[REF] slow np.isclose to math.isclose

camelot/parsers/base.py

@@ -1,9 +1,9 @@
 """Defines a base parser. As well as generic methods for other parsers."""
 
+import math
 import os
 import warnings
 
-import numpy as np
 import pandas as pd
 
 from ..core import Table
@@ -319,7 +319,7 @@ def _group_rows(text, row_tol=2):
  # if type(obj) is LTChar]):
  if row_y is None:
  row_y = t.y0
- elif not np.isclose(row_y, t.y0, atol=row_tol):
+ elif not math.isclose(row_y, t.y0, abs_tol=row_tol):
  rows.append(sorted(temp, key=lambda t: t.x0))
  temp = []
  # We update the row's bottom as we go, to be forgiving if there
@@ -352,8 +352,8 @@ def _merge_columns(cl, column_tol=0):
  else:
  lower = merged[-1]
  if column_tol >= 0:
- if higher[0] <= lower[1] or np.isclose(
- higher[0], lower[1], atol=column_tol
+ if higher[0] <= lower[1] or math.isclose(
+ higher[0], lower[1], abs_tol=column_tol
  ):
  upper_bound = max(lower[1], higher[1])
  lower_bound = min(lower[0], higher[0])
@@ -362,7 +362,7 @@ def _merge_columns(cl, column_tol=0):
  merged.append(higher)
  elif column_tol < 0:
  if higher[0] <= lower[1]:
- if np.isclose(higher[0], lower[1], atol=abs(column_tol)):
+ if math.isclose(higher[0], lower[1], abs_tol=abs(column_tol)):
  merged.append(higher)
  else:
  upper_bound = max(lower[1], higher[1])

camelot/utils.py

@@ -1,6 +1,7 @@
 """General helper utilities to parse the pdf tables."""
 
 import atexit
+import math
 import os
 import random
 import re
@@ -855,7 +856,7 @@ def merge_close_lines(ar, line_tol=2):
  ret.append(a)
  else:
  temp = ret[-1]
- if np.isclose(temp, a, atol=line_tol):
+ if math.isclose(temp, a, abs_tol=line_tol):
  temp = (temp + a) / 2.0
  ret[-1] = temp
  else: