Refactor in modules, add gemtext parser

6 files changed, 402 insertions, 0 deletions

diff --git a/blog_experiment/blog/__init__.py b/blog_experiment/blog/__init__.py
new file mode 100644
index 00000000..4b1e0ba5
--- /dev/null
+++ b/blog_experiment/blog/__init__.py
@@ -0,0 +1,16 @@
+import pathlib
+import re
+
+import bicephalus
+
+from blog import blog_pages, page
+
+
+def handler(request: bicephalus.Request) -> bicephalus.Response:
+    if request.path == "/":
+        return blog_pages.Root(request).response()
+    if re.match(r"/\d{4}/\d{2}/.*/", request.path):
+        blog_file = pathlib.Path("content") / (request.path[1:-1] + ".gmi")
+        if blog_file.exists():
+            return blog_pages.EntryPage(request, blog_file).response()
+    return page.NotFound(request).response()
diff --git a/blog_experiment/blog/__main__.py b/blog_experiment/blog/__main__.py
new file mode 100644
index 00000000..b936500d
--- /dev/null
+++ b/blog_experiment/blog/__main__.py
@@ -0,0 +1,17 @@
+import logging
+
+from bicephalus import main as bicephalus_main
+from bicephalus import otel
+from bicephalus import ssl
+
+import blog
+
+
+def main():
+    otel.configure_logging(logging.INFO)
+    with ssl.temporary_ssl_context("localhost") as ssl_context:
+        bicephalus_main.main(blog.handler, ssl_context, 8000)
+
+
+if __name__ == "__main__":
+    main()
diff --git a/blog_experiment/blog/blog_pages.py b/blog_experiment/blog/blog_pages.py
new file mode 100644
index 00000000..7808c16d
--- /dev/null
+++ b/blog_experiment/blog/blog_pages.py
@@ -0,0 +1,80 @@
+import datetime
+import itertools
+import pathlib
+import textwrap
+
+import bicephalus
+
+import htmlgenerator as h
+
+from blog import html, page
+
+
+class Entry:
+    def __init__(self, path: pathlib.Path):
+        assert path.is_relative_to(pathlib.Path("content")), f"bad path {path}"
+        self.path = path
+        self.content = path.read_text()
+
+    @property
+    def title(self):
+        return self.content.splitlines()[0][2:]
+
+    @property
+    def posted(self):
+        return datetime.date.fromisoformat(self.content.splitlines()[1])
+
+    @property
+    def uri(self):
+        return f"/{self.path.parts[1]}/{self.path.parts[2]}/{self.path.stem}/"
+
+
+class Root(page.BasePage):
+    def entries(self):
+        entries = map(Entry, pathlib.Path("content").glob("*/*/*.gmi"))
+        return sorted(entries, key=lambda e: e.posted, reverse=True)
+
+    def get_gemini_content(self):
+        posts = "\n".join([f"=> {e.uri} {e.posted} {e.title}" for e in self.entries()])
+        content = (
+            textwrap.dedent(
+                """\
+                # El blog es mío
+
+                ## Hay otros como él, pero este es el mío
+
+                ____
+                """
+            )
+            + posts
+        )
+        return bicephalus.Status.OK, "text/gemini", content
+
+    def get_http_content(self):
+        posts = [
+            (h.H3(h.A(f"{e.title} ({e.posted})", href=e.uri))) for e in self.entries()
+        ]
+        return (
+            bicephalus.Status.OK,
+            "text/html",
+            html.html_template(*itertools.chain(posts)),
+        )
+
+
+class EntryPage(page.BasePage):
+    def __init__(self, request, path):
+        super().__init__(request)
+        self.path = path
+        self.entry = Entry(path)
+
+    def get_gemini_content(self):
+        return bicephalus.Status.OK, "text/gemini", self.entry.content
+
+    def get_http_content(self):
+        return (
+            bicephalus.Status.OK,
+            "text/html",
+            html.html_template(
+                h.PRE(self.entry.content),
+            ),
+        )
diff --git a/blog_experiment/blog/gemtext.py b/blog_experiment/blog/gemtext.py
new file mode 100644
index 00000000..66298e3f
--- /dev/null
+++ b/blog_experiment/blog/gemtext.py
@@ -0,0 +1,223 @@
+import dataclasses
+import re
+import typing
+
+
+def parse(s):
+    """
+    >>> parse('''# Header 1
+    ... 
