madduck's git repository

Every one of the projects in this repository is available at the canonical URL git://git.madduck.net/madduck/pub/<projectpath> — see each project's metadata for the exact URL.

All patches and comments are welcome. Please squash your changes to logical commits before using git-format-patch and git-send-email to patches@git.madduck.net. If you'd read over the Git project's submission guidelines and adhered to them, I'd be especially grateful.

SSH access, as well as push access can be individually arranged.

If you use my repositories frequently, consider adding the following snippet to ~/.gitconfig and using the third clone URL listed for each project:

[url "git://git.madduck.net/madduck/"]
  insteadOf = madduck:

Allow same RHS expressions in annotated assignments as in regular assignments (#1835)
[etc/vim.git] / docs / conf.py
index aeb57f99fcc63784e75db695d17393fef961659d..7381c9d6423f37a97148789e8c3c42cce3092f7f 100644 (file)
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.
 #
-import ast
 from pathlib import Path
 import re
-import shutil
 import string
+from typing import Callable, Dict, List, Optional, Pattern, Tuple, Set
+from dataclasses import dataclass
+import logging
 
-from recommonmark.parser import CommonMarkParser
+from pkg_resources import get_distribution
 
+logging.basicConfig(format="%(levelname)s: %(message)s", level=logging.INFO)
+
+LOG = logging.getLogger(__name__)
 
 CURRENT_DIR = Path(__file__).parent
+README = CURRENT_DIR / ".." / "README.md"
+REFERENCE_DIR = CURRENT_DIR / "reference"
+STATIC_DIR = CURRENT_DIR / "_static"
+
+
+@dataclass
+class SrcRange:
+    """Tracks which part of a file to get a section's content.
+
+    Data:
+        start_line: The line where the section starts (i.e. its sub-header) (inclusive).
+        end_line: The line where the section ends (usually next sub-header) (exclusive).
+    """
+
+    start_line: int
+    end_line: int
+
+
+@dataclass
+class DocSection:
+    """Tracks information about a section of documentation.
 
+    Data:
+        name: The section's name. This will used to detect duplicate sections.
+        src: The filepath to get its contents.
+        processors: The processors to run before writing the section to CURRENT_DIR.
+        out_filename: The filename to use when writing the section to CURRENT_DIR.
+        src_range: The line range of SRC to gets its contents.
+    """
 
-def get_version():
-    black_py = CURRENT_DIR / ".." / "black.py"
-    _version_re = re.compile(r"__version__\s+=\s+(?P<version>.*)")
-    with open(str(black_py), "r", encoding="utf8") as f:
-        version = _version_re.search(f.read()).group("version")
-    return str(ast.literal_eval(version))
+    name: str
+    src: Path
+    src_range: SrcRange = SrcRange(0, 1_000_000)
+    out_filename: str = ""
+    processors: Tuple[Callable, ...] = ()
 
+    def get_out_filename(self) -> str:
+        if not self.out_filename:
+            return self.name + ".md"
+        else:
+            return self.out_filename
 
-def make_pypi_svg(version):
-    template = CURRENT_DIR / "_static" / "pypi_template.svg"
-    target = CURRENT_DIR / "_static" / "pypi.svg"
+
+def make_pypi_svg(version: str) -> None:
+    template: Path = CURRENT_DIR / "_static" / "pypi_template.svg"
+    target: Path = CURRENT_DIR / "_static" / "pypi.svg"
     with open(str(template), "r", encoding="utf8") as f:
-        svg = string.Template(f.read()).substitute(version=version)
+        svg: str = string.Template(f.read()).substitute(version=version)
     with open(str(target), "w", encoding="utf8") as f:
         f.write(svg)
 
 
-def make_filename(line):
-    non_letters = re.compile(r"[^a-z]+")
-    filename = line[3:].rstrip().lower()
+def make_filename(line: str) -> str:
+    non_letters: Pattern = re.compile(r"[^a-z]+")
+    filename: str = line[3:].rstrip().lower()
     filename = non_letters.sub("_", filename)
     if filename.startswith("_"):
         filename = filename[1:]
@@ -52,71 +89,180 @@ def make_filename(line):
     return filename + ".md"
 
