summaryrefslogtreecommitdiff
path: root/third_party/llvm-project/include/llvm/ADT/edit_distance.h
diff options
context:
space:
mode:
authorAlon Zakai <azakai@google.com>2019-12-19 09:04:08 -0800
committerGitHub <noreply@github.com>2019-12-19 09:04:08 -0800
commit4d28d3f32e7f213e300b24bc61c3f0ac9d6e1ab6 (patch)
tree91bffc2d47b1fe4bba01e7ada77006ef340bd138 /third_party/llvm-project/include/llvm/ADT/edit_distance.h
parent0048f5b004ddf50e750aa335d0be314a73852058 (diff)
downloadbinaryen-4d28d3f32e7f213e300b24bc61c3f0ac9d6e1ab6.tar.gz
binaryen-4d28d3f32e7f213e300b24bc61c3f0ac9d6e1ab6.tar.bz2
binaryen-4d28d3f32e7f213e300b24bc61c3f0ac9d6e1ab6.zip
DWARF parsing and writing support using LLVM (#2520)
This imports LLVM code for DWARF handling. That code has the Apache 2 license like us. It's also the same code used to emit DWARF in the common toolchain, so it seems like a safe choice. This adds two passes: --dwarfdump which runs the same code LLVM runs for llvm-dwarfdump. This shows we can parse it ok, and will be useful for debugging. And --dwarfupdate writes out the DWARF sections (unchanged from what we read, so it just roundtrips - for updating we need #2515). This puts LLVM in thirdparty which is added here. All the LLVM code is behind USE_LLVM_DWARF, which is on by default, but off in JS for now, as it increases code size by 20%. This current approach imports the LLVM files directly. This is not how they are intended to be used, so it required a bunch of local changes - more than I expected actually, for the platform-specific stuff. For now this seems to work, so it may be good enough, but in the long term we may want to switch to linking against libllvm. A downside to doing that is that binaryen users would need to have an LLVM build, and even in the waterfall builds we'd have a problem - while we ship LLVM there anyhow, we constantly update it, which means that binaryen would need to be on latest llvm all the time too (which otherwise, given DWARF is quite stable, we might not need to constantly update). An even larger issue is that as I did this work I learned about how DWARF works in LLVM, and while the reading code is easy to reuse, the writing code is trickier. The main code path is heavily integrated with the MC layer, which we don't have - we might want to create a "fake MC layer" for that, but it sounds hard. Instead, there is the YAML path which is used mostly for testing, and which can convert DWARF to and from YAML and from binary. Using the non-YAML parts there, we can convert binary DWARF to the YAML layer's nice Info data, then convert that to binary. This works, however, this is not the path LLVM uses normally, and it supports only some basic DWARF sections - I had to add ranges support, in fact. So if we need more complex things, we may end up needing to use the MC layer approach, or consider some other DWARF library. However, hopefully that should not affect the core binaryen code which just calls a library for DWARF stuff. Helps #2400
Diffstat (limited to 'third_party/llvm-project/include/llvm/ADT/edit_distance.h')
-rw-r--r--third_party/llvm-project/include/llvm/ADT/edit_distance.h102
1 files changed, 102 insertions, 0 deletions
diff --git a/third_party/llvm-project/include/llvm/ADT/edit_distance.h b/third_party/llvm-project/include/llvm/ADT/edit_distance.h
new file mode 100644
index 000000000..4f5134008
--- /dev/null
+++ b/third_party/llvm-project/include/llvm/ADT/edit_distance.h
@@ -0,0 +1,102 @@
+//===-- llvm/ADT/edit_distance.h - Array edit distance function --- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines a Levenshtein distance function that works for any two
+// sequences, with each element of each sequence being analogous to a character
+// in a string.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_ADT_EDIT_DISTANCE_H
+#define LLVM_ADT_EDIT_DISTANCE_H
+
+#include "llvm/ADT/ArrayRef.h"
+#include <algorithm>
+#include <memory>
+
+namespace llvm {
+
+/// Determine the edit distance between two sequences.
+///
+/// \param FromArray the first sequence to compare.
+///
+/// \param ToArray the second sequence to compare.
+///
+/// \param AllowReplacements whether to allow element replacements (change one
+/// element into another) as a single operation, rather than as two operations
+/// (an insertion and a removal).
+///
+/// \param MaxEditDistance If non-zero, the maximum edit distance that this
+/// routine is allowed to compute. If the edit distance will exceed that
+/// maximum, returns \c MaxEditDistance+1.
+///
+/// \returns the minimum number of element insertions, removals, or (if
+/// \p AllowReplacements is \c true) replacements needed to transform one of
+/// the given sequences into the other. If zero, the sequences are identical.
+template<typename T>
+unsigned ComputeEditDistance(ArrayRef<T> FromArray, ArrayRef<T> ToArray,
+ bool AllowReplacements = true,
+ unsigned MaxEditDistance = 0) {
+ // The algorithm implemented below is the "classic"
+ // dynamic-programming algorithm for computing the Levenshtein
+ // distance, which is described here:
+ //
+ // http://en.wikipedia.org/wiki/Levenshtein_distance
+ //
+ // Although the algorithm is typically described using an m x n
+ // array, only one row plus one element are used at a time, so this
+ // implementation just keeps one vector for the row. To update one entry,
+ // only the entries to the left, top, and top-left are needed. The left
+ // entry is in Row[x-1], the top entry is what's in Row[x] from the last
+ // iteration, and the top-left entry is stored in Previous.
+ typename ArrayRef<T>::size_type m = FromArray.size();
+ typename ArrayRef<T>::size_type n = ToArray.size();
+
+ const unsigned SmallBufferSize = 64;
+ unsigned SmallBuffer[SmallBufferSize];
+ std::unique_ptr<unsigned[]> Allocated;
+ unsigned *Row = SmallBuffer;
+ if (n + 1 > SmallBufferSize) {
+ Row = new unsigned[n + 1];
+ Allocated.reset(Row);
+ }
+
+ for (unsigned i = 1; i <= n; ++i)
+ Row[i] = i;
+
+ for (typename ArrayRef<T>::size_type y = 1; y <= m; ++y) {
+ Row[0] = y;
+ unsigned BestThisRow = Row[0];
+
+ unsigned Previous = y - 1;
+ for (typename ArrayRef<T>::size_type x = 1; x <= n; ++x) {
+ int OldRow = Row[x];
+ if (AllowReplacements) {
+ Row[x] = std::min(
+ Previous + (FromArray[y-1] == ToArray[x-1] ? 0u : 1u),
+ std::min(Row[x-1], Row[x])+1);
+ }
+ else {
+ if (FromArray[y-1] == ToArray[x-1]) Row[x] = Previous;
+ else Row[x] = std::min(Row[x-1], Row[x]) + 1;
+ }
+ Previous = OldRow;
+ BestThisRow = std::min(BestThisRow, Row[x]);
+ }
+
+ if (MaxEditDistance && BestThisRow > MaxEditDistance)
+ return MaxEditDistance + 1;
+ }
+
+ unsigned Result = Row[n];
+ return Result;
+}
+
+} // End llvm namespace
+
+#endif