Merge remote-tracking branch 'upstream/branch-24.12' into pylibcudf/s…

…trings/convert_intergers
rapidsai · Oct 7, 2024 · 7d93e26 · 7d93e26
2 parents 101be17 + f926a61
commit 7d93e26
Show file tree

Hide file tree

Showing 25 changed files with 364 additions and 106 deletions.
diff --git a/.github/workflows/pr_issue_status_automation.yml b/.github/workflows/pr_issue_status_automation.yml
@@ -62,3 +62,33 @@ jobs:
         UPDATE_ITEM: true
         UPDATE_LINKED_ISSUES: true
       secrets: inherit
+
+    process-branch-name:
+      if: ${{ github.event.pull_request.state == 'open' && needs.get-project-id.outputs.ITEM_PROJECT_ID != '' }}
+      needs: get-project-id
+      runs-on: ubuntu-latest
+      outputs:
+        branch-name: ${{ steps.process-branch-name.outputs.branch-name }}
+      steps:
+        - name: Extract branch name
+          id: process-branch-name
+          run: |
+            branch=${{ github.event.pull_request.base.ref }}
+            release=${branch#branch-}
+            echo "branch-name=$release" >> "$GITHUB_OUTPUT"
+
+    update-release:
+      # This job sets the PR and its linked issues to the release they are targeting
+      uses: rapidsai/shared-workflows/.github/workflows/[email protected]
+      if: ${{ github.event.pull_request.state == 'open' && needs.get-project-id.outputs.ITEM_PROJECT_ID != '' }}
+      needs: [get-project-id, process-branch-name]
+      with:
+        PROJECT_ID: "PVT_kwDOAp2shc4AiNzl"
+        SINGLE_SELECT_FIELD_ID: "PVTSSF_lADOAp2shc4AiNzlzgg52UQ"
+        SINGLE_SELECT_FIELD_NAME: "Release"
+        SINGLE_SELECT_OPTION_VALUE: "${{ needs.process-branch-name.outputs.branch-name }}"
+        ITEM_PROJECT_ID: "${{ needs.get-project-id.outputs.ITEM_PROJECT_ID }}"
+        ITEM_NODE_ID: "${{ github.event.pull_request.node_id }}"
+        UPDATE_ITEM: true
+        UPDATE_LINKED_ISSUES: true
+      secrets: inherit
diff --git a/ci/test_python_cudf.sh b/ci/test_python_cudf.sh
@@ -9,7 +9,7 @@ source ./ci/test_python_common.sh test_python_cudf
 
 rapids-logger "Check GPU usage"
 nvidia-smi
-
+rapids-print-env
 EXITCODE=0
 trap "EXITCODE=1" ERR
 set +e

diff --git a/conda/environments/all_cuda-118_arch-x86_64.yaml b/conda/environments/all_cuda-118_arch-x86_64.yaml
@@ -63,7 +63,7 @@ dependencies:
 - openpyxl
 - packaging
 - pandas
-- pandas>=2.0,<2.2.3dev0
+- pandas>=2.0,<2.2.4dev0
 - pandoc
 - polars>=1.8,<1.9
 - pre-commit

diff --git a/conda/environments/all_cuda-125_arch-x86_64.yaml b/conda/environments/all_cuda-125_arch-x86_64.yaml
@@ -61,7 +61,7 @@ dependencies:
 - openpyxl
 - packaging
 - pandas
-- pandas>=2.0,<2.2.3dev0
+- pandas>=2.0,<2.2.4dev0
 - pandoc
 - polars>=1.8,<1.9
 - pre-commit

diff --git a/conda/recipes/cudf/meta.yaml b/conda/recipes/cudf/meta.yaml
@@ -78,7 +78,7 @@ requirements:
   run:
     - python
     - typing_extensions >=4.0.0
-    - pandas >=2.0,<2.2.3dev0
+    - pandas >=2.0,<2.2.4dev0
     - cupy >=12.0.0
     - numba-cuda >=0.0.13
     - numpy >=1.23,<3.0a0