 
-def generate_sections_from_readme():
-    target_dir = CURRENT_DIR / "_build" / "generated"
-    readme = CURRENT_DIR / ".." / "README.md"
-    shutil.rmtree(str(target_dir), ignore_errors=True)
-    target_dir.mkdir(parents=True)
-
-    output = None
-    target_dir = target_dir.relative_to(CURRENT_DIR)
-    with open(str(readme), "r", encoding="utf8") as f:
-        for line in f:
+def get_contents(section: DocSection) -> str:
+    """Gets the contents for the DocSection."""
+    contents: List[str] = []
+    src: Path = section.src
+    start_line: int = section.src_range.start_line
+    end_line: int = section.src_range.end_line
+    with open(src, "r", encoding="utf-8") as f:
+        for lineno, line in enumerate(f, start=1):
+            if lineno >= start_line and lineno < end_line:
+                contents.append(line)
+    result = "".join(contents)
+    # Let's make Prettier happy with the amount of trailing newlines in the sections.
+    if result.endswith("\n\n"):
+        result = result[:-1]
+    if not result.endswith("\n"):
+        result = result + "\n"
+    return result
+
+
+def get_sections_from_readme() -> List[DocSection]:
+    """Gets the sections from README so they can be processed by process_sections.
+
+    It opens README and goes down line by line looking for sub-header lines which
+    denotes a section. Once it finds a sub-header line, it will create a DocSection
+    object with all of the information currently available. Then on every line, it will
+    track the ending line index of the section. And it repeats this for every sub-header
+    line it finds.
+    """
+    sections: List[DocSection] = []
+    section: Optional[DocSection] = None
+    with open(README, "r", encoding="utf-8") as f:
+        for lineno, line in enumerate(f, start=1):
             if line.startswith("## "):
-                if output is not None:
-                    output.close()
                 filename = make_filename(line)
-                output_path = CURRENT_DIR / filename
-                if output_path.is_symlink() or output_path.is_file():
-                    output_path.unlink()
-                output_path.symlink_to(target_dir / filename)
-                output = open(str(output_path), "w", encoding="utf8")
-                output.write(
-                    "[//]: # (NOTE: THIS FILE IS AUTOGENERATED FROM README.md)\n\n"
+                section_name = filename[:-3]
+                section = DocSection(
+                    name=str(section_name),
+                    src=README,
+                    src_range=SrcRange(lineno, lineno),
+                    out_filename=filename,
+                    processors=(fix_headers,),
                 )
-
-            if output is None:
-                continue
-
-            if line.startswith("##"):
-                line = line[1:]
-
-            output.write(line)
+                sections.append(section)
+            if section is not None:
+                section.src_range.end_line += 1
+    return sections
+
+
+def fix_headers(contents: str) -> str:
+    """Fixes the headers of sections copied from README.
+
+    Removes one octothorpe (#) from all headers since the contents are no longer nested
+    in a root document (i.e. the README).
+    """
+    lines: List[str] = contents.splitlines()
+    fixed_contents: List[str] = []
+    for line in lines:
+        if line.startswith("##"):
+            line = line[1:]
+        fixed_contents.append(line + "\n")  # splitlines strips the leading newlines
+    return "".join(fixed_contents)
+
+
+def process_sections(
+    custom_sections: List[DocSection], readme_sections: List[DocSection]
+) -> None:
+    """Reads, processes, and writes sections to CURRENT_DIR.
+
+    For each section, the contents will be fetched, processed by processors
+    required by the section, and written to CURRENT_DIR. If it encounters duplicate
+    sections (i.e. shares the same name attribute), it will skip processing the
+    duplicates.
+
+    It processes custom sections before the README generated sections so sections in the
+    README can be overwritten with custom options.
+    """
+    processed_sections: Dict[str, DocSection] = {}
+    modified_files: Set[Path] = set()
+    sections: List[DocSection] = custom_sections
+    sections.extend(readme_sections)
+    for section in sections:
+        if section.name in processed_sections:
+            LOG.warning(
+                f"Skipping '{section.name}' from '{section.src}' as it is a duplicate"
+                f" of a custom section from '{processed_sections[section.name].src}'"
+            )
+            continue
+
+        LOG.info(f"Processing '{section.name}' from '{section.src}'")
+        target_path: Path = CURRENT_DIR / section.get_out_filename()
+        if target_path in modified_files:
+            LOG.warning(
+                f"{target_path} has been already written to, its contents will be"
+                " OVERWRITTEN and notices will be duplicated"
+            )
+        contents: str = get_contents(section)
+
+        # processors goes here
+        if fix_headers in section.processors:
+            contents = fix_headers(contents)
+
+        with open(target_path, "w", encoding="utf-8") as f:
+            if section.src.suffix == ".md" and section.src != target_path:
+                rel = section.src.resolve().relative_to(CURRENT_DIR.parent)
+                f.write(f'[//]: # "NOTE: THIS FILE WAS AUTOGENERATED FROM {rel}"\n\n')
+            f.write(contents)
+        processed_sections[section.name] = section
+        modified_files.add(target_path)
 
