Validate source text size and fix empty buffer bugs. (#1113)

There's currently a bug with empty files, in that it initializes SourceBuffer with an invalid StringRef that results in a crash. That got me looking at the std::optional TODO, but the issue is that there are really three states:

- Buffered
- mmapped (not buffered)
- Moved out of (no longer initialized)

Technically an optional could work if we initialize the buffer on move out, indicating the mmap is gone. But the mode setup felt better to me.

And then this also adds the size check. Which is really how I started looking at this.

toolchain/source/source_buffer.h

@@ -35,34 +35,18 @@ class SourceBuffer {
  public:
   static auto CreateFromText(llvm::Twine text,
                              llvm::StringRef filename = "/text")
-      -> SourceBuffer;
+      -> llvm::Expected<SourceBuffer>;
   static auto CreateFromFile(llvm::StringRef filename)
       -> llvm::Expected<SourceBuffer>;
 
   // Use one of the factory functions above to create a source buffer.
   SourceBuffer() = delete;
 
-  // Cannot copy as there may be non-trivial owned file data, see the class
+  // Cannot copy as there may be non-trivial owned file data; see the class
   // comment for details.
   SourceBuffer(const SourceBuffer& arg) = delete;
 
-  SourceBuffer(SourceBuffer&& arg) noexcept
-      : filename_(std::move(arg.filename_)),
-        text_(arg.text_),
-        is_string_rep_(arg.is_string_rep_) {
-    // The easy case in when we don't need to transfer an allocated string
-    // representation.
-    if (!arg.is_string_rep_) {
-      // Take ownership of a non-string representation by clearing its text.
-      arg.text_ = llvm::StringRef();
-      return;
-    }
-
-    // If the argument is using a string rep we need to move that storage over
-    // and recreate our text `StringRef` to point at our storage.
-    new (&string_storage_) std::string(std::move(arg.string_storage_));
-    text_ = string_storage_;
-  }
+  SourceBuffer(SourceBuffer&& arg) noexcept;
 
   ~SourceBuffer();
 
@@ -71,28 +55,21 @@ class SourceBuffer {
   [[nodiscard]] auto Text() const -> llvm::StringRef { return text_; }
 
  private:
-  SourceBuffer(llvm::StringRef fake_filename, std::string buffer_text)
-      : filename_(fake_filename.str()),
-        is_string_rep_(true),
-        string_storage_(std::move(buffer_text)) {
-    text_ = string_storage_;
-  }
-
-  explicit SourceBuffer(llvm::StringRef filename)
-      : filename_(filename.str()), text_(), is_string_rep_(false) {}
-
-  std::string filename_;
-
-  llvm::StringRef text_;
-
-  bool is_string_rep_;
-
-  // We use a transparent union to avoid constructing the storage.
-  // FIXME: We should replace this and the boolean with an optional which would
-  // be much simpler.
-  union {
-    std::string string_storage_;
+  enum class ContentMode {
+    Uninitialized,
+    MMapped,
+    Owned,
   };
+
+  // Constructor for mmapped content.
+  SourceBuffer(std::string filename, llvm::StringRef text);
+  // Constructor for owned content.
+  SourceBuffer(std::string filename, std::string text);
+
+  ContentMode content_mode_;
+  std::string filename_;
+  std::string text_storage_;
+  llvm::StringRef text_;
 };
 
 }  // namespace Carbon