diff --git a/conda/recipes/pylibcudf/meta.yaml b/conda/recipes/pylibcudf/meta.yaml
@@ -77,7 +77,7 @@ requirements:
   run:
     - python
     - typing_extensions >=4.0.0
-    - pandas >=2.0,<2.2.3dev0
+    - pandas >=2.0,<2.2.4dev0
     - numpy >=1.23,<3.0a0
     - pyarrow>=14.0.0,<18.0.0a0
     - {{ pin_compatible('rmm', max_pin='x.x') }}

diff --git a/cpp/src/io/json/write_json.cu b/cpp/src/io/json/write_json.cu
@@ -170,6 +170,9 @@ struct escape_strings_fn {
                                               rmm::cuda_stream_view stream,
                                               rmm::device_async_resource_ref mr)
   {
+    if (column_v.is_empty()) {  // empty begets empty
+      return make_empty_column(type_id::STRING);
+    }
     auto [offsets_column, chars] =
       cudf::strings::detail::make_strings_children(*this, column_v.size(), stream, mr);
 

diff --git a/cpp/tests/io/json/json_writer.cpp b/cpp/tests/io/json/json_writer.cpp
@@ -70,6 +70,43 @@ TEST_F(JsonWriterTest, EmptyInput)
   EXPECT_EQ(expected_lines, std::string(out_buffer.data(), out_buffer.size()));
 }
 
+TEST_F(JsonWriterTest, EmptyLeaf)
+{
+  cudf::test::strings_column_wrapper col1{""};
+  cudf::test::fixed_width_column_wrapper<cudf::size_type> offsets{0, 0};
+  auto col2 = make_lists_column(1,
+                                offsets.release(),
+                                cudf::test::strings_column_wrapper{}.release(),
+                                0,
+                                rmm::device_buffer{},
+                                cudf::test::get_default_stream());
+  auto col3 = cudf::test::lists_column_wrapper<int>::make_one_empty_row_column();
+  cudf::table_view tbl_view{{col1, *col2, col3}};
+  cudf::io::table_metadata mt{{{"col1"}, {"col2"}, {"col3"}}};
+
+  std::vector<char> out_buffer;
+  auto destination = cudf::io::sink_info(&out_buffer);
+  auto out_options = cudf::io::json_writer_options_builder(destination, tbl_view)
+                       .include_nulls(true)
+                       .metadata(mt)
+                       .lines(false)
+                       .na_rep("null")
+                       .build();
+
+  // Empty columns in table
+  cudf::io::write_json(out_options, cudf::test::get_default_stream());
+  std::string const expected = R"([{"col1":"","col2":[],"col3":[]}])";
+  EXPECT_EQ(expected, std::string(out_buffer.data(), out_buffer.size()));
+
+  // Empty columns in table - JSON Lines
+  out_buffer.clear();
+  out_options.enable_lines(true);
+  cudf::io::write_json(out_options, cudf::test::get_default_stream());
+  std::string const expected_lines = R"({"col1":"","col2":[],"col3":[]})"
+                                     "\n";
+  EXPECT_EQ(expected_lines, std::string(out_buffer.data(), out_buffer.size()));
+}
+
 TEST_F(JsonWriterTest, ErrorCases)
 {
   cudf::test::strings_column_wrapper col1{"a", "b", "c"};

diff --git a/dependencies.yaml b/dependencies.yaml
@@ -602,7 +602,7 @@ dependencies:
         packages:
           - fsspec>=0.6.0
           - &numpy numpy>=1.23,<3.0a0
-          - pandas>=2.0,<2.2.3dev0
+          - pandas>=2.0,<2.2.4dev0
   run_pylibcudf:
     common:
       - output_types: [conda, requirements, pyproject]
@@ -748,6 +748,10 @@ dependencies:
             packages:
               - *numba-cuda-dep
               - pandas==2.0.*
+          - matrix: {dependencies: "latest"}
+            packages:
+              - numba-cuda==0.0.15
+              - pandas==2.2.3
           - matrix:
             packages:
       - output_types: conda

diff --git a/python/cudf/cudf/_lib/string_casting.pyx b/python/cudf/cudf/_lib/string_casting.pyx
@@ -6,7 +6,6 @@ from cudf._lib.scalar import as_device_scalar
 from cudf._lib.types import SUPPORTED_NUMPY_TO_LIBCUDF_TYPES
 
