[GraphBolt][Standalone] Reduce and mark dependencies on DGL. (#7499)

python/dgl/graphbolt/__init__.py

@@ -4,6 +4,7 @@
 
 import torch
 
+### FROM DGL @todo
 from .._ffi import libinfo
 
 
@@ -46,12 +47,13 @@ def load_graphbolt():
 from .itemset import *
 from .item_sampler import *
 from .minibatch_transformer import *
+from .internal_utils import *
 from .negative_sampler import *
 from .sampled_subgraph import *
 from .subgraph_sampler import *
+from .external_utils import add_reverse_edges, exclude_seed_edges
 from .internal import (
     compact_csc_format,
     unique_and_compact,
     unique_and_compact_csc_formats,
 )
-from .utils import add_reverse_edges, exclude_seed_edges

python/dgl/graphbolt/base.py

@@ -20,7 +20,7 @@
 from torch.utils.data import functional_datapipe
 from torchdata.datapipes.iter import IterDataPipe
 
-from ..utils import recursive_apply
+from .internal_utils import recursive_apply
 
 __all__ = [
     "CANONICAL_ETYPE_DELIMITER",

python/dgl/graphbolt/impl/fused_csc_sampling_graph.py

@@ -7,12 +7,8 @@
 
 import torch
 
-from dgl.utils import recursive_apply
-
-from ...base import EID, ETYPE, NID, NTYPE
-from ...convert import to_homogeneous
-from ...heterograph import DGLGraph
 from ..base import etype_str_to_tuple, etype_tuple_to_str, ORIGINAL_EDGE_ID
+from ..internal_utils import recursive_apply
 from ..sampling_graph import SamplingGraph
 from .sampled_subgraph_impl import CSCFormatBase, SampledSubgraphImpl
 
@@ -1538,11 +1534,15 @@ def load_from_shared_memory(
 
 
 def from_dglgraph(
-    g: DGLGraph,
+    DGLGraphInstance,
     is_homogeneous: bool = False,
     include_original_edge_id: bool = False,
 ) -> FusedCSCSamplingGraph:
     """Convert a DGLGraph to FusedCSCSamplingGraph."""
+    from dgl.base import EID, ETYPE, NID, NTYPE
+    from dgl.convert import to_homogeneous
+
+    g = DGLGraphInstance
 
     homo_g, ntype_count, _ = to_homogeneous(
         g, ndata=g.ndata, edata=g.edata, return_count=True

python/dgl/graphbolt/impl/legacy_dataset.py

@@ -2,7 +2,6 @@
 
 from typing import List, Union
 
-from dgl.data import AsNodePredDataset, DGLDataset
 from ..base import etype_tuple_to_str
 from ..dataset import Dataset, Task
 from ..itemset import HeteroItemSet, ItemSet
@@ -16,7 +15,7 @@
 class LegacyDataset(Dataset):
     """A Graphbolt dataset for legacy DGLDataset."""
 
-    def __init__(self, legacy: DGLDataset):
+    def __init__(self, legacy):
         # Only supports single graph cases.
         assert len(legacy) == 1
         graph = legacy[0]
@@ -28,7 +27,7 @@ def __init__(self, legacy: DGLDataset):
         else:
             self._init_as_heterogeneous_node_pred(legacy)
 
-    def _init_as_heterogeneous_node_pred(self, legacy: DGLDataset):
+    def _init_as_heterogeneous_node_pred(self, legacy):
         def _init_item_set_dict(idx, labels):
             item_set_dict = {}
             for key in idx.keys():
@@ -87,7 +86,9 @@ def _init_item_set_dict(idx, labels):
                 "Only support heterogeneous ogn node pred dataset"
             )
 
-    def _init_as_homogeneous_node_pred(self, legacy: DGLDataset):
+    def _init_as_homogeneous_node_pred(self, legacy):
+        from dgl.data import AsNodePredDataset
+
         legacy = AsNodePredDataset(legacy)
 
         # Initialize tasks.

python/dgl/graphbolt/impl/ondisk_dataset.py

@@ -13,8 +13,6 @@
 import torch
 import yaml
 
-from ...base import dgl_warning
-from ...data.utils import download, extract_archive
 from ..base import etype_str_to_tuple, ORIGINAL_EDGE_ID
 from ..dataset import Dataset, Task
 from ..internal import (
@@ -25,6 +23,7 @@
     read_data,
     read_edges,
 )
+from ..internal_utils import download, extract_archive, gb_warning
 from ..itemset import HeteroItemSet, ItemSet
 from ..sampling_graph import SamplingGraph
 from .fused_csc_sampling_graph import (
@@ -457,7 +456,7 @@ def preprocess_ondisk_dataset(
                 is_feature=True,
             )
         if has_edge_feature_data and not include_original_edge_id:
-            dgl_warning("Edge feature is stored, but edge IDs are not saved.")
+            gb_warning("Edge feature is stored, but edge IDs are not saved.")
 
     # 6. Save tasks and train/val/test split according to the output_config.
     if input_config.get("tasks", None):
@@ -831,7 +830,7 @@ def _init_tasks(
         if selected_tasks:
             not_found_tasks = set(selected_tasks) - task_names
             if len(not_found_tasks):
-                dgl_warning(
+                gb_warning(
                     f"Below tasks are not found in YAML: {not_found_tasks}. Skipped."
                 )
         return ret
@@ -887,7 +886,7 @@ def _init_tvt_set(
 
     def _init_all_nodes_set(self, graph) -> Union[ItemSet, HeteroItemSet]:
         if graph is None:
-            dgl_warning(
+            gb_warning(
                 "`all_nodes_set` is returned as None, since graph is None."
             )
             return None

python/dgl/graphbolt/impl/ondisk_metadata.py

@@ -5,7 +5,8 @@
 
 import pydantic
 
-from ...utils import version
+from ..internal_utils import version
+
 
 __all__ = [
     "OnDiskFeatureDataFormat",