+    ... ## Header 2
+    ... 
+    ... ### Header 3
+    ... 
+    ... * List 1
+    ... * List 2
+    ... 
+    ... > First line quote.
+    ... > Second line of quote.
+    ... 
+    ... ```
+    ... Fenced
+    ... Lines
+    ... ```
+    ... 
+    ... Paragraph.
+    ... 
+    ... Another paragraph.
+    ... ''')
+    [Header(level=1, text='Header 1'),
+     Line(text=''),
+     Header(level=2, text='Header 2'),
+     Line(text=''),
+     Header(level=3, text='Header 3'),
+     Line(text=''),
+     List(items=[ListItem(text='List 1'),
+                 ListItem(text='List 2')]),
+     Line(text=''),
+     BlockQuote(lines=[BlockQuoteLine(text='First line quote.'),
+                       BlockQuoteLine(text='Second line of quote.')]),
+     Line(text=''),
+     Pre(content='Fenced\\nLines\\n'),
+     Line(text=''),
+     Line(text='Paragraph.'),
+     Line(text=''),
+     Line(text='Another paragraph.')]
+    """
+
+    lines = s.splitlines()
+
+    i = 0
+    gem = []
+
+    while i < len(lines):
+        line = parse_line(lines[i])
+
+        if isinstance(line, Link):
+            gem.append(line)
+            i = i + 1
+            continue
+
+        if isinstance(line, Header):
+            gem.append(line)
+            i = i + 1
+            continue
+
+        if isinstance(line, ListItem):
+            items = []
+            while i < len(lines) and isinstance(parse_line(lines[i]), ListItem):
+                items.append(parse_line(lines[i]))
+                i = i + 1
+            gem.append(List(items))
+            continue
+
+        if isinstance(line, BlockQuoteLine):
+            quotes = []
+            while i < len(lines) and isinstance(parse_line(lines[i]), BlockQuoteLine):
+                quotes.append(parse_line(lines[i]))
+                i = i + 1
+            gem.append(BlockQuote(quotes))
+            continue
+
+        if isinstance(line, PreFence):
+            content = ""
+            i = i + 1
+            while i < len(lines) and not isinstance(parse_line(lines[i]), PreFence):
+                content += lines[i]
+                content += "\n"
+                i = i + 1
+            gem.append(Pre(content))
+            i = i + 1
+            continue
+
+        gem.append(line)
+        i = i + 1
+
+    return gem
+
+
+def parse_line(l):
+    if Link.is_link(l):
+        return Link(l)
+    if Header.is_header(l):
+        return Header(l)
+    if ListItem.is_list_item(l):
+        return ListItem(l)
+    if BlockQuoteLine.is_block_quote_line(l):
+        return BlockQuoteLine(l)
+    if PreFence.is_pre_fence(l):
+        return PreFence()
+    return Line(l)
+
+
+@dataclasses.dataclass
+class Link:
+    """
+    >>> Link("=> http://example.com")
+    Link(url='http://example.com', text=None)
+
+    >>> Link("=> http://example.com Example text")
+    Link(url='http://example.com', text='Example text')
+    """
+
+    url: str
+    text: typing.Optional[str]
+
+    def __init__(self, line: str):
+        assert Link.is_link(line)
+        parts = line.split(None, 2)
+        self.url = parts[1]
+        self.text = parts[2] if len(parts) > 2 else None
+
+    @staticmethod
+    def is_link(line: str):
+        return line.startswith("=>")
+
+@dataclasses.dataclass
+class Header:
+    """
+    >>> Header("# Level one")
+    Header(level=1, text='Level one')
+
+    >>> Header("## Level two")
+    Header(level=2, text='Level two')
+
+    >>> Header("### Level three")
+    Header(level=3, text='Level three')
+    """
+
+    level: int