 from libcpp.memory cimport unique_ptr
-from libcpp.string cimport string
 from libcpp.utility cimport move
 
 from pylibcudf.libcudf.column.column cimport column
@@ -15,11 +14,6 @@ from pylibcudf.libcudf.strings.convert.convert_floats cimport (
     from_floats as cpp_from_floats,
     to_floats as cpp_to_floats,
 )
-from pylibcudf.libcudf.strings.convert.convert_ipv4 cimport (
-    integers_to_ipv4 as cpp_integers_to_ipv4,
-    ipv4_to_integers as cpp_ipv4_to_integers,
-    is_ipv4 as cpp_is_ipv4,
-)
 from pylibcudf.libcudf.types cimport data_type, type_id
 
 from cudf._lib.types cimport underlying_type_t_type_id
@@ -504,12 +498,11 @@ def timedelta2int(Column input_col, dtype, format):
 
     """
     dtype = dtype_to_pylibcudf_type(dtype)
-    cdef string c_timestamp_format = format.encode('UTF-8')
     return Column.from_pylibcudf(
         plc.strings.convert.convert_durations.to_durations(
             input_col.to_pylibcudf(mode="read"),
             dtype,
-            c_timestamp_format
+            format
         )
     )
 
@@ -528,12 +521,10 @@ def int2timedelta(Column input_col, str format):
     A Column with Timedelta represented in string format
 
     """
-
-    cdef string c_duration_format = format.encode('UTF-8')
     return Column.from_pylibcudf(
         plc.strings.convert.convert_durations.from_durations(
             input_col.to_pylibcudf(mode="read"),
-            c_duration_format
+            format
         )
     )
 
@@ -551,14 +542,10 @@ def int2ip(Column input_col):
     A Column with integer represented in string ipv4 format
 
     """
-
-    cdef column_view input_column_view = input_col.view()
-    cdef unique_ptr[column] c_result
-    with nogil:
-        c_result = move(
-            cpp_integers_to_ipv4(input_column_view))
-
-    return Column.from_unique_ptr(move(c_result))
+    plc_column = plc.strings.convert.convert_ipv4.integers_to_ipv4(
+        input_col.to_pylibcudf(mode="read")
+    )
+    return Column.from_pylibcudf(plc_column)
 
 
 def ip2int(Column input_col):
@@ -574,14 +561,10 @@ def ip2int(Column input_col):
     A Column with ipv4 represented as integer
 
     """
-
-    cdef column_view input_column_view = input_col.view()
-    cdef unique_ptr[column] c_result
-    with nogil:
-        c_result = move(
-            cpp_ipv4_to_integers(input_column_view))
-
-    return Column.from_unique_ptr(move(c_result))
+    plc_column = plc.strings.convert.convert_ipv4.ipv4_to_integers(
+        input_col.to_pylibcudf(mode="read")
+    )
+    return Column.from_pylibcudf(plc_column)
 
 
 def is_ipv4(Column source_strings):
@@ -590,15 +573,10 @@ def is_ipv4(Column source_strings):
     that have strings in IPv4 format. This format is nnn.nnn.nnn.nnn
     where nnn is integer digits in [0,255].
     """
-    cdef unique_ptr[column] c_result
-    cdef column_view source_view = source_strings.view()
-
-    with nogil:
-        c_result = move(cpp_is_ipv4(
-            source_view
-        ))
-
-    return Column.from_unique_ptr(move(c_result))
+    plc_column = plc.strings.convert.convert_ipv4.is_ipv4(
+        source_strings.to_pylibcudf(mode="read")
+    )
+    return Column.from_pylibcudf(plc_column)
 
 
 def htoi(Column input_col):

diff --git a/python/cudf/cudf/core/_compat.py b/python/cudf/cudf/core/_compat.py
@@ -3,7 +3,7 @@
 import pandas as pd
 from packaging import version
 
-PANDAS_CURRENT_SUPPORTED_VERSION = version.parse("2.2.2")
+PANDAS_CURRENT_SUPPORTED_VERSION = version.parse("2.2.3")
 PANDAS_VERSION = version.parse(pd.__version__)
 
