""" ================================= | | This file is part of | | staticoso | | Just a simple Static Site Generator | | | | Licensed under the AGPLv3 license | | Copyright (C) 2022, OctoSpacc | | ================================= """ from datetime import datetime from multiprocessing import Pool, cpu_count from Libs.bs4 import BeautifulSoup from Modules.Config import * from Modules.Elements import * from Modules.HTML import * from Modules.Logging import * from Modules.Markdown import * from Modules.Pug import * from Modules.Utils import * # Generate HTML tree/nested list from our internal metaformat, such as: # :Item 1 \\
  • Item 1
  • # :Item 4 //
  • Item 4
  • def GenHTMLTreeList(MetaList:str, Type:str='ul'): HTML = '' Lines = MetaList.splitlines() CurDepth, NextDepth, PrevDepth = 0, 0, 0 for i,e in enumerate(Lines): CurDepth = e.find(':') NextDepth = Lines[i+1].find(':') if i+1 < len(Lines) else 0 HTML += '\n
  • ' + e[CurDepth+1:] if NextDepth == CurDepth: HTML += '
  • ' elif NextDepth > CurDepth: HTML += f'\n<{Type}>' * (NextDepth - CurDepth) elif NextDepth < CurDepth: HTML += f'\n' * (CurDepth - NextDepth) + '' PrevDepth = CurDepth return f'<{Type}>{HTML}\n' # Menu styles: # - Simple: Default, Flat, Line # - Others: Excerpt, Image, Preview (Excerpt + Image), Full def GetHTMLPagesList(Pages, BlogName, SiteRoot, PathPrefix, CallbackFile=None, Unite=[], Type=None, Limit=None, PathFilter='', Category=None, For='Menu', MarkdownExts=(), MenuStyle='Default', ShowPaths=True): Flatten, SingleLine, DoneCount, PrevDepth = False, False, 0, 0 if MenuStyle == 'Flat': Flatten = True elif MenuStyle == 'Line': ShowPaths, SingleLine = False, True List, ToPop, LastParent = '', [], [] IndexPages = Pages.copy() for e in IndexPages: if e[3]['Index'] == 'False' or e[3]['Index'] == 'None': IndexPages.remove(e) for i,e in enumerate(IndexPages): if Type and e[3]['Type'] != Type: ToPop += [i] ToPop = RevSort(ToPop) for i in ToPop: IndexPages.pop(i) if Type == 'Page': IndexPages = OrderPages(IndexPages) for i,e in enumerate(Unite): if e: IndexPages.insert(i,[e,None,None,{'Type':Type,'Index':'True','Order':'Unite'}]) for File, Content, Titles, Meta in IndexPages: # Allow for the virtual "Pages/" prefix to be used in path filtering TmpPathFilter = PathFilter if TmpPathFilter.startswith('Pages/'): TmpPathFilter = TmpPathFilter[len('Pages/'):] if File.startswith('Posts/'): continue if (not Type or (Meta['Type'] == Type and CanIndex(Meta['Index'], For))) and (not Category or Category in Meta['Categories']) and File.startswith(TmpPathFilter) and File != CallbackFile and (not Limit or Limit > DoneCount): Depth = (File.count('/') + 1) if Meta['Order'] != 'Unite' else 1 # Folder names are handled here if Depth > 1 and Meta['Order'] != 'Unite': CurParent = File.split('/')[:-1] for i,s in enumerate(CurParent): if LastParent != CurParent and ShowPaths: LastParent = CurParent Levels = '.' * ((Depth-2+i) if not Flatten else 0) + ':' # If search node endswith index, it's a page; else, it's a folder if StripExt(File).endswith('index'): Title = MakeListTitle(File, Meta, Titles, 'HTMLTitle', SiteRoot, BlogName, PathPrefix) DoneCount += 1 else: Title = CurParent[Depth-2+i] if SingleLine: List += ' ' + Title + ' ' else: List += Levels + Title + '\n' # Pages with any other path if not (Depth > 1 and StripExt(File).split('/')[-1] == 'index'): Levels = '.' * ((Depth-1) if not Flatten else 0) + ':' DoneCount += 1 if Meta['Order'] == 'Unite': Title = markdown(MarkdownHTMLEscape(File, MarkdownExts), extensions=MarkdownExts).removeprefix('

    ').removesuffix('

    ') else: Title = MakeListTitle(File, Meta, Titles, 'HTMLTitle', SiteRoot, BlogName, PathPrefix) if SingleLine: List += ' ' + Title + ' ' else: List += Levels + Title + '\n' if MenuStyle in ('Default', 'Flat'): return GenHTMLTreeList(List) elif MenuStyle in ('Line', 'Excerpt', 'Image', 'Preview', 'Full'): return List def CheckHTMLCommentLine(Line): if Line.startswith(''): return Line return None def TemplatePreprocessor(Text): Meta, MetaDefault = '', { 'MenuStyle': 'Default'} for l in Text.splitlines(): ll = l.lstrip().rstrip() lll = CheckHTMLCommentLine(ll) if lll: if lll.startswith('%'): Meta += lll[1:-3].lstrip().rstrip() + '\n' Meta = dict(ReadConf(LoadConfStr('[Meta]\n' + Meta), 'Meta')) for i in MetaDefault: if not i in Meta: Meta.update({i:MetaDefault[i]}) return Meta def PagePreprocessor(Path, TempPath, Type, SiteTemplate, SiteRoot, GlobalMacros, CategoryUncategorized, LightRun=False): File = ReadFile(Path) Path = Path.lower() Content, Titles, DashyTitles, HTMLTitlesFound, Macros, Meta, MetaDefault = '', [], [], False, '', '', { 'Template': SiteTemplate, 'Style': '', 'Type': Type, 'Index': 'Unspecified', 'Feed': 'True', 'Title': '', 'HTMLTitle': '', 'Description': '', 'Image': '', 'Macros': {}, 'Categories': [], 'URLs': [], 'CreatedOn': '', 'UpdatedOn': '', 'EditedOn': '', 'Order': None} # Find all positions of '', add them in a list=[[pos0,pos1,line0,line1],...] for l in File.splitlines(): ll = l.lstrip().rstrip() lll = CheckHTMLCommentLine(ll) if ll.startswith('//') or lll: # Find preprocessor lines lll = ll[2:].lstrip() if lll.startswith('%'): Meta += lll[1:].lstrip() + '\n' elif lll.startswith('$'): Macros += lll[1:].lstrip() + '\n' #if ll.startswith(''): # Find comment and code blocks # IgnoreBlocksStart += [l] else: # Find headings #if line in ignore block: # continue Headings = ('h1', 'h2', 'h3', 'h4', 'h5', 'h6') if Path.endswith(FileExtensions['HTML']) and not HTMLTitlesFound: Soup = BeautifulSoup(File, 'html.parser') Tags = Soup.find_all() for t in Tags: if t.name in Headings: Title = '#'*int(t.name[1]) + ' ' + str(t.text) DashTitle = DashifyTitle(Title.lstrip('#'), DashyTitles) DashyTitles += [DashTitle] Titles += [Title] t.replace_with(MakeLinkableTitle(None, Title, DashTitle, 'md')) Content = str(Soup.prettify(formatter=None)) HTMLTitlesFound = True elif Path.endswith(FileExtensions['Markdown']): lsuffix = '' if ll.startswith(('-', '+', '*')): lsuffix += ll[0] ll = ll[1:].lstrip() if ll.startswith('#') or (ll.startswith('<') and ll[1:].startswith(Headings)): if ll.startswith('#'): Title = ll elif ll.startswith('<'): if ll[3:].startswith((" class='NoTitle", ' class="NoTitle')): Content += l + '\n' continue else: Title = '#'*h + str(ll[3:]) DashTitle = DashifyTitle(MkSoup(Title.lstrip('#')).get_text(), DashyTitles) DashyTitles += [DashTitle] Titles += [Title] Title = MakeLinkableTitle(None, Title, DashTitle, 'md') Title = Title.replace('> Max: Max = int(Curr) else: NoOrder += [e] New = [None] * (Max+1) for i,e in enumerate(Old): Curr = e[3]['Order'] if Curr: New[int(Curr)] = e while None in New: New.remove(None) return New + NoOrder def CanIndex(Index, For): if Index in ('False', 'None'): return False elif Index in ('True', 'All', 'Unlinked'): return True else: return True if