+    text: str
+
+    def __init__(self, line: str):
+        assert Header.is_header(line)
+        hashes, self.text = line.split(None, 1)
+        self.level = len(hashes)
+
+    @staticmethod
+    def is_header(line: str):
+        return re.match("#{1,3} .*", line)
+
+@dataclasses.dataclass
+class ListItem:
+    """
+    >>> ListItem("* foo")
+    ListItem(text='foo')
+    """
+
+    text: str
+
+    def __init__(self, line: str):
+        assert ListItem.is_list_item(line)
+        self.text = line[2:]
+
+    @staticmethod
+    def is_list_item(line: str):
+        return line.startswith("* ")
+
+
+@dataclasses.dataclass
+class BlockQuoteLine:
+    """
+    >>> BlockQuoteLine("> foo")
+    BlockQuoteLine(text='foo')
+
+    >>> BlockQuoteLine(">foo")
+    BlockQuoteLine(text='foo')
+    """
+
+    text: str
+
+    def __init__(self, line: str):
+        assert BlockQuoteLine.is_block_quote_line(line)
+        self.text = line[2:] if line.startswith("> ") else line[1:]
+
+    @staticmethod
+    def is_block_quote_line(line: str):
+        return line.startswith(">")
+
+
+class PreFence:
+    @staticmethod
+    def is_pre_fence(line: str):
+        return line == "```"
+
+
+@dataclasses.dataclass
+class Line:
+    text: str
+
+
+@dataclasses.dataclass
+class List:
+    items: typing.List[ListItem]
+
+
+@dataclasses.dataclass
+class BlockQuote:
+    lines: typing.List[BlockQuoteLine]
+
+
+@dataclasses.dataclass
+class Pre:
+    content: str
diff --git a/blog_experiment/blog/html.py b/blog_experiment/blog/html.py
new file mode 100644
index 00000000..7293d395
--- /dev/null
+++ b/blog_experiment/blog/html.py
@@ -0,0 +1,29 @@
+import subprocess
+
+import htmlgenerator as h
+
+
+def tidy(s):
+    p = subprocess.run(
+        ["tidy", "--indent", "yes", "-q", "-wrap", "160"],
+        input=s,
+        stdout=subprocess.PIPE,
+        encoding="UTF8",
+    )
+    return p.stdout
+
+
+def html_template(*content):
+    return tidy(
+        h.render(
+            h.HTML(
+                h.HEAD(h.TITLE("El blog es mío")),
+                h.BODY(
+                    h.H1("El blog es mío"),
+                    h.H2("Hay otros como él, pero este es el mío"),
+                    *content,
+                ),
+            ),
+            {},
+        )
+    )
diff --git a/blog_experiment/blog/page.py b/blog_experiment/blog/page.py
new file mode 100644
index 00000000..fcc4841a
--- /dev/null
+++ b/blog_experiment/blog/page.py
@@ -0,0 +1,37 @@
+import bicephalus
+
+
+class BasePage:
+    def __init__(self, request):
+        self.request = request
+
+    def response(self):
+        if self.request.proto == bicephalus.Proto.GEMINI:
+            status, content_type, content = self.get_gemini_content()
+        elif self.request.proto == bicephalus.Proto.HTTP:
+            status, content_type, content = self.get_http_content()
+        else:
+            assert False, f"unknown protocol {self.request.proto}"
+
+        return bicephalus.Response(
+            content=content.encode("utf8"),
+            content_type=content_type,
+            status=bicephalus.Status.OK,
+        )
+
+
+class NotFound(BasePage):
+    def get_gemini_content(self):
+        # TODO: does not work!
+        return (
+            bicephalus.Status.NOT_FOUND,
+            "text/gemini",
+            f"{self.request.path} not found",
+        )
+
+    def get_http_content(self):
+        return (
+            bicephalus.Status.NOT_FOUND,
+            "text/html",
+            f"{self.request.path} not found",
+        )