troggle-unchained/core/views_expo.py

import os
import re
from pathlib import Path
from urllib.parse import urljoin, unquote as urlunquote
from urllib.request import urlopen

from django.shortcuts import render, redirect
from django.http import HttpResponse, HttpResponseRedirect, Http404
from django.urls import reverse, resolve
from django.template import Context, loader
from django.views.decorators.csrf import ensure_csrf_cookie

import django.forms as forms

from troggle.helper import login_required_if_public
from troggle.core.models_caves import Cave
import troggle.core.views_caves
import troggle.settings as settings

'''Formerly a separate package  'flatpages' written by Martin Green 2011. 
This was NOT django.contrib.flatpages which stores HTML in the database, so the name was chnaged to expopages.
Then it was incorporated into troggle directly, rather than being an unnecessary external package.
'''

def expofiles_redirect(request, path):
    '''This is used only when running as a test system without a local copy of /expofiles/
    '''
    return redirect(urljoin('http://expo.survex.com/expofiles/', path))

def expofilessingle(request, filepath):
    '''sends a single binary file to the user,
    '''
    fn=urlunquote(filepath)
    fn = Path(settings.EXPOFILES,filepath)
    if fn.is_dir():
        return expofilesdir(request, Path(fn), Path(filepath))
    # print(" - expofilessingle {}:{}:{}:".format(filepath, fn, getmimetype(fn)))
    return HttpResponse(content=open(fn, "rb"),content_type=getmimetype(filepath)) # any file

def expofilesdir(request, dirpath, filepath):
    '''does a directory display. If there is an index.html file we should display that.
    - dirpath is a Path() and it does not have /expofiles/ in it
    '''
    # print(" - expofilesdir {}".format(dirpath))
    urlpath = 'expofiles' / Path(filepath)
    fileitems = []
    diritems = []
    for f in dirpath.iterdir():
        if f.is_dir():
             diritems.append((urlpath / f.parts[-1], str(f.parts[-1])))
        else:
            # if f.parts[-1].lower() == 'index.htm' or f.parts[-1].lower() == 'index.html': # css cwd problem
                # return HttpResponse(content=open(f, "rb"),content_type=getmimetype(filepath)) # any file
                # return expofilessingle(request, str(Path(filepath /  f.parts[-1])))             
            fileitems.append((Path(urlpath) / f.parts[-1], str(f.parts[-1]), getmimetype(f)))
    return render(request, 'dirdisplay.html', { 'filepath': urlpath, 'fileitems':fileitems, 'diritems': diritems,'settings': settings })

def flatpage(request, path):
    '''Either renders an HTML page from expoweb with all the menus,
    or serves an unadorned binary file with mime type
    
    This is a horrible mess and some code is redundant and unreachable because of urls.py setup
    '''
    # print(" - EXPOPAGES delivering the file: {} as MIME type: {}".format(path,getmimetype(path)),flush=True)

    if path.startswith("noinfo") and settings.PUBLIC_SITE and not request.user.is_authenticated():
        # print((" - EXPOPAGES redirect to logon: flat path noinfo", path))
        return HttpResponseRedirect(urljoin(reverse("auth_login"),'?next={}'.format(request.path)))

    expowebpath = Path(settings.EXPOWEB)
    if path.endswith("/") or path == "":
        # print(" - EXPOPAGES the file: {} ENDSWITH ...".format(path))

        try:
            o = open(os.path.normpath(expowebpath / path / "index.html"), "rb")
            path = path + "index.html"
        except IOError:
            try:
                o = open(os.path.normpath(expowebpath / path / "index.htm"), "rb")
                path = path + "index.htm"
            except IOError:
                return render(request, 'pagenotfound.html', {'path': path})
    else:
        # print(" - EXPOPAGES the file: '{}'  ...".format(path))
        if path.startswith('site_media'):
            # print(" - MEDIA_ROOT: {}  ...{}".format(settings.MEDIA_ROOT, path))
            path = path.replace("site_media", settings.MEDIA_ROOT)
            filetobeopened = os.path.normpath(path)
        elif path.startswith("static"):
            # print(" - STATIC_ROOT: {}  ...{}".format(settings.MEDIA_ROOT, path))
            path = path.replace("static", settings.MEDIA_ROOT)
            filetobeopened = os.path.normpath(path)
        else:
            # print(" - NO    _ROOT: {}  ...".format(expowebpath))
            filetobeopened = os.path.normpath(expowebpath / path)
        
