IDEA-CR-58836: PY-39526 SciView of dataframes converts integers to floats

Use by default ".5f" formatting for floats and "%s" for other types. The user-visible formatting will be "%s"

PY-39526 Rename function

Signed-off-by: Elizaveta Shashkova <Elizaveta.Shashkova@jetbrains.com>

GitOrigin-RevId: a8af904b55e703848592efad7d0642ccef12a0d1

python/helpers/pydev/_pydevd_bundle/pydevd_constants.py

@@ -83,6 +83,12 @@ IS_PYTHON_STACKLESS = "stackless" in sys.version.lower()
 CYTHON_SUPPORTED = False
 
 NUMPY_NUMERIC_TYPES = "biufc"
+NUMPY_FLOATING_POINT_TYPES = "fc"
+# b boolean
+# i signed integer
+# u unsigned integer
+# f floating-point
+# c complex floating-point
 
 try:
     import platform

python/helpers/pydev/_pydevd_bundle/pydevd_thrift.py

@@ -15,7 +15,8 @@ from _pydevd_bundle.pydevd_constants import dict_iter_items, dict_keys, IS_PY3K,
     NUMPY_NUMERIC_TYPES
 from _pydevd_bundle.pydevd_extension_api import TypeResolveProvider, StrPresentationProvider
 from _pydevd_bundle.pydevd_utils import take_first_n_coll_elements, is_numeric_container, is_pandas_container, pandas_to_str, is_string
-from _pydevd_bundle.pydevd_vars import get_label, array_default_format, is_able_to_format_number, MAXIMUM_ARRAY_SIZE
+from _pydevd_bundle.pydevd_vars import get_label, array_default_format, is_able_to_format_number, MAXIMUM_ARRAY_SIZE, \
+    get_column_formatter_by_type, DEFAULT_DF_FORMAT
 from pydev_console.pydev_protocol import DebugValue, GetArrayResponse, ArrayData, ArrayHeaders, ColHeader, RowHeader, \
     UnsupportedArrayTypeException, ExceedingArrayDimensionsException
 
@@ -529,7 +530,7 @@ def dataframe_to_thrift_struct(df, name, roffset, coffset, rows, cols, format):
                     kind = "O"
             format = array_default_format(kind)
         else:
-            format = array_default_format("f")
+            format = array_default_format(DEFAULT_DF_FORMAT)
     array_chunk.format = "%" + format
 
     if (rows, cols) == (-1, -1):
@@ -560,7 +561,7 @@ def dataframe_to_thrift_struct(df, name, roffset, coffset, rows, cols, format):
     cols = df.shape[1] if dim > 1 else 1
 
     def col_to_format(c):
-        return format if dtypes[c] in NUMPY_NUMERIC_TYPES and format else array_default_format(dtypes[c])
+        return get_column_formatter_by_type(format, dtypes[c])
 
     iat = df.iat if dim == 1 or len(df.columns.unique()) == len(df.columns) else df.iloc
 

python/helpers/pydev/_pydevd_bundle/pydevd_vars.py

@@ -4,9 +4,9 @@
 import math
 import pickle
 
-from _pydev_imps._pydev_saved_modules import thread
 from _pydev_bundle.pydev_imports import quote
-from _pydevd_bundle.pydevd_constants import get_frame, get_current_thread_id, xrange, NUMPY_NUMERIC_TYPES
+from _pydev_imps._pydev_saved_modules import thread
+from _pydevd_bundle.pydevd_constants import get_frame, get_current_thread_id, xrange, NUMPY_NUMERIC_TYPES, NUMPY_FLOATING_POINT_TYPES
 from _pydevd_bundle.pydevd_custom_frames import get_custom_frame
 from _pydevd_bundle.pydevd_xml import ExceptionOnEvaluate, get_type, var_to_xml
 
@@ -28,6 +28,7 @@ from _pydev_bundle.pydev_imports import Exec, execfile
 from _pydevd_bundle.pydevd_utils import to_string, VariableWithOffset
 
