Add v2 of the API (#4)

CMakeLists.txt

@@ -1,12 +1,12 @@
 cmake_minimum_required(VERSION 3.15)
+project(mgcxx VERSION 0.0.3)
 set (CMAKE_CXX_STANDARD 20)
 if (NOT CMAKE_BUILD_TYPE)
     set(CMAKE_BUILD_TYPE "Debug")
 endif()
 set(CMAKE_EXPORT_COMPILE_COMMANDS ON)
 include(ExternalProject)
 include(FetchContent)
-project(cxxtantivy)
 option(ENABLE_TESTS "Enable tests" ON)
 
 # NOTE: Be careful with moving this outside of the if block (it should not be

README.md

@@ -1,11 +1,17 @@
-# cxxtantivy
+# mgcxx (experimental)
 
-## Work in Progress
+A collection of C++ wrappers around Rust libraries.
+The list includes:
+  * full-text search enabled by [tantivy](https://github.com/quickwit-oss/tantivy)
+
+## text_search
 
 ### TODOs
 
-- [ ] Figure out the right API ⏳
-  - [ ] All READ methods (`search`, `aggregate`, `find`) depend on the exact schema -> make it robust
+- [ ] Implement full API
+    - [ ] delete
+    - [ ] update
+- [ ] Polish & test all error messages
 - [ ] Write unit / integration test to compare STRING vs JSON fiels search query syntax.
 - [ ] Figure out what's the right search syntax for a property graph
 - [ ] Add some notion of pagination
@@ -29,14 +35,21 @@
         - [ ] Note [DocAddress](https://docs.rs/tantivy/latest/tantivy/struct.DocAddress.html) is composed of 2 u32 but the `SegmentOrdinal` is tied to the `Searcher` -> is it possible/wise to cache the address (`SegmentId` is UUID)
             - [ ] A [searcher](https://docs.rs/tantivy/latest/tantivy/struct.IndexReader.html#method.searcher) per transaction -> cache `DocAddress` inside Memgraph's `ElementAccessors`?
 - [ ] Implement the stress test by adding & searching to the same index concurrently + large dataset generator.
+- [ ] Consider implementing panic! handler preventing outside process to crash (optionally).
 
 ### NOTEs
 
 * if a field doesn't get specified in the schema, it's ignored
-* `TEXT` means the field will be tokenized and indexed (required to be able to search)
-* Tantivy add_json_object accepts serde_json::map::Map<String, serde_json::value::Value>.
+* `TEXT` means the field will be tokenized and indexed (required to be able to
+  search)
+* Tantivy add_json_object accepts serde_json::map::Map<String, serde_json::value::Value>
+* C++ text-search API is snake case because it's implemented in Rust
+* Writing each document and then committing (writing to disk) will be
+  expensive. In a standard OLTP workload that's a common case -> introduce some
+  form of batching.
 
 ## Resources
 
 * https://fulmicoton.com/posts/behold-tantivy-part2
 * https://stackoverflow.com/questions/37924383/combining-several-static-libraries-into-one-using-cmake
+    --> decided to have 2 separate libraries user code has to link

text_search/ci.sh

@@ -29,23 +29,35 @@ echo "Run config:"
 echo "  full   : $MGCXX_TEXT_SEARCH_CI_FULL"
 echo "  release: $MGCXX_TEXT_SEARCH_CI_RELEASE"
 
+cd "$SCRIPT_DIR/.."
+FILES_TO_FIX=$({ git diff --name-only ; git diff --name-only --staged ; } | sort | uniq | egrep "\.c$|\.cpp$|.cxx$|\.h$|\.hpp$|\.hxx$" || true)
+if [ ! -z "$FILES_TO_FIX" ]; then
+  for file in "${FILES_TO_FIX}"; do
+    clang-format -i -verbose ${file}
+  done
+fi
 cd "$SCRIPT_DIR"
-# TODO(gitbuda): Add clang-format call here.
 cargo fmt
 
 mkdir -p "$SCRIPT_DIR/../build"
 cd "$SCRIPT_DIR/../build"
 if [ "$MGCXX_TEXT_SEARCH_CI_FULL" = true ]; then
   rm -rf ./* && rm -rf .cache
+  # Added here because Cargo.lock is ignored for libraries, but it's not
+  # located under build folder. Rebuilding from scratch should also start clean
+  # from cargo perspective.
+  rm "$SCRIPT_DIR/Cargo.lock" || true
 else
   rm -rf index*
 fi
+
 if [ "$MGCXX_TEXT_SEARCH_CI_RELEASE" = true ]; then
   cmake -DCMAKE_BUILD_TYPE=Release ..
 else
   cmake ..
 fi
 make -j8
+
 cd "$SCRIPT_DIR/../build/text_search"
 ./test_unit
 ./test_bench