Repository with sources and generator of https://larlet.fr/david/ https://larlet.fr/david/
Nelze vybrat více než 25 témat Téma musí začínat písmenem nebo číslem, může obsahovat pomlčky („-“) a může být dlouhé až 35 znaků.

site.py 6.8KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206
  1. #!/usr/bin/env python3
  2. import fnmatch
  3. import json
  4. import locale
  5. import os
  6. from dataclasses import dataclass
  7. from datetime import date, datetime, timedelta
  8. from html import escape
  9. from pathlib import Path
  10. from time import perf_counter
  11. import mistune
  12. from jinja2 import Environment as Env
  13. from jinja2 import FileSystemLoader
  14. from minicli import cli, run, wrap
  15. from mistune.directives import DirectiveInclude
  16. from slugify import slugify
  17. # Useful for dates rendering within Jinja2.
  18. locale.setlocale(locale.LC_ALL, "fr_FR.UTF-8")
  19. HERE = Path(".")
  20. DAVID = HERE / "david"
  21. DOMAIN = "https://larlet.fr"
  22. LOCAL_DOMAIN = "http://larlet.test:3579"
  23. # Hardcoding publication at 12 in Paris timezone.
  24. NORMALIZED_STRFTIME = "%Y-%m-%dT12:00:00+01:00"
  25. TODAY = datetime.today() + timedelta(hours=6)
  26. class H2AnchorsHTMLRenderer(mistune.HTMLRenderer):
  27. def heading(self, text, level):
  28. # Set an anchor to h2 headings.
  29. if level == 2:
  30. slug = slugify(text)
  31. return (
  32. f'<h2 id="{slug}">'
  33. f"{text} "
  34. f'<a href="#{slug}" title="Ancre vers cette partie" '
  35. f'aria-hidden="true">#</a>'
  36. f"</h2>"
  37. )
  38. else:
  39. return super().heading(text, level)
  40. # We want a custom renderer to create a hash/link for each H2 headings.
  41. markdown_with_h2_anchors = mistune.create_markdown(
  42. renderer=H2AnchorsHTMLRenderer(escape=False), plugins=[DirectiveInclude()]
  43. )
  44. # The second markdown is pertinent to generate articles for the feed,
  45. # we do not need anchors in that case.
  46. markdown = mistune.create_markdown(
  47. renderer=mistune.HTMLRenderer(escape=False), plugins=[DirectiveInclude()]
  48. )
  49. # This is the jinja2 configuration to locate templates.
  50. environment = Env(loader=FileSystemLoader(str(DAVID / "templates")))
  51. def neighborhood(iterable, first=None, last=None):
  52. """
  53. Yield the (previous, current, next) items given an iterable.
  54. You can specify a `first` and/or `last` item for bounds.
  55. """
  56. iterator = iter(iterable)
  57. previous = first
  58. current = next(iterator) # Throws StopIteration if empty.
  59. for next_ in iterator:
  60. yield (previous, current, next_)
  61. previous = current
  62. current = next_
  63. yield (previous, current, last)
  64. def each_markdown_from(source_dir, file_name="*.md"):
  65. """Walk across the `source_dir` and return the md file paths."""
  66. for filename in fnmatch.filter(os.listdir(source_dir), file_name):
  67. yield os.path.join(source_dir, filename)
  68. @dataclass
  69. class Page:
  70. title: str
  71. content: str
  72. file_path: str
  73. lang: str = "fr"
  74. def __post_init__(self):
  75. suffix = len(".md")
  76. prefix = len("YYYY/MM-DD") + suffix
  77. date_str = self.file_path[-prefix:-suffix].replace("-", "/")
  78. self.url = f"/david/{date_str}/"
  79. self.date = datetime.strptime(date_str, "%Y/%m/%d").date()
  80. self.full_url = f"{DOMAIN}{self.url}"
  81. self.normalized_date = self.date.strftime(NORMALIZED_STRFTIME)
  82. self.escaped_title = escape(self.title)
  83. self.escaped_content = escape(
  84. self.content.replace('href="/', f'href="{DOMAIN}/')
  85. .replace('src="/', f'src="{DOMAIN}/')
  86. .replace('href="#', f'href="{self.full_url}#')
  87. )
  88. # Extract first paragraph.