        # print(" - EXPOPAGES full path : {}  ...".format(filetobeopened))
        try:
            o = open(filetobeopened, "rb")
            #print(" - EXPOPAGES full path no error: {}  ...".format(filetobeopened))
        except IOError:
            #print(" - EXPOPAGES ERROR: {}  ...".format(filetobeopened))
            #o.close() 
            return render(request, 'pagenotfound.html', {'path': path})
 
 
    if path.endswith(".htm") or path.endswith(".html"):
        # add the menus etc.
        with open(os.path.normpath(expowebpath / path), "rb") as o:   
            html = o.read()
        
        m = re.search(rb'(.*)<\s*head([^>]*)>(.*)<\s*/head\s*>(.*)<\s*body([^>]*)>(.*)<\s*/body\s*>(.*)', html, re.DOTALL + re.IGNORECASE)
        if m:
            preheader, headerattrs, head, postheader, bodyattrs, body, postbody = m.groups()
        else:
            return HttpResponse(html + "HTML Parsing failure: Page could not be split into header and body: failed in expopages.views.py")
        m = re.search(rb"<title>(.*)</title>", head, re.DOTALL + re.IGNORECASE)
        if m:
            title, = m.groups()
        else:
            title = ""
        m = re.search(rb"<meta([^>]*)noedit", head, re.DOTALL + re.IGNORECASE)
        if m:
            editable = False
        else:
            editable = True
        
        has_menu = False
        menumatch = re.match(rb'(.*)<div id="menu">', body, re.DOTALL + re.IGNORECASE)
        if menumatch:
            has_menu = True
        menumatch = re.match(rb'(.*)<ul id="links">', body, re.DOTALL + re.IGNORECASE)
        if menumatch:
            has_menu = True
            #body, = menumatch.groups()
#        if re.search(rb"iso-8859-1", html):
#            body = str(body, "iso-8859-1")
#            body.strip
        return render(request, 'flatpage.html', {'editable': editable, 'path': path, 'title': title, 
                    'body': body, 'homepage': (path == "index.htm"), 'has_menu': has_menu})
    else:
        # print(" - EXPOPAGES delivering the file: {} as MIME type: {}".format(path,getmimetype(path)))
        return HttpResponse(content=open(filetobeopened, "rb"), content_type=getmimetype(path))
        #return HttpResponse(content=open(singlescan.ffile,"rb"), content_type=getmimetype(path))

def getmimetype(path):
    path = str(path)
    if path.lower().endswith(".css"): return "text/css"
    if path.lower().endswith(".txt"): return "text/css"
    if path.lower().endswith(".js"): return "application/javascript"
    if path.lower().endswith(".json"): return "application/javascript"
    if path.lower().endswith(".ico"): return "image/vnd.microsoft.icon"
    if path.lower().endswith(".png"): return "image/png"
    if path.lower().endswith(".tif"): return "image/tif"
    if path.lower().endswith(".gif"): return "image/gif"
    if path.lower().endswith(".jpeg"): return "image/jpeg"
    if path.lower().endswith(".jpg"): return "image/jpeg"
    if path.lower().endswith("svg"): return "image/svg+xml"
    if path.lower().endswith("xml"): return "application/xml" # we use "text/xhtml" for tunnel files
    if path.lower().endswith(".pdf"): return "application/pdf"
    if path.lower().endswith(".ps"): return "application/postscript"
    if path.lower().endswith(".svx"): return "application/x-survex-svx"
    if path.lower().endswith(".3d"): return "application/x-survex-3d"
    if path.lower().endswith(".pos"): return "application/x-survex-pos"
    if path.lower().endswith(".err"): return "application/x-survex-err"
    if path.lower().endswith(".odt"): return "application/vnd.oasis.opendocument.text"
    if path.lower().endswith(".ods"): return "application/vnd.oasis.opendocument.spreadsheet"
    if path.lower().endswith(".docx"): return "application/vnd.openxmlformats-officedocument.wordprocessingml.document"
    if path.lower().endswith(".xslx"): return "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
    if path.lower().endswith(".gz"): return "application/x-7z-compressed"
    if path.lower().endswith(".7z"): return "application/x-7z-compressed"
    if path.lower().endswith(".zip"): return "application/zip"
    return ""

