rgapi 0.1.27__tar.gz → 0.1.28__tar.gz

{rgapi-0.1.27 → rgapi-0.1.28}/Cargo.lock

@@ -330,7 +330,7 @@ checksum = "d6f6ff9a378485b298a5286656da665ba74413d36db0979633275d2e708145d4"
 
 [[package]]
 name = "rgapi"
-version = "0.1.27"
+version = "0.1.28"
 dependencies = [
  "crc32fast",
  "globset",
@@ -443,9 +443,9 @@ checksum = "adb6935a6f5c20170eeceb1a3835a49e12e19d792f6dd344ccc76a985ca5a6ca"
 
 [[package]]
 name = "unicode-ident"
-version = "1.0.25"
+version = "1.0.26"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ab72a15cf68d77cb0987d3684aa8a45c5ef827e8cb49ee2f30bfd7ba2feb519f"
+checksum = "d245f478577f809a851594d02313b640fb437e0bb33866753cff937863096954"
 
 [[package]]
 name = "walkdir"

{rgapi-0.1.27 → rgapi-0.1.28}/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "rgapi"
-version = "0.1.27"
+version = "0.1.28"
 edition = "2024"
 rust-version = "1.91"
 license = "Apache-2.0"

{rgapi-0.1.27 → rgapi-0.1.28}/DEV.md

@@ -38,7 +38,7 @@ The GitHub workflow builds wheels for Python 3.10-3.13 on Linux and macOS and pu
 
 ## Design notes
 
-Python discovery and `paths=True` results contain absolute `pathlib.Path` objects. Structured search rows retain root-relative string labels with `/` separators. Traversal uses `ignore::WalkParallel`, so result order is not part of the API contract. Search results are structured rows; collected result lists use rg-style `str()` and notebook display. `SearchLine.lnhash` is computed with the same CRC-32-based line-content hash format as exhash (`lineno|hash|`, low 16 bits of CRC-32 over the line's UTF-8 bytes); `lnhashs=True` only changes row display, not `line_number` or matching behavior. Path regexes filter returned/searched paths; `skip_dir` and `skip_dir_re` prune traversal through `ignore::WalkBuilder::filter_entry`. Depth, size, filesystem, hidden, and ignore options use `ignore::WalkBuilder` settings. Discovery checks root links with `symlink_metadata` and returns an unfollowed link through a ready stream without a worker. This also supports dangling roots, which the underlying walker would reject. Other discovery roots use absolute paths without canonicalizing. Content searches retain canonical root resolution. `rg_iter` exposes the same parallel search stream that `rg` collects by default; `paths=True` and `count=True` consume that stream with different reducers. Text search skips binary files and invalid UTF-8 content.
+Python discovery and `paths=True` results contain absolute `pathlib.Path` objects. Structured search rows retain root-relative string labels with `/` separators. Traversal uses `ignore::WalkParallel`, so result order is not part of the API contract. Search results are structured rows; collected result lists use rg-style `str()` and notebook display. `SearchLine.lnhash` is computed with the same CRC-32-based line-content hash format as exhash (`lineno|hash|`, low 12 bits of CRC-32 over the line's UTF-8 bytes, encoded as two Base64url characters); `lnhashs=True` only changes row display, not `line_number` or matching behavior. Path regexes filter returned/searched paths; `skip_dir` and `skip_dir_re` prune traversal through `ignore::WalkBuilder::filter_entry`. Depth, size, filesystem, hidden, and ignore options use `ignore::WalkBuilder` settings. Discovery checks root links with `symlink_metadata` and returns an unfollowed link through a ready stream without a worker. This also supports dangling roots, which the underlying walker would reject. Other discovery roots use absolute paths without canonicalizing. Content searches retain canonical root resolution. `rg_iter` exposes the same parallel search stream that `rg` collects by default; `paths=True` and `count=True` consume that stream with different reducers. Text search skips binary files and invalid UTF-8 content.
 
 Streaming engine: `walk.rs` owns the generic machinery. `StreamIter<T>` is the worker-thread-plus-bounded-channel iterator (`sync_channel(8192)`, so producers block rather than buffer without limit when a consumer lags), and `spawn_walk` owns the shared scaffold: walker config, panic catching, cancel flag, and worker thread. `rg_iter` (`T = SearchLine`), `block_iter` (`T = SearchBlock`), `nb_iter` (`T = NbCell`), and `find_iter` (`T = PathBuf`, the path walk) plug entry closures into that engine. Block search reads each file once, searches it once, groups nonblank lines into blocks, maps matching lines to their blocks, and expands context by block index. Each `SearchBlock` carries numeric boundaries plus hashes for its first and last source lines. Python keeps both and chooses the displayed address without another file read.
 

{rgapi-0.1.27 → rgapi-0.1.28}/PKG-INFO

@@ -1,9 +1,9 @@
 Metadata-Version: 2.4
 Name: rgapi
-Version: 0.1.27
+Version: 0.1.28
 Classifier: Programming Language :: Rust
 Classifier: Programming Language :: Python :: Implementation :: CPython
-Requires-Dist: fastcore>=1.14.6
+Requires-Dist: fastcore>=2.2.29
 Requires-Dist: fastship>=0.0.12 ; extra == 'dev'
 Requires-Dist: maturin>=1.0,<2.0 ; extra == 'dev'
 Requires-Dist: pytest ; extra == 'dev'
@@ -168,7 +168,7 @@ rg("TODO", ".", summary=True, context=1, maxlen=120)
 
 The result is `BlockResults`, a list of `SearchBlock` objects. Each block has `path`, `block_index`, `start_line`, `end_line`, `start_lnhash`, `end_lnhash`, `kind`, full `source`, and `matches`.
 
-Matches display as `path:start-end:source`. Context displays as `path:start-end-source`. With `lnhashs=True`, the range uses copyable boundary addresses such as `path:4|a3f2|,6|b1c3|:source`. Newline runs display as `¶`. `maxlen` limits the displayed text without changing `source` or `asdict()`.
+Matches display as `path:start-end:source`. Context displays as `path:start-end-source`. With `lnhashs=True`, the range uses copyable boundary addresses such as `path:4|Py|,6|HD|:source`. Newline runs display as `¶`. `maxlen` limits the displayed text without changing `source` or `asdict()`.
 
 In summary mode, `before_context`, `after_context`, and `context` count neighbouring blocks. `max_results` counts matching blocks and retains their context. `summary=True` cannot be combined with `paths` or `count`. It can be combined with `lnhash` for copyable block boundaries.
 

{rgapi-0.1.27 → rgapi-0.1.28}/README.md

@@ -147,7 +147,7 @@ rg("TODO", ".", summary=True, context=1, maxlen=120)
 
 The result is `BlockResults`, a list of `SearchBlock` objects. Each block has `path`, `block_index`, `start_line`, `end_line`, `start_lnhash`, `end_lnhash`, `kind`, full `source`, and `matches`.
 
-Matches display as `path:start-end:source`. Context displays as `path:start-end-source`. With `lnhashs=True`, the range uses copyable boundary addresses such as `path:4|a3f2|,6|b1c3|:source`. Newline runs display as `¶`. `maxlen` limits the displayed text without changing `source` or `asdict()`.
+Matches display as `path:start-end:source`. Context displays as `path:start-end-source`. With `lnhashs=True`, the range uses copyable boundary addresses such as `path:4|Py|,6|HD|:source`. Newline runs display as `¶`. `maxlen` limits the displayed text without changing `source` or `asdict()`.
 
 In summary mode, `before_context`, `after_context`, and `context` count neighbouring blocks. `max_results` counts matching blocks and retains their context. `summary=True` cannot be combined with `paths` or `count`. It can be combined with `lnhash` for copyable block boundaries.
 

{rgapi-0.1.27 → rgapi-0.1.28}/pyproject.toml

@@ -14,7 +14,7 @@ classifiers = [
     "Programming Language :: Rust",
     "Programming Language :: Python :: Implementation :: CPython",
 ]
-dependencies = ["fastcore>=1.14.6"]
+dependencies = ["fastcore>=2.2.29"]
 
 [project.optional-dependencies]
 dev = ["fastship>=0.0.12", "maturin>=1.0,<2.0", "pytest"]

{rgapi-0.1.27 → rgapi-0.1.28}/src/search.rs

@@ -185,9 +185,11 @@ pub fn compile_regex(pattern: &str, case_sensitive: Option<bool>, smart_case: bo
     builder.build(pattern).map_err(|e| RgApiError::new(e.to_string()))
 }
 
-fn line_hash_u16(line: &str) -> u16 { (crc32fast::hash(line.as_bytes()) & 0xffff) as u16 }
-
-pub(crate) fn format_lnhash(lineno: u64, line: &str) -> String { format!("{}|{:04x}|", lineno, line_hash_u16(line)) }
+pub(crate) fn format_lnhash(lineno: u64, line: &str) -> String {
+    const ALPHABET: &[u8; 64] = b"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_";
+    let hash = crc32fast::hash(line.as_bytes()) as usize;
+    format!("{}|{}{}|", lineno, ALPHABET[(hash >> 6) & 63] as char, ALPHABET[hash & 63] as char)
+}
 
 pub fn search_path(
     path: &Path,

{rgapi-0.1.27 → rgapi-0.1.28}/tests/test_rgapi.py

@@ -185,11 +185,11 @@ def test_depth_size_and_filesystem_options(tmp_path):
     assert [r.path for r in rg("TODO", str(tmp_path), min_depth=2, max_filesize=5)] == ["sub/small.txt"]
 
 
-def test_lnhash_matches_stdlib_crc32(tmp_path):
-    import zlib
+def test_lnhash_matches_fastcore(tmp_path):
+    from fastcore.tools import lnhash as py_hash
     make_tree(tmp_path)
-    row = rg("TODO", str(tmp_path))[0]
-    assert row.lnhash == f"{row.line_number}|{zlib.crc32(row.line.encode()) & 0xffff:04x}|"
+    for row in rg(".", str(tmp_path)):
+        assert row.lnhash == py_hash(row.line_number, row.line)
 
 
 def test_rg_returns_structured_matches_context_and_relative_paths(tmp_path):
@@ -214,8 +214,8 @@ def test_rg_returns_structured_matches_context_and_relative_paths(tmp_path):
     except AssertionError as e: assert "mutually exclusive" in str(e)
     else: assert False
     addr = res[1].lnhash.split("|")
-    assert addr[0] == "2" and len(addr[1]) == 4 and addr[2:] == [""]
-    assert int(addr[1], 16) >= 0
+    assert addr[0] == "2" and len(addr[1]) == 2 and addr[2:] == [""]
+    assert all(c in 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_' for c in addr[1])
     expected = (f'SearchLine(kind="match", path="src/app.py", line_number=2, lnhash="{res[1].lnhash}", '
         'line="TODO here", matches=[(0, 4)])')
     assert repr(res[1]) == expected