  89. self.extract = self.content.split("</p>", 1)[0] + "</p>"
  90. def __lt__(self, other: "Page"):
  91. if not isinstance(other, Page):
  92. return NotImplemented
  93. return self.date < other.date
  94. @staticmethod
  95. def all(source: Path, only_published=True, with_h2_anchors=True):
  96. """Retrieve all pages sorted by desc."""
  97. page_list = []
  98. md = markdown_with_h2_anchors if with_h2_anchors else markdown
  99. for file_path in each_markdown_from(source):
  100. result = md.read(file_path)
  101. # Extract (and remove) the title from the generated page.
  102. title, content = result.split("</h1>", 1)
  103. h1_opening_size = len("<h1>")
  104. title = title[h1_opening_size:]
  105. page = Page(title, content, file_path)
  106. if only_published and page.is_draft:
  107. continue
  108. page_list.append(page)
  109. return sorted(page_list, reverse=True)
  110. @property
  111. def is_draft(self):
  112. return (
  113. datetime(year=self.date.year, month=self.date.month, day=self.date.day)
  114. > TODAY
  115. )
  116. @cli
  117. def orphans():
  118. """Print out fragments not linked to any page."""
  119. linked_fragments = []
  120. for file_path in each_markdown_from(DAVID / "2020"):
  121. for line in open(file_path).readlines():
  122. if line.startswith(".. include:: fragments/"):
  123. linked_fragments.append(line[len(".. include:: fragments/") : -1])
  124. all_fragments = []
  125. for file_path in each_markdown_from(DAVID / "2020" / "fragments"):
  126. all_fragments.append(file_path[len("david/2020/fragments/") :])
  127. for fragment_filename in set(all_fragments) - set(linked_fragments):
  128. # Prepending path for easy command+click from fish.
  129. print(f"Orphan: {DAVID / '2020' / 'fragments' / fragment_filename}")
  130. @cli
  131. def pages():
  132. """Build the agregations from fragments."""
  133. root_path = DAVID / "2020"
  134. for previous, page, next_ in neighborhood(
  135. reversed(Page.all(source=root_path, only_published=False)),
  136. first={
  137. "url": "/david/stream/",
  138. "title": "Streams 2009-2019",
  139. "is_draft": False,
  140. },
  141. ):
  142. template = environment.get_template("article_2020.html")
  143. content = template.render(page=page, prev=previous, next=next_,)
  144. target_path = Path(page.url[1:])
  145. target_path.mkdir(parents=True, exist_ok=True)
  146. open(target_path / "index.html", "w").write(content)
  147. if page.is_draft:
  148. print(f"Draft: {LOCAL_DOMAIN}{page.url} ({page.title})")
  149. template = environment.get_template("archives_2020.html")
  150. content = template.render(page_list=Page.all(source=root_path))
  151. open(root_path / "index.html", "w").write(content)
  152. @cli
  153. def home():
  154. """Build the home page with last published items."""
  155. template = environment.get_template("profil.html")
  156. content = template.render(page_list=Page.all(source=DAVID / "2020"),)
  157. open(DAVID / "index.html", "w").write(content)
  158. @cli
  159. def feed():
  160. """Generate a feed from last published items."""
  161. template = environment.get_template("feed.xml")
  162. content = template.render(
  163. page_list=Page.all(source=DAVID / "2020", with_h2_anchors=False),
  164. current_dt=TODAY.strftime(NORMALIZED_STRFTIME),
  165. BASE_URL=f"{DOMAIN}/david/",
  166. )
  167. open(DAVID / "log" / "index.xml", "w").write(content)
  168. @wrap
  169. def perf_wrapper():
  170. start = perf_counter()
  171. yield
  172. elapsed = perf_counter() - start
  173. print(f"Done in {elapsed:.5f} seconds.")
  174. if __name__ == "__main__":
  175. run()