SiteGenerator/gsitegen/generate.py

from html5tagger import Document, E, HTML
import shutil
import markdown
import html
from pathlib import Path

#TODO re-think navigation bar. Think of way to add option to display navigation bar entry differently if you're currently on that page.
def generateNavigationBar(lines):
  global navbar
  navbar = E
  rawhtml = False
  htmlstring = ""
  for id, line in enumerate(lines):

    #parse raw HTML
    rawhtml, navbar, htmlstring = parseRawHTML(navbar, line, htmlstring, rawhtml, id, len(lines))

    #parse markdown
    if not rawhtml:
      if ";" in line:
        title, link = line.split(";", 1)

		#div class for styling
        navbar = navbar(HTML("<div class='" + title.replace(" ", "-").lower() + "'>"))

        #link and end of div
        navbar = navbar.li(HTML("<a href='" + link.strip() + "'>" + title + "</a></div>"))

def generateFooter(lines):
  global footer
  footer = E
  rawhtml = False
  htmlstring = ""
  for id, line in enumerate(lines):

    #parse raw HTML
    rawhtml, footer, htmlstring = parseRawHTML(footer, line, htmlstring, rawhtml, id, len(lines))

    #parse markdown
    if not rawhtml:
      footer = footer.li(HTML(parseMarkdown(footer, line)))

def parseRawHTML(doc, line, htmlstring, rawhtml, id, maxlines):

  #raw html start
  if line.strip() == ">":
    rawhtml = True
    htmlstring = ""

  #parse indented raw html
  elif rawhtml:
    if line.startswith("  "):

      #experimental markdown inside HTML support
      htmlstring = htmlstring + html.unescape(markdown.markdown(html.escape(line.strip())))

      #if indented html was the last line, this is needed for it to not be ignored
      #since this is the end of the file, we will not set rawhtml to False.
      if maxlines - id == 1:
        doc.div(HTML(htmlstring))
    else:
      rawhtml = False
      doc.div(HTML(htmlstring))
      htmlstring = ""

  return rawhtml, doc, htmlstring

def parseMarkdown(doc, line):

  #do not allow HTML in markdown
  line = html.escape(line.strip())
  return markdown.markdown(line)

def generateLines(title, lines):
  title = title.replace(".page", "")
  doc = Document(title, lang="en")
  rawhtml = False
  htmlstring = ""
  for id, line in enumerate(lines):

    #parse raw HTML
    rawhtml, doc, htmlstring = parseRawHTML(doc, line, htmlstring, rawhtml, id, len(lines))

    #parse markdown
    if not rawhtml:
      doc = doc(HTML(parseMarkdown(doc, line)))

  generatePage(title, doc)

def generatePage(title, doc):
  global pages
  global titles
  if 'pages' not in globals():
    pages = []
  if 'titles' not in globals():
    titles = []

  if 'navbar' in globals():
    if 'footer' in globals():
      pages.append(str(E.ul(navbar)) + str(doc) + str(E.ul(footer)))
    else:
      pages.append(str(E.ul(navbar)) + str(doc))
  else:
    if 'footer' in globals():
      pages.append(str(doc) + str(E.ul(footer)))
    else:
      pages.append(str(doc))
  titles.append(title)

def writePages(homepage):
  global pages
  global titles
    #TODO only delete files that aren't present in newest site generation
    #deleting contents of folder without deleting the folder, to increase compatibility with various systems
    #for root, dirs, files in os.walk('./website-output'):
    #  for f in files:
    #    os.unlink(os.path.join(root, f))
    #  for d in dirs:
    #    rmtree(os.path.join(root, d))

  for id, page in enumerate(pages):
    foldername = ""

    #creates ./website-output/pagetitle/index.html file if it is not homepage
    if titles[id] != homepage:
      foldername = titles[id].replace(" ", "-").lower()

    outputpath = Path(__file__).parent.joinpath("website-output")

    dirpath = outputpath.joinpath(foldername)

    dirpath.mkdir(parents=True, exist_ok=True)

    filepath = dirpath.joinpath("index.html")

    if filepath.exists():
      with filepath.open('r') as newpage:
        if newpage.read() == page:
          print("Page not changed: " + titles[id])
          continue

    with filepath.open('w') as newpage:
      newpage.write(page)
      print("Written changed page: " + titles[id])

  #resources
  respath = Path(__file__).parent.joinpath("resources")
  if respath.exists():

    #shutil.copytree copies timestamp too
    shutil.copytree(respath, outputpath, dirs_exist_ok=True)

    #check if folders are named correctly
    for folder in outputpath.iterdir():
      if folder.is_dir():

        #rename folder name format from "About Page" to "about-page"
        bettername = folder.name.replace(" ", "-").lower()

        if folder.name != bettername:
          newpath = outputpath.joinpath(bettername)

          #example: resources folder is "About", page got auto-created earlier as "about" (from About.page), let's copy "About" resources to "about" and delete "About"
          if newpath.exists():
            shutil.copytree(folder, newpath, dirs_exist_ok=True)
            shutil.rmtree(folder)
          else:
            shutil.move(str(folder), newpath)

def main():

  #if homepage is at Home.page, set homepage to "Home"
  homepage = "Home"

  navbarfile = Path(__file__).parent.joinpath('navbar')

  if navbarfile.exists():
    with navbarfile.open('r') as navbarfile:
      generateNavigationBar(navbarfile.readlines())
  else:
    print("No 'navbar' file found, there will be no navigation bar.")

  footerfile = Path(__file__).parent.joinpath('footer')

  if footerfile.exists():
    with footerfile.open('r') as footerfile:
      generateFooter(footerfile.readlines())
  else:
    print("No 'footer' file found, there will be no footer.")

  pagescount = 0

  for file in Path(__file__).parent.iterdir():
    if file.is_file():
      if file.suffix == ".page":
        pagescount += 1
        with file.open('r') as page:
          generateLines(file.stem, page.readlines())
  print("Found " + str(pagescount) + " pages")

  #write all pages to files
  writePages(homepage)

if __name__ == "__main__":
  main()