Merge pull request #13211 from xokdvium/pos-table

libexpr: Actually cache line information in PosTable

src/libutil/include/nix/util/lru-cache.hh

@@ -33,6 +33,18 @@ private:
     Data data;
     LRU lru;
 
+    /**
+     * Move this item to the back of the LRU list.
+     */
+    void promote(LRU::iterator it)
+    {
+        /* Think of std::list iterators as stable pointers to the list node,
+         * which never get invalidated. Thus, we can reuse the same lru list
+         * element and just splice it to the back of the list without the need
+         * to update its value in the key -> list iterator map. */
+        lru.splice(/*pos=*/lru.end(), /*other=*/lru, it);
+    }
+
 public:
 
     LRUCache(size_t capacity)
@@ -83,7 +95,9 @@ public:
     /**
      * Look up an item in the cache. If it exists, it becomes the most
      * recently used item.
-     * */
+     *
+     * @returns corresponding cache entry, std::nullopt if it's not in the cache
+     */
     template<typename K>
     std::optional<Value> get(const K & key)
     {
@@ -91,20 +105,30 @@ public:
         if (i == data.end())
             return {};
 
-        /**
-         * Move this item to the back of the LRU list.
-         *
-         * Think of std::list iterators as stable pointers to the list node,
-         * which never get invalidated. Thus, we can reuse the same lru list
-         * element and just splice it to the back of the list without the need
-         * to update its value in the key -> list iterator map.
-         */
         auto & [it, value] = i->second;
-        lru.splice(/*pos=*/lru.end(), /*other=*/lru, it.it);
-
+        promote(it.it);
         return value;
     }
 
+    /**
+     * Look up an item in the cache. If it exists, it becomes the most
+     * recently used item.
+     *
+     * @returns mutable pointer to the corresponding cache entry, nullptr if
+     * it's not in the cache
+     */
+    template<typename K>
+    Value * getOrNullptr(const K & key)
+    {
+        auto i = data.find(key);
+        if (i == data.end())
+            return nullptr;
+
+        auto & [it, value] = i->second;
+        promote(it.it);
+        return &value;
+    }
+
     size_t size() const noexcept
     {
         return data.size();

src/libutil/include/nix/util/pos-table.hh

@@ -4,6 +4,7 @@
 #include <cstdint>
 #include <vector>
 
+#include "nix/util/lru-cache.hh"
 #include "nix/util/pos-idx.hh"
 #include "nix/util/position.hh"
 #include "nix/util/sync.hh"
@@ -37,10 +38,20 @@ public:
     };
 
 private:
+    /**
+     * Vector of byte offsets (in the virtual input buffer) of initial line character's position.
+     * Sorted by construction. Binary search over it allows for efficient translation of arbitrary
+     * byte offsets in the virtual input buffer to its line + column position.
+     */
     using Lines = std::vector<uint32_t>;
+    /**
+     * Cache from byte offset in the virtual buffer of Origins -> @ref Lines in that origin.
+     */
+    using LinesCache = LRUCache<uint32_t, Lines>;
 
     std::map<uint32_t, Origin> origins;
-    mutable Sync<std::map<uint32_t, Lines>> lines;
+
+    mutable Sync<LinesCache> linesCache;
 
     const Origin * resolve(PosIdx p) const
     {
@@ -56,6 +67,11 @@ private:
     }
 
 public:
+    PosTable(std::size_t linesCacheCapacity = 65536)
+        : linesCache(linesCacheCapacity)
+    {
+    }
+
     Origin addOrigin(Pos::Origin origin, size_t size)
     {
         uint32_t offset = 0;

src/libutil/pos-table.cc

@@ -15,21 +15,35 @@ Pos PosTable::operator[](PosIdx p) const
     const auto offset = origin->offsetOf(p);
 
     Pos result{0, 0, origin->origin};
-    auto lines = this->lines.lock();
-    auto linesForInput = (*lines)[origin->offset];
+    auto linesCache = this->linesCache.lock();
 
-    if (linesForInput.empty()) {
-        auto source = result.getSource().value_or("");
-        const char * begin = source.data();
-        for (Pos::LinesIterator it(source), end; it != end; it++)
-            linesForInput.push_back(it->data() - begin);
-        if (linesForInput.empty())
-            linesForInput.push_back(0);
+    /* Try the origin's line cache */
+    const auto * linesForInput = linesCache->getOrNullptr(origin->offset);
+
+    auto fillCacheForOrigin = [](std::string_view content) {
+        auto contentLines = Lines();
+
+        const char * begin = content.data();
+        for (Pos::LinesIterator it(content), end; it != end; it++)
+            contentLines.push_back(it->data() - begin);
+        if (contentLines.empty())
+            contentLines.push_back(0);
+
+        return contentLines;
+    };
+
+    /* Calculate line offsets and fill the cache */
+    if (!linesForInput) {
+        auto originContent = result.getSource().value_or("");
+        linesCache->upsert(origin->offset, fillCacheForOrigin(originContent));
+        linesForInput = linesCache->getOrNullptr(origin->offset);
     }
-    // as above: the first line starts at byte 0 and is always present
-    auto lineStartOffset = std::prev(std::upper_bound(linesForInput.begin(), linesForInput.end(), offset));
 
-    result.line = 1 + (lineStartOffset - linesForInput.begin());
+    assert(linesForInput);
+
+    // as above: the first line starts at byte 0 and is always present
+    auto lineStartOffset = std::prev(std::upper_bound(linesForInput->begin(), linesForInput->end(), offset));
+    result.line = 1 + (lineStartOffset - linesForInput->begin());
     result.column = 1 + (offset - *lineStartOffset);
     return result;
 }