 

diff --git a/python/cudf/pyproject.toml b/python/cudf/pyproject.toml
@@ -28,7 +28,7 @@ dependencies = [
     "numpy>=1.23,<3.0a0",
     "nvtx>=0.2.1",
     "packaging",
-    "pandas>=2.0,<2.2.3dev0",
+    "pandas>=2.0,<2.2.4dev0",
     "ptxcompiler",
     "pyarrow>=14.0.0,<18.0.0a0",
     "pylibcudf==24.12.*,>=0.0.0a0",

diff --git a/python/dask_cudf/pyproject.toml b/python/dask_cudf/pyproject.toml
@@ -23,7 +23,7 @@ dependencies = [
     "cupy-cuda11x>=12.0.0",
     "fsspec>=0.6.0",
     "numpy>=1.23,<3.0a0",
-    "pandas>=2.0,<2.2.3dev0",
+    "pandas>=2.0,<2.2.4dev0",
     "rapids-dask-dependency==24.12.*,>=0.0.0a0",
 ] # This list was generated by `rapids-dependency-file-generator`. To make changes, edit ../../dependencies.yaml and run `rapids-dependency-file-generator`.
 classifiers = [

diff --git a/python/pylibcudf/pylibcudf/libcudf/strings/convert/convert_durations.pxd b/python/pylibcudf/pylibcudf/libcudf/strings/convert/convert_durations.pxd
@@ -10,7 +10,7 @@ from pylibcudf.libcudf.types cimport data_type
 cdef extern from "cudf/strings/convert/convert_durations.hpp" namespace \
         "cudf::strings" nogil:
     cdef unique_ptr[column] to_durations(
-        const column_view & strings_col,
+        const column_view & input,
         data_type duration_type,
         const string & format) except +
 

diff --git a/python/pylibcudf/pylibcudf/libcudf/strings/convert/convert_ipv4.pxd b/python/pylibcudf/pylibcudf/libcudf/strings/convert/convert_ipv4.pxd
@@ -8,11 +8,11 @@ from pylibcudf.libcudf.column.column_view cimport column_view
 cdef extern from "cudf/strings/convert/convert_ipv4.hpp" namespace \
         "cudf::strings" nogil:
     cdef unique_ptr[column] ipv4_to_integers(
-        column_view input_col) except +
+        column_view input) except +
 
     cdef unique_ptr[column] integers_to_ipv4(
-        column_view input_col) except +
+        column_view integers) except +
 
     cdef unique_ptr[column] is_ipv4(
-        column_view source_strings
+        column_view input
     ) except +
diff --git a/python/pylibcudf/pylibcudf/strings/convert/CMakeLists.txt b/python/pylibcudf/pylibcudf/strings/convert/CMakeLists.txt
@@ -13,7 +13,7 @@
 # =============================================================================
 
 set(cython_sources convert_booleans.pyx convert_datetime.pyx convert_durations.pyx
-                   convert_fixed_point.pyx convert_integers.pyx
+                   convert_fixed_point.pyx convert_integers.pyx convert_ipv4.pyx
 )
 
 set(linked_libraries cudf::cudf)

diff --git a/python/pylibcudf/pylibcudf/strings/convert/__init__.pxd b/python/pylibcudf/pylibcudf/strings/convert/__init__.pxd
@@ -5,4 +5,5 @@ from . cimport (
     convert_durations,
     convert_fixed_point,
     convert_integers,
+    convert_ipv4,
 )
diff --git a/python/pylibcudf/pylibcudf/strings/convert/__init__.py b/python/pylibcudf/pylibcudf/strings/convert/__init__.py
@@ -5,4 +5,5 @@
     convert_durations,
     convert_fixed_point,
     convert_integers,
+    convert_ipv4,
 )
diff --git a/python/pylibcudf/pylibcudf/strings/convert/convert_durations.pxd b/python/pylibcudf/pylibcudf/strings/convert/convert_durations.pxd
@@ -8,10 +8,10 @@ from pylibcudf.types cimport DataType
 cpdef Column to_durations(
     Column input,
     DataType duration_type,
-    const string& format
+    str format
 )
 
 cpdef Column from_durations(
-    Column input,
-    const string& format
+    Column durations,
+    str format=*
 )