleError)):
                         raise e
@@ -435,7 +437,7 @@ class run_context:
                 cls._notebook_meta = metadata  # type: ignore
             else:
                 msg = msg_manual_init.format(notebook_path=notebook_path_str)
-                raise UpdateNbWithNonInteractiveEditorError(msg)
+                raise UpdateNbWithNonInteractiveEditor(msg)
 
         if _env in ("lab", "notebook"):
             # save the notebook in case that title was updated
@@ -450,7 +452,7 @@ class run_context:
             is_interactive = _seconds_modified(_filepath) < 1.5  # should be ~1 sec
             if not is_interactive and needs_init:
                 msg = msg_manual_init.format(notebook_path=_filepath)
-                raise UpdateNbWithNonInteractiveEditorError(msg)
+                raise UpdateNbWithNonInteractiveEditor(msg)
 
             nbproject_id = metadata["id"]
             nbproject_version = metadata["version"]
@@ -509,7 +511,7 @@ class run_context:
                 cls._notebook_meta = metadata  # type: ignore
             else:
                 msg = msg_manual_init.format(notebook_path=filepath)
-                raise UpdateNbWithNonInteractiveEditorError(msg)
+                raise UpdateNbWithNonInteractiveEditor(msg)
         else:
             from lamin_cli._transform import update_transform_source_metadata
 

{lamindb-0.65.0 → lamindb-0.65.1}/lamindb/dev/_track_environment.py

@@ -6,9 +6,13 @@ from lnschema_core.models import Run
 
 
 def track_environment(run: Run) -> None:
-    filepath = ln_setup.settings.storage.cache_dir / f"run_env_pip_{run.uid}"
+    filepath = ln_setup.settings.storage.cache_dir / f"run_env_pip_{run.uid}.txt"
     # create a requirements.txt
     # we don't create a conda environment.yml mostly for its slowness
-    result = subprocess.run(f"pip freeze > {str(filepath)}", shell=True)
-    if result.returncode == 0:
+    try:
+        result = subprocess.run(f"pip freeze > {str(filepath)}", shell=True)
+    except OSError as e:
+        result = None
+        logger.warning(f"could not run pip freeze with error {e}")
+    if result is not None and result.returncode == 0:
         logger.info(f"tracked pip freeze > {str(filepath)}")

{lamindb-0.65.0 → lamindb-0.65.1}/pyproject.toml

@@ -17,7 +17,7 @@ classifiers = [
 dependencies = [
     # Lamin PINNED packages
     "lnschema_core==0.60.0",
-    "lamindb_setup==0.63.0",
+    "lamindb_setup==0.63.1",
     "lamin_utils==0.13.0",
     "lamin_cli==0.5.0",
     # others

{lamindb-0.65.0 → lamindb-0.65.1}/sub/lamin-cli/lamin_cli/_transform.py

@@ -66,17 +66,26 @@ def update_transform_source_metadata(
     else:
         is_notebook = False
         stem_uid, version = get_script_metadata(filepath)
-    logger.important(
-        f"transform is tracked with stem_uid='{stem_uid}', version: '{version}'"
+    from lamin_utils._base62 import encodebytes
+    import hashlib
+    # the following line is duplicated with get_transform_kwargs_from_stem_uid
+    # in lamindb - we should move it, e.g., to lamin-utils
+    # it also occurs a few lines below
+    uid_ext = encodebytes(hashlib.md5(version.encode()).digest())[:4]
+    # it simply looks better here to not use the logger because we won't have an emoji also for the subsequent
+    # input question
+    print(
+        f"Transform is tracked with stem_uid='{stem_uid}' & version='{version}' (uid='{stem_uid}{uid_ext}')"
     )
     updated = False
-    # ask for generating a new uid prefix
+    # ask for generating a new stem uid
+    response = "bump"
     if not bump_version:
         if os.getenv("LAMIN_TESTING") is None:
-            response = input("Do you want to generate a new uid prefix? (y/n) ")
+            response = input("To create a new stem uid, type 'new'. To bump the version, type 'bump' or a custom version: ")
         else:
-            response = "y"
-        if response == "y":
+            response = "new"
+        if response == "new":
             new_stem_uid = nbproject_id()
             updated = True
         else:
@@ -84,21 +93,27 @@ def update_transform_source_metadata(
     new_version = version
     if bump_version:
         new_stem_uid = stem_uid
-        if os.getenv("LAMIN_TESTING") is None:
-            new_version = input(
-                f"The current version is '{version}' - please type the new version: "
-            )
+        if response == "bump":
+            try:
+                new_version = str(int(version) + 1)
+            except ValueError:
+                new_version = input(
+                    f"The current version is '{version}' - please type the new version: "
+                )
         else:
-            new_version = str(int(version) + 1)
+            new_version = response
         updated = new_version != version
     if updated and run_from_cli:
+        display_info = f"version='{new_version}'" if bump_version else f"stem_uid='{new_stem_uid}'"
+        new_uid_ext = encodebytes(hashlib.md5(new_version.encode()).digest())[:4]
+        display_info += f" (uid='{new_stem_uid}{new_uid_ext}')"
         if is_notebook:
-            logger.save("updated notebook")
+            logger.save(f"updated notebook: {display_info}")
             content.metadata["nbproject"]["id"] = new_stem_uid
             content.metadata["nbproject"]["version"] = new_version
             write_notebook(content, filepath)
         else:
-            logger.save("updated script")
+            logger.save(f"updated script: {display_info}")
             old_metadata = (
                 f'__transform_stem_uid__ = "{stem_uid}"\n__version__ = "{version}"\n'
             )
@@ -284,9 +299,9 @@ def save(filepath: str) -> Optional[str]:
         source_code.save()
         transform.source_code = source_code
     # track environment
-    filepath = lamindb_setup.settings.storage.cache_dir / f"run_env_pip_{run.uid}"
+    filepath = lamindb_setup.settings.storage.cache_dir / f"run_env_pip_{run.uid}.txt"
     if filepath.exists():
-        artifact = ln.Artifact(filepath, description="requirements.txt")
+        artifact = ln.Artifact(filepath, description="requirements.txt", visibility=0)
         if artifact._state.adding:
             artifact.save()
         run.environment = artifact