Introduce a simple line-by-line iterator type into the Support library.

This is an iterator which you can build around a MemoryBuffer. It will
iterate through the non-empty, non-comment lines of the buffer as
a forward iterator. It should be small and reasonably fast (although it
could be made much faster if anyone cares, I don't really...).

This will be used to more simply support the text-based sample
profile file format, and is largely based on the original patch by
Diego. I've re-worked the style of it and separated it from the work of
producing a MemoryBuffer from a file which both simplifies the interface
and makes it easier to test.

The style of the API follows the C++ standard naming conventions to fit
in better with iterators in general, much like the Path and FileSystem
interfaces follow standard-based naming conventions.

llvm-svn: 198068

1 files changed, 68 insertions, 0 deletions

diff --git a/llvm/lib/Support/LineIterator.cpp b/llvm/lib/Support/LineIterator.cpp
new file mode 100644
index 00000000000..c14f96bf714
--- /dev/null
+++ b/llvm/lib/Support/LineIterator.cpp
@@ -0,0 +1,68 @@
+//===- LineIterator.cpp - Implementation of line iteration ----------------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/LineIterator.h"
+#include "llvm/Support/MemoryBuffer.h"
+
+using namespace llvm;
+
+line_iterator::line_iterator(const MemoryBuffer &Buffer, char CommentMarker)
+    : Buffer(Buffer.getBufferSize() ? &Buffer : 0),
+      CommentMarker(CommentMarker), LineNumber(1),
+      CurrentLine(Buffer.getBufferSize() ? Buffer.getBufferStart() : 0, 0) {
+  // Ensure that if we are constructed on a non-empty memory buffer that it is
+  // a null terminated buffer.
+  if (Buffer.getBufferSize()) {
+    assert(Buffer.getBufferEnd()[0] == '\0');
+    advance();
+  }
+}
+
+void line_iterator::advance() {
+  assert(Buffer && "Cannot advance past the end!");
+
+  const char *Pos = CurrentLine.end();
+  assert(Pos == Buffer->getBufferStart() || *Pos == '\n' || *Pos == '\0');
+  size_t Length = 0;
+
+  if (CommentMarker == '\0') {
+    // If we're not stripping comments, this is simpler.
+    while (Pos[Length] == '\n')
+      ++Length;
+    Pos += Length;
+    LineNumber += Length;
+    Length = 0;
+  } else {
+    // Skip comments and count line numbers, which is a bit more complex.
+    for (;;) {
+      if (*Pos == CommentMarker)
+        do {
+          ++Pos;
+        } while (*Pos != '\0' && *Pos != '\n');
+      if (*Pos != '\n')
+        break;
+      ++Pos;
+      ++LineNumber;
+    }
+  }
+
+  if (*Pos == '\0') {
+    // We've hit the end of the buffer, reset ourselves to the end state.
+    Buffer = 0;
+    CurrentLine = StringRef();
+    return;
+  }
+
+  // Measure the line.
+  do {
+    ++Length;
+  } while (Pos[Length] != '\0' && Pos[Length] != '\n');
+
+  CurrentLine = StringRef(Pos, Length);
+}