commit 3c1ba2fb29430577d062eb151db8d2ed279ab8e9
author Andy Phan <andyphan@chromium.org> Sat Aug 17 00:55:02 2024 +0000
committer Pdfium LUCI CQ <pdfium-scoped@luci-project-accounts.iam.gserviceaccount.com> Sat Aug 17 00:55:02 2024 +0000
tree 0024da63ea8ead4e16b8d7f8e82825af5c9c65bb
parent 6e56b3afb19b8d7abc3f87aa7c39673791aa402c

Refactor CPDF_SimpleParser part 3

Remove the size variable, which is usually just `cur_pos_ - start_pos`.
Handle special cases accordingly.

Avoid stepping back the current position. Instead, prefer processing the
current character before incrementing the position. This makes the logic
easier to follow and makes the size easier to understand.

Also update a double negative conditional to use positive instead.

Change-Id: I2b8a9bccd8b4e8b7c3ccb260ff46e898d2965328
Reviewed-on: https://pdfium-review.googlesource.com/c/pdfium/+/123412
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Andy Phan <andyphan@chromium.org>

core/fpdfapi/parser/cpdf_simple_parser.cpp

diff --git a/core/fpdfapi/parser/cpdf_simple_parser.cpp b/core/fpdfapi/parser/cpdf_simple_parser.cpp
index 9856e2f..38c1bdb 100644
--- a/core/fpdfapi/parser/cpdf_simple_parser.cpp
+++ b/core/fpdfapi/parser/cpdf_simple_parser.cpp
@@ -48,35 +48,29 @@
     }
   }
 
-  uint8_t size = 0;
   uint32_t start_position = cur_position_ - 1;
   if (PDFCharIsDelimiter(cur_char)) {
     // Find names
     if (cur_char == '/') {
-      while (true) {
-        if (cur_position_ >= data_.size()) {
-          break;
+      while (cur_position_ < data_.size()) {
+        cur_char = data_[cur_position_];
+        // Stop parsing after encountering a whitespace or delimiter.
+        if (PDFCharIsWhitespace(cur_char) || PDFCharIsDelimiter(cur_char)) {
+          return ByteStringView(
+              data_.subspan(start_position, cur_position_ - start_position));
         }
-
-        cur_char = data_[cur_position_++];
-        if (!PDFCharIsOther(cur_char) && !PDFCharIsNumeric(cur_char)) {
-          cur_position_--;
-          size = cur_position_ - start_position;
-          break;
-        }
+        ++cur_position_;
       }
-      return ByteStringView(data_.subspan(start_position, size));
+      return ByteStringView();
     }
 
-    size = 1;
     if (cur_char == '<') {
       if (cur_position_ >= data_.size()) {
-        return ByteStringView(data_.subspan(start_position, size));
+        return ByteStringView(
+            data_.subspan(start_position, cur_position_ - start_position));
       }
       cur_char = data_[cur_position_++];
-      if (cur_char == '<') {
-        size = 2;
-      } else {
+      if (cur_char != '<') {
         while (cur_position_ < data_.size() && data_[cur_position_] != '>') {
           cur_position_++;
         }
@@ -84,18 +78,14 @@
         if (cur_position_ < data_.size()) {
           cur_position_++;
         }
-
-        size = cur_position_ - start_position;
       }
     } else if (cur_char == '>') {
       if (cur_position_ >= data_.size()) {
-        return ByteStringView(data_.subspan(start_position, size));
+        return ByteStringView(
+            data_.subspan(start_position, cur_position_ - start_position));
       }
-      cur_char = data_[cur_position_++];
-      if (cur_char == '>') {
-        size = 2;
-      } else {
-        cur_position_--;
+      if (data_[cur_position_] == '>') {
+        ++cur_position_;
       }
     } else if (cur_char == '(') {
       int level = 1;
@@ -124,21 +114,18 @@
       if (cur_position_ < data_.size()) {
         cur_position_++;
       }
-
-      size = cur_position_ - start_position;
     }
-    return ByteStringView(data_.subspan(start_position, size));
+    return ByteStringView(
+        data_.subspan(start_position, cur_position_ - start_position));
   }
 
-  size = 1;
   while (cur_position_ < data_.size()) {
-    cur_char = data_[cur_position_++];
-
+    cur_char = data_[cur_position_];
     if (PDFCharIsDelimiter(cur_char) || PDFCharIsWhitespace(cur_char)) {
-      cur_position_--;
       break;
     }
-    size++;
+    ++cur_position_;
   }
-  return ByteStringView(data_.subspan(start_position, size));
+  return ByteStringView(
+      data_.subspan(start_position, cur_position_ - start_position));
 }