 SENTINEL_VALUE = []
+DEFAULT_DF_FORMAT = "s"
 
 # ------------------------------------------------------------------------------------------------------ class for errors
 
@@ -584,6 +585,16 @@ def array_to_meta_xml(array, name, format):
     return array, slice_to_xml(slice, rows, cols, format, type, bounds), rows, cols, format
 
 
+def get_column_formatter_by_type(initial_format, column_type):
+    if column_type in NUMPY_NUMERIC_TYPES and initial_format:
+        if column_type in NUMPY_FLOATING_POINT_TYPES and initial_format.strip() == DEFAULT_DF_FORMAT:
+            # use custom formatting for floats when default formatting is set
+            return array_default_format(column_type)
+        return initial_format
+    else:
+        return array_default_format(column_type)
+
+
 def array_default_format(type):
     if type == 'f':
         return '.5f'
@@ -625,7 +636,7 @@ def dataframe_to_xml(df, name, roffset, coffset, rows, cols, format):
                     kind = 'O'
             format = array_default_format(kind)
         else:
-            format = array_default_format('f')
+            format = array_default_format(DEFAULT_DF_FORMAT)
 
     xml = slice_to_xml(name, num_rows, num_cols, format, "", (0, 0))
 
@@ -657,7 +668,7 @@ def dataframe_to_xml(df, name, roffset, coffset, rows, cols, format):
     cols = df.shape[1] if dim > 1 else 1
 
     def col_to_format(c):
-        return format if dtypes[c] in NUMPY_NUMERIC_TYPES and format else array_default_format(dtypes[c])
+        return get_column_formatter_by_type(format, dtypes[c])
 
     iat = df.iat if dim == 1 or len(df.columns.unique()) == len(df.columns) else df.iloc
 

python/testData/debug/test_dataframe.py

@@ -2,8 +2,8 @@ import pandas as pd
 import numpy as np
 df1 = pd.DataFrame({'row': [0, 1, 2],
                     'One_X': [1.1, 1.1, 1.1],
-                    'One_Y': [1.2, 1.2, 1.2],
-                    'Two_X': [1.11, 1.11, 1.11],
+                    'Year': [2018, 2019, 2020],
+                    'Winner': [True, False, True],
                     'Two_Y': [1.22, 1.22, 1.22]})
 print(df1)  ###line 8
 

python/testSrc/com/jetbrains/env/python/PythonDataViewerTest.java

@@ -111,15 +111,33 @@ public class PythonDataViewerTest extends PyEnvTestCase {
 
   @Test
   @Staging
-  public void testDataFrameFormatting() {
+  public void testDataFrameFloatFormatting() {
     runPythonTest(new PyDataFrameDebuggerTask(getRelativeTestDataPath(), "test_dataframe.py", ImmutableSet.of(7)) {
       @Override
       public void testing() throws Exception {
         doTest("df1", 3, 5, (varName, session) -> getChunk(varName, "%.2f", session), arrayChunk -> {
           Object[][] data = arrayChunk.getData();
           assertEquals("'1.10'", data[0][1].toString());
-          assertEquals("'1.20'", data[0][2].toString());
           assertEquals("'1.22'", data[1][4].toString());
+          assertEquals("'2019.00'", data[1][2].toString());
+          assertEquals("'1.00'", data[2][3].toString());
+        });
+      }
+    });
+  }
+
+  @Test
+  @Staging
+  public void testDataFrameDefaultFormatting() {
+    runPythonTest(new PyDataFrameDebuggerTask(getRelativeTestDataPath(), "test_dataframe.py", ImmutableSet.of(7)) {
+      @Override
+      public void testing() throws Exception {
+        doTest("df1", 3, 5, (varName, session) -> getChunk(varName, "%", session), arrayChunk -> {
+          Object[][] data = arrayChunk.getData();
+          assertEquals("'1.10000'", data[0][1].toString());
+          assertEquals("'1.22000'", data[1][4].toString());
+          assertEquals("'2019'", data[1][2].toString());
+          assertEquals("'True'", data[2][3].toString());
         });
       }
     });