@login_required_if_public
@ensure_csrf_cookie
def editflatpage(request, path):
    try:
        r = Cave.objects.get(url = path)
        return troggle.core.views_caves.editCave(request, r.cave.slug)
    except Cave.DoesNotExist:
        pass


    try:
        filepath = Path(settings.EXPOWEB) / path
        o = open(filepath, "r")
        html = o.read()
        autogeneratedmatch = re.search(r"\<\!--\s*(.*?(Do not edit|auto-generated).*?)\s*--\>", html, re.DOTALL + re.IGNORECASE)
        if autogeneratedmatch:
            return HttpResponse(autogeneratedmatch.group(1))
        m = re.search(r"(.*)<head([^>]*)>(.*)</head>(.*)<body([^>]*)>(.*)</body>(.*)", html, re.DOTALL + re.IGNORECASE)
        if m:
            filefound = True
            preheader, headerargs, head, postheader, bodyargs, body, postbody = m.groups()
            linksmatch = re.match(r'(.*)(<ul\s+id="links">.*)', body, re.DOTALL + re.IGNORECASE)
            if linksmatch:
                body, links = linksmatch.groups()
#            if re.search(r"iso-8859-1", html):
#                body = str(body, "iso-8859-1")
        else:
            return HttpResponse("Page could not be split into header and body")
    except IOError:
        print("### File not found ### ", filepath)
        filefound = False
        

    if request.method == 'POST': # If the form has been submitted...
        flatpageForm = FlatPageForm(request.POST) # A form bound to the POST data
        if flatpageForm.is_valid():# Form valid therefore write file
            print("### \n", str(flatpageForm)[0:300])
            print("### \n csrfmiddlewaretoken: ",request.POST['csrfmiddlewaretoken'])
            if filefound:
                headmatch =  re.match(r"(.*)<title>.*</title>(.*)", head, re.DOTALL + re.IGNORECASE)
                if headmatch:
                    head = headmatch.group(1) + "<title>" + flatpageForm.cleaned_data["title"] + "</title>" + headmatch.group(2)
                else:
                    head = "<title>" + flatpageForm.cleaned_data["title"] + "</title>"
            else:
                head = "<title>" + flatpageForm.cleaned_data["title"] + "</title>"
                preheader = "<html>"
                headerargs = ""
                postheader = ""
                bodyargs = ""
                postbody = "</html>" 
            body = flatpageForm.cleaned_data["html"]
            body = body.replace("\r", "")
            result = "%s<head%s>%s</head>%s<body%s>\n%s</body>%s" % (preheader, headerargs, head, postheader, bodyargs, body, postbody)
            f = open(filepath, "w")
            f.write(result)
            f.close()
            return HttpResponseRedirect(reverse('flatpage', args=[path])) # Redirect after POST
    else:
        if filefound:
            m = re.search(r"<title>(.*)</title>", head, re.DOTALL + re.IGNORECASE)
            if m: 
                title, = m.groups()
            else:
                title = ""
            flatpageForm = FlatPageForm({"html": body, "title": title})
        else:
            body = "### File not found ###\n"    + str(filepath)
            flatpageForm = FlatPageForm({"html": body, "title": "Missing"})
    return render(request, 'editflatpage.html', {'path': path, 'form': flatpageForm, })