Index == For else False def PatchHTML(File, HTML, StaticPartsText, DynamicParts, DynamicPartsText, HTMLPagesList, PagePath, Content, Titles, Meta, SiteDomain, SiteRoot, SiteName, BlogName, FolderRoots, Categories, SiteLang, Locale, LightRun): HTMLTitles = FormatTitles(Titles) BodyDescription, BodyImage = '', '' if not File.lower().endswith('.txt'): Soup = BeautifulSoup(Content, 'html.parser') if not BodyDescription and Soup.p: BodyDescription = Soup.p.get_text()[:150].replace('\n', ' ').replace('"', "'") + '...' if not BodyImage and Soup.img and Soup.img['src']: BodyImage = Soup.img['src'] #Content = SquareFnrefs(Content) if '', '[', ']') if any(_ in Content for _ in ('', '', '', '')): Content = DictReplWithEsc( Content, { '': '', '': '', '': '', '': ''}) Title = GetTitle(File.split('/')[-1], Meta, Titles, 'MetaTitle', BlogName) Description = GetDescription(Meta, BodyDescription, 'MetaDescription') Image = GetImage(Meta, BodyImage, 'MetaImage') ContentHeader = MakeContentHeader(Meta, Locale, MakeCategoryLine(File, Meta)) TimeNow = datetime.now().strftime('%Y-%m-%d %H:%M') RelativeRoot = GetPathLevels(PagePath) if 'staticoso:DynamicPart:' in HTML: # Reduce risk of unnecessary cycles for Line in HTML.splitlines(): Line = Line.lstrip().rstrip() if (Line.startswith('[staticoso:DynamicPart:') and Line.endswith(']')) or (Line.startswith('')): Path = Line[len('", Text) for e in StaticPartsText: HTML = ReplWithEsc(HTML, f"[staticoso:StaticPart:{e}]", StaticPartsText[e]) HTML = ReplWithEsc(HTML, f"", StaticPartsText[e]) if LightRun: HTML = None else: HTML = DictReplWithEsc(HTML, { '[staticoso:Site:Menu]': HTMLPagesList, '': HTMLPagesList, '[staticoso:Page:Lang]': SiteLang, '': SiteLang, '[staticoso:Page:Chapters]': HTMLTitles, '': HTMLTitles, '[staticoso:Page:Title]': Title, '': Title, '[staticoso:Page:Description]': Description, '': Description, '[staticoso:Page:Image]': Image, '': Image, '[staticoso:Page:Path]': PagePath, '': PagePath, '[staticoso:Page:Style]': Meta['Style'], '': Meta['Style'], # NOTE: Content is injected in page only at this point! Keep in mind for other substitutions '[staticoso:Page:Content]': Content, '': Content, '[staticoso:Page:ContentInfo]': ContentHeader, '': ContentHeader, '[staticoso:BuildTime]': TimeNow, '': TimeNow, '': SiteDomain, '[staticoso:Site:Name]': SiteName, '': SiteName, '[staticoso:Site:AbsoluteRoot]': SiteRoot, '': SiteRoot, '[staticoso:Site:RelativeRoot]': RelativeRoot, '': RelativeRoot}) for e in Meta['Macros']: HTML = ReplWithEsc(HTML, f"[:{e}:]", Meta['Macros'][e]) for e in FolderRoots: HTML = DictReplWithEsc(HTML, { f"[staticoso:CustomPath:{e}]": FolderRoots[e], f"": FolderRoots[e], #DEPRECATED f"[staticoso:Folder:{e}:AbsoluteRoot]": FolderRoots[e], f"": FolderRoots[e]}) for e in Categories: HTML = DictReplWithEsc(HTML, { f"[staticoso:Category:{e}]": Categories[e], f"": Categories[e], f"": Categories[e], #DEPRECATED f"[staticoso:Category:{e}]": Categories[e]}) # TODO: Clean this doubling? ContentHTML = Content ContentHTML = DictReplWithEsc(ContentHTML, { '[staticoso:Page:Title]': Title, '': Title, '[staticoso:Page:Description]': Description, '': Description, '': SiteDomain, '[staticoso:Site:Name]': SiteName, '': SiteName, '[staticoso:Site:AbsoluteRoot]': SiteRoot, '': SiteRoot, '[staticoso:Site:RelativeRoot]': RelativeRoot, '': RelativeRoot}) for e in Meta['Macros']: ContentHTML = ReplWithEsc(ContentHTML, f"[:{e}:]", Meta['Macros'][e]) for