ENH: Optimize nrows in read_excel

pandas/io/excel/_base.py

@@ -380,7 +380,7 @@ def get_sheet_by_index(self, index):
         pass
 
     @abc.abstractmethod
-    def get_sheet_data(self, sheet, convert_float):
+    def get_sheet_data(self, sheet, convert_float, header, skiprows, nrows):
         pass
 
     def parse(
@@ -436,7 +436,7 @@ def parse(
             else:  # assume an integer if not a string
                 sheet = self.get_sheet_by_index(asheetname)
 
-            data = self.get_sheet_data(sheet, convert_float)
+            data = self.get_sheet_data(sheet, convert_float, header, skiprows, nrows)
             usecols = _maybe_convert_usecols(usecols)
 
             if not data:

pandas/io/excel/_odfreader.py

@@ -1,11 +1,12 @@
-from typing import List
+from typing import List, Optional, Sequence
 
-from pandas._typing import FilePathOrBuffer, Scalar
+from pandas._typing import FilePathOrBuffer, Scalar, Union
 from pandas.compat._optional import import_optional_dependency
 
 import pandas as pd
 
 from pandas.io.excel._base import _BaseExcelReader
+from pandas.io.parsers import _validate_integer
 
 
 class _ODFReader(_BaseExcelReader):
@@ -63,7 +64,14 @@ def get_sheet_by_name(self, name: str):
 
         raise ValueError(f"sheet {name} not found")
 
-    def get_sheet_data(self, sheet, convert_float: bool) -> List[List[Scalar]]:
+    def get_sheet_data(
+        self,
+        sheet,
+        convert_float: bool,
+        header: Optional[Union[int, Sequence[int]]],
+        skiprows: Optional[Union[int, Sequence[int]]],
+        nrows: Optional[int],
+    ) -> List[List[Scalar]]:
         """
         Parse an ODF Table into a list of lists
         """
@@ -79,7 +87,29 @@ def get_sheet_data(self, sheet, convert_float: bool) -> List[List[Scalar]]:
 
         table: List[List[Scalar]] = []
 
+        if nrows is not None:
+            _validate_integer("nrows", nrows)
+        header = 0 if header is None else header
+        skiprows = 0 if skiprows is None else skiprows
+        if isinstance(header, list) or isinstance(skiprows, list):
+            nrows = None
+
         for i, sheet_row in enumerate(sheet_rows):
+
+            if nrows is not None:
+                if header > 1:
+                    header -= 1
+                    table.append([])
+                    continue
+                elif skiprows > 0:
+                    skiprows -= 1
+                    table.append([])
+                    continue
+                if nrows >= 0:
+                    nrows -= 1
+                else:
+                    break
+
             sheet_cells = [x for x in sheet_row.childNodes if x.qname in cell_names]
             empty_cells = 0
             table_row: List[Scalar] = []

pandas/io/excel/_openpyxl.py

@@ -1,12 +1,13 @@
-from typing import List
+from typing import List, Optional, Sequence
 
 import numpy as np
 
-from pandas._typing import FilePathOrBuffer, Scalar
+from pandas._typing import FilePathOrBuffer, Scalar, Union
 from pandas.compat._optional import import_optional_dependency
 
 from pandas.io.excel._base import ExcelWriter, _BaseExcelReader
 from pandas.io.excel._util import _validate_freeze_panes
+from pandas.io.parsers import _validate_integer
 
 
 class _OpenpyxlWriter(ExcelWriter):
@@ -524,9 +525,39 @@ def _convert_cell(self, cell, convert_float: bool) -> Scalar:
 
         return cell.value
 
-    def get_sheet_data(self, sheet, convert_float: bool) -> List[List[Scalar]]:
+    def get_sheet_data(
+        self,
+        sheet,
+        convert_float: bool,
+        header: Optional[Union[int, Sequence[int]]],
+        skiprows: Optional[Union[int, Sequence[int]]],
+        nrows: Optional[int],
+    ) -> List[List[Scalar]]:
         data: List[List[Scalar]] = []
+
+        if nrows is not None:
+            _validate_integer("nrows", nrows)
+        header = 0 if header is None else header
+        skiprows = 0 if skiprows is None else skiprows
+        if isinstance(header, list) or isinstance(skiprows, list):
+            nrows = None
+
         for row in sheet.rows:
+
+            if nrows is not None:
+                if header > 1:
+                    header -= 1
+                    data.append([])
+                    continue
+                elif skiprows > 0:
+                    skiprows -= 1
+                    data.append([])
+                    continue
+                if nrows >= 0:
+                    nrows -= 1
+                else:
+                    break
+
             data.append([self._convert_cell(cell, convert_float) for cell in row])
 
         return data

pandas/io/excel/_pyxlsb.py

@@ -62,7 +62,9 @@ def _convert_cell(self, cell, convert_float: bool) -> Scalar:
 
         return cell.v
 
-    def get_sheet_data(self, sheet, convert_float: bool) -> List[List[Scalar]]:
+    def get_sheet_data(
+        self, sheet, convert_float: bool, header, skiprows, nrows
+    ) -> List[List[Scalar]]:
         return [
             [self._convert_cell(c, convert_float) for c in r]
             for r in sheet.rows(sparse=False)

pandas/io/excel/_xlrd.py

@@ -1,10 +1,13 @@
 from datetime import time
+from typing import List, Optional, Sequence
 
 import numpy as np
 
+from pandas._typing import Scalar, Union
 from pandas.compat._optional import import_optional_dependency
 
 from pandas.io.excel._base import _BaseExcelReader
+from pandas.io.parsers import _validate_integer
 
 
 class _XlrdReader(_BaseExcelReader):
@@ -46,7 +49,14 @@ def get_sheet_by_name(self, name):
     def get_sheet_by_index(self, index):
         return self.book.sheet_by_index(index)
 
-    def get_sheet_data(self, sheet, convert_float):
+    def get_sheet_data(
+        self,
+        sheet,
+        convert_float,
+        header: Optional[Union[int, Sequence[int]]],
+        skiprows: Optional[Union[int, Sequence[int]]],
+        nrows: Optional[int],
+    ) -> List[List[Scalar]]:
         from xlrd import (
             xldate,
             XL_CELL_DATE,
@@ -95,9 +105,30 @@ def _parse_cell(cell_contents, cell_typ):
                     cell_contents = val
             return cell_contents
 
-        data = []
+        data: List[List[Scalar]] = []
 
+        if nrows is not None:
+            _validate_integer("nrows", nrows)
+        header = 0 if header is None else header
+        skiprows = 0 if skiprows is None else skiprows
+        if isinstance(header, list) or isinstance(skiprows, list):
+            nrows = None
         for i in range(sheet.nrows):
+
+            if nrows is not None:
+                if header > 1:
+                    header -= 1
+                    data.append([])
+                    continue
+                elif skiprows > 0:
+                    skiprows -= 1
+                    data.append([])
+                    continue
+                if nrows >= 0:
+                    nrows -= 1
+                else:
+                    break
+
             row = [
                 _parse_cell(value, typ)
                 for value, typ in zip(sheet.row_values(i), sheet.row_types(i))