 
 # -- Project information -----------------------------------------------------
 
 project = "Black"
-copyright = "2018, Łukasz Langa and contributors to Black"
+copyright = "2020, Łukasz Langa and contributors to Black"
 author = "Łukasz Langa and contributors to Black"
 
 # Autopopulate version
-# The full version, including alpha/beta/rc tags.
-release = get_version()
+# The version, including alpha/beta/rc tags, but not commit hash and datestamps
+release = get_distribution("black").version.split("+")[0]
 # The short X.Y version.
 version = release
 for sp in "abcfr":
     version = version.split(sp)[0]
+
+custom_sections = [
+    DocSection("the_black_code_style", CURRENT_DIR / "the_black_code_style.md"),
+    DocSection("editor_integration", CURRENT_DIR / "editor_integration.md"),
+    DocSection("blackd", CURRENT_DIR / "blackd.md"),
+    DocSection("black_primer", CURRENT_DIR / "black_primer.md"),
+    DocSection("contributing_to_black", CURRENT_DIR / ".." / "CONTRIBUTING.md"),
+    DocSection("change_log", CURRENT_DIR / ".." / "CHANGES.md"),
+]
+
+# Sphinx complains when there is a source file that isn't referenced in any of the docs.
+# Since some sections autogenerated from the README are unused warnings will appear.
+#
+# Sections must be listed to what their name is when passed through make_filename().
+blocklisted_sections_from_readme = {
+    "license",
+    "pragmatism",
+    "testimonials",
+    "used_by",
+}
+
 make_pypi_svg(release)
-generate_sections_from_readme()
+readme_sections = get_sections_from_readme()
+readme_sections = [
+    x for x in readme_sections if x.name not in blocklisted_sections_from_readme
+]
+
+process_sections(custom_sections, readme_sections)
 
 
 # -- General configuration ---------------------------------------------------
 
 # If your documentation needs a minimal Sphinx version, state it here.
-#
-# needs_sphinx = '1.0'
+needs_sphinx = "3.0"
 
 # Add any Sphinx extension module names here, as strings. They can be
 # extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
 # ones.
-extensions = ["sphinx.ext.autodoc", "sphinx.ext.intersphinx", "sphinx.ext.napoleon"]
+extensions = [
+    "sphinx.ext.autodoc",
+    "sphinx.ext.intersphinx",
+    "sphinx.ext.napoleon",
+    "recommonmark",
+]
+
+# If you need extensions of a certain version or higher, list them here.
+needs_extensions = {"recommonmark": "0.5"}
 
 # Add any paths that contain templates here, relative to this directory.
 templates_path = ["_templates"]
 
-source_parsers = {".md": CommonMarkParser}
-
 # The suffix(es) of source filenames.
 # You can specify multiple suffix as a list of string:
 source_suffix = [".rst", ".md"]
@@ -134,6 +280,7 @@ language = None
 # List of patterns, relative to source directory, that match files and
 # directories to ignore when looking for source files.
 # This pattern also affects html_static_path and html_extra_path .
+
 exclude_patterns = ["_build", "Thumbs.db", ".DS_Store"]
 
 # The name of the Pygments (syntax highlighting) style to use.
@@ -161,12 +308,13 @@ html_theme_options = {
     "show_related": False,
     "description": "“Any color you like.”",
     "github_button": True,
-    "github_user": "ambv",
+    "github_user": "psf",
     "github_repo": "black",
     "github_type": "star",
     "show_powered_by": True,
     "fixed_sidebar": True,
     "logo": "logo2.png",
+    "travis_button": True,
 }