e in FolderRoots: ContentHTML = DictReplWithEsc(ContentHTML, { f"[staticoso:CustomPath:{e}]": FolderRoots[e], f"": FolderRoots[e], #DEPRECATED f"[staticoso:Folder:{e}:AbsoluteRoot]": FolderRoots[e], f"": FolderRoots[e]}) for e in Categories: ContentHTML = DictReplWithEsc(ContentHTML, { f"[staticoso:Category:{e}]": Categories[e], f"": Categories[e], f"": Categories[e], #DEPRECATED f"[staticoso:Category:{e}]": Categories[e]}) return HTML, ContentHTML, Description, Image def HandlePage(Flags, Page, Pages, Categories, LimitFiles, Snippets, ConfMenu, Locale): File, Content, Titles, Meta = Page OutDir, MarkdownExts, Sorting, MinifyKeepComments = Flags['OutDir'], Flags['MarkdownExts'], Flags['Sorting'], Flags['MinifyKeepComments'] SiteName, BlogName, SiteTagline = Flags['SiteName'], Flags['BlogName'], Flags['SiteTagline'] SiteTemplate, SiteLang = Flags['SiteTemplate'], Flags['SiteLang'] SiteDomain, SiteRoot, FolderRoots = Flags['SiteDomain'], Flags['SiteRoot'], Flags['FolderRoots'] AutoCategories, CategoryUncategorized = Flags['CategoriesAutomatic'], Flags['CategoriesUncategorized'] ImgAltToTitle, ImgTitleToAlt = Flags['ImgAltToTitle'], Flags['ImgTitleToAlt'] DynamicParts, DynamicPartsText, StaticPartsText, TemplatesText = Flags['DynamicParts'], Snippets['DynamicParts'], Snippets['StaticParts'], Snippets['Templates'] FileLower = File.lower() PagePath = f"{OutDir}/{StripExt(File)}.html" LightRun = False if LimitFiles == False or File in LimitFiles else True if FileLower.endswith(FileExtensions['Markdown']): Content = markdown(PagePostprocessor('md', Content, Meta), extensions=MarkdownExts) elif FileLower.endswith(('.pug')): Content = PagePostprocessor('pug', ReadFile(PagePath), Meta) elif FileLower.endswith(('.txt')): Content = '

    ' + html.escape(Content) + '
    ' elif FileLower.endswith(FileExtensions['HTML']): Content = ReadFile(PagePath) if LightRun: HTMLPagesList = None else: TemplateMeta = TemplatePreprocessor(TemplatesText[Meta['Template']]) HTMLPagesList = GetHTMLPagesList( Pages=Pages, BlogName=BlogName, SiteRoot=SiteRoot, PathPrefix=GetPathLevels(File), Unite=ConfMenu, Type='Page', For='Menu', MarkdownExts=MarkdownExts, MenuStyle=TemplateMeta['MenuStyle']) HTML, ContentHTML, Description, Image = PatchHTML( File=File, HTML=TemplatesText[Meta['Template']], StaticPartsText=StaticPartsText, DynamicParts=DynamicParts, DynamicPartsText=DynamicPartsText, HTMLPagesList=HTMLPagesList, PagePath=PagePath[len(f"{OutDir}/"):], Content=Content, Titles=Titles, Meta=Meta, SiteDomain=SiteDomain, SiteRoot=SiteRoot, SiteName=SiteName, BlogName=BlogName, FolderRoots=FolderRoots, Categories=Categories, SiteLang=SiteLang, Locale=Locale, LightRun=LightRun) HTML = ReplWithEsc(HTML, f"", GetHTMLPagesList( Limit=Flags['FeedEntries'], Type='Post', Category=None if Flags['FeedCategoryFilter'] == '*' else Flags['FeedCategoryFilter'], Pages=Pages, BlogName=BlogName, SiteRoot=SiteRoot, PathPrefix=GetPathLevels(File), For='Categories', MarkdownExts=MarkdownExts, MenuStyle='Flat', ShowPaths=False)) if 'staticoso:DirectoryList:' in HTML: # Reduce risk of unnecessary cycles for Line in HTML.splitlines(): Line = Line.lstrip().rstrip() if Line.startswith(''): Path = Line[len('", DirectoryList) if Flags['MinifyOutput']: if not LightRun: HTML = DoMinifyHTML(HTML, MinifyKeepComments) ContentHTML = DoMinifyHTML(ContentHTML, MinifyKeepComments) if Flags['NoScripts'] and ("