Fix nits in CPDF_Function and derived classes.

- Switch away from FX_Alloc2D() when it is safe to do so.
- Use FX_SAFE_SIZE_T to avoid a potential underflow with FX_Realloc().
- Mark the v_Call() input as const.
- Check Call() and v_Call() return results.

Change-Id: I6ec3122b23e0f137f88aa8d85a9675154a1b1706
Reviewed-on: https://pdfium-review.googlesource.com/30802
Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>

1 files changed, 1 insertions, 1 deletions

diff --git a/core/fpdfapi/page/cpdf_sampledfunc.cpp b/core/fpdfapi/page/cpdf_sampledfunc.cpp
index 78c4f2c755..205ce91eaa 100644
--- a/core/fpdfapi/page/cpdf_sampledfunc.cpp
+++ b/core/fpdfapi/page/cpdf_sampledfunc.cpp
@@ -91,7 +91,7 @@ bool CPDF_SampledFunc::v_Init(CPDF_Object* pObj,
   return true;
 }
 
-bool CPDF_SampledFunc::v_Call(float* inputs, float* results) const {
+bool CPDF_SampledFunc::v_Call(const float* inputs, float* results) const {
   int pos = 0;
   CFX_FixedBufGrow<float, 16> encoded_input_buf(m_nInputs);
   float* encoded_input = encoded_input_buf;