Rewrite TreeDom __str__ and __repr__

Bump version - "the first initial release"

[run test]

Author: awolverp

Cargo.toml

@@ -17,7 +17,7 @@ crate-type = ["cdylib"]
 members = ["treedom", "matching"]
 
 [workspace.package]
-version = "0.1.0-beta1"
+version = "0.1.0"
 edition = "2021"
 readme = "README.md"
 license = "MIT"
@@ -28,7 +28,7 @@ authors = ["awolverp"]
 [profile.release]
 codegen-units = 1
 debug = false
-# incremental = false
+incremental = false
 lto = "fat"
 panic = "abort"
 strip = "symbols"

README.md

@@ -76,7 +76,7 @@ print(root.serialize())
 ```
 
 # TODO List
-- [ ] Rewrite TreeDom `__repr__` and `__str__`
+- [x] Rewrite TreeDom `__repr__` and `__str__`
 - [ ] Add benchmarks
 - [ ] Add memory usage report
 - [ ] Add PyPI version, downloads, test coverage, and python versions badges

python/markupever/_display.py

@@ -0,0 +1,51 @@
+import typing
+
+
+class _DisplayCharacterToken:
+    __slots__ = ("siblings", "children")
+
+    def __init__(self, siblings: bool):
+        self.siblings = siblings
+        self.children = False
+
+    def __str__(self):
+        # match (siblings, children)
+        # (true, true) => "│   ",
+        # (true, false) => "├── ",
+        # (false, true) => "    ",
+        # (false, false) => "└── ",
+
+        if self.children:
+            return "│   " if self.siblings else "    "
+
+        if self.siblings:
+            return "├── "
+
+        return "└── "
+
+
+class _Indentation:
+    __slots__ = ("tokens", "ignore_root")
+
+    def __init__(self, ignore_root: bool):
+        self.tokens: typing.List[_DisplayCharacterToken] = []
+        self.ignore_root = ignore_root
+
+    def indent(self, siblings: bool):
+        length = len(self.tokens)
+        if length > 0:
+            self.tokens[length - 1].children = True
+
+        self.tokens.append(_DisplayCharacterToken(siblings))
+        return self
+
+    def deindent(self):
+        self.tokens.pop()
+        return self
+
+    def __str__(self) -> str:
+        s = ""
+        for token in self.tokens[int(self.ignore_root) :]:
+            s += str(token)
+
+        return s

python/markupever/dom.py

@@ -60,10 +60,32 @@ def __len__(self) -> int:
         return len(self._raw)
 
     def __str__(self) -> str:
-        return str(self._raw)
+        from ._display import _Indentation
 
-    def __repr__(self) -> str:
-        return repr(self._raw)
+        res = ""
+
+        indent = _Indentation(True)
+
+        for edge in self.root().traverse():
+            if edge.closed:
+                if edge.node.has_children:
+                    indent.deindent()
+
+                continue
+
+            if edge.node.has_children:
+                indent.indent(edge.node.next_sibling is not None)
+                res += str(indent) + str(edge.node) + "\n"
+
+            else:
+                indent.indent(edge.node.next_sibling is not None)
+                res += str(indent) + str(edge.node) + "\n"
+                indent.deindent()
+
+        return res[:-1]  # remove the last '\n'
+
+    def __repr__(self):
+        return f"TreeDom(len={len(self)}, namespaces={self.namespaces()})"
 
 
 class _ConfigNode:
@@ -550,10 +572,11 @@ def __gt__(self, value):  # pragma: no cover
 
 class AttrsList:
     """
-    This type is only designed for communicating with element attributes. 
-    
+    This type is only designed for communicating with element attributes.
+
     Really it's a list, but has a behaviour between dictionary and list to provide you easy-to-use management.
     """
+
     __slots__ = ("__raw",)
 
     def __init__(self, attrs: _rustlib.AttrsList):
@@ -729,14 +752,10 @@ def __setitem__(
         self.__raw.update_item(index, val[0], val[1])
 
     @typing.overload
-    def __getitem__(
-        self, index: typing.Union[str, _rustlib.QualName]
-    ) -> str: ...
+    def __getitem__(self, index: typing.Union[str, _rustlib.QualName]) -> str: ...
 
     @typing.overload
-    def __getitem__(
-        self, index: int
-    ) -> typing.Tuple[_rustlib.QualName, str]: ...
+    def __getitem__(self, index: int) -> typing.Tuple[_rustlib.QualName, str]: ...
 
     def __getitem__(self, index):
         if not isinstance(index, int):
@@ -889,6 +908,7 @@ class ProcessingInstruction(BaseNode):
     a Node which embeds an instruction targeting a specific application but that can
     be ignored by any other applications which don't recognize the instruction.
     """
+
     _CONFIG = _ConfigNode(_rustlib.ProcessingInstruction, (Ordering.APPEND, Ordering.PREPEND))
 
     @property

python/tests/test_dom.py

@@ -34,7 +34,9 @@ def test_basenode_init():
     _test_rustlib_node_convert(_rustlib.Doctype, markupever.dom.Doctype, dom, "name", "", "")
     _test_rustlib_node_convert(_rustlib.Comment, markupever.dom.Comment, dom, "content")
     _test_rustlib_node_convert(_rustlib.Text, markupever.dom.Text, dom, "content")
-    _test_rustlib_node_convert(_rustlib.Element, markupever.dom.Element, dom, "name", [], False, False)
+    _test_rustlib_node_convert(
+        _rustlib.Element, markupever.dom.Element, dom, "name", [], False, False
+    )
     _test_rustlib_node_convert(
         _rustlib.ProcessingInstruction, markupever.dom.ProcessingInstruction, dom, "name", "data"
     )
@@ -55,7 +57,9 @@ def test_connect_node():
     dom = markupever.dom.TreeDom()
     root = dom.root()
 
-    html = root.create_element(markupever.dom.QualName("html", "html"), {"lang": "en"}, False, False)
+    html = root.create_element(
+        markupever.dom.QualName("html", "html"), {"lang": "en"}, False, False
+    )
     assert isinstance(html, markupever.dom.Element)
     assert html.parent == root
     assert html.name == "html"

python/tests/test_parser.py

@@ -56,7 +56,9 @@ def test_parser():  # this is a copy of test_rustlib.test_parser for markupever.
 
 
 def test_parse_function():
-    assert isinstance(markupever.parse("<html></html>", markupever.XmlOptions()), markupever.dom.TreeDom)
+    assert isinstance(
+        markupever.parse("<html></html>", markupever.XmlOptions()), markupever.dom.TreeDom
+    )
 
 
 def test_parse_file_function(tmp_path):
@@ -78,6 +80,8 @@ def test_parse_file_function(tmp_path):
         markupever.parse_file(str(file), markupever.HtmlOptions())
 
     file.write_bytes(b"<body></body>")
-    assert isinstance(markupever.parse_file(str(file), markupever.HtmlOptions()), markupever.dom.TreeDom)
+    assert isinstance(
+        markupever.parse_file(str(file), markupever.HtmlOptions()), markupever.dom.TreeDom
+    )
 
     markupever.parse_file(file, markupever.HtmlOptions())

src/nodes.rs

@@ -710,7 +710,12 @@ impl PyText {
     }
 }
 
-#[pyo3::pyclass(name = "AttrsListItems", module = "markupever._rustlib", mapping, frozen)]
+#[pyo3::pyclass(
+    name = "AttrsListItems",
+    module = "markupever._rustlib",
+    mapping,
+    frozen
+)]
 pub struct PyAttrsListItems {
     guard: NodeGuard,
     index: std::sync::atomic::AtomicUsize,

treedom/src/dom.rs

@@ -96,7 +96,7 @@ impl<'a> Serializer<'a> {
 
     fn serialize_iter<S>(
         &self,
-        iter: impl Iterator<Item=ego_tree::iter::Edge<'a, interface::Interface>>,
+        iter: impl Iterator<Item = ego_tree::iter::Edge<'a, interface::Interface>>,
         serializer: &mut S,
     ) -> std::io::Result<()>
     where
@@ -145,11 +145,9 @@ impl<'a> Serializer<'a> {
     }
 }
 
-fn skip_last<T>(mut iter: impl Iterator<Item=T>) -> impl Iterator<Item=T> {
+fn skip_last<T>(mut iter: impl Iterator<Item = T>) -> impl Iterator<Item = T> {
     let last = iter.next();
-    iter.scan(last, |state, item| {
-        std::mem::replace(state, Some(item))
-    })
+    iter.scan(last, |state, item| std::mem::replace(state, Some(item)))
 }
 
 impl markup5ever::serialize::Serialize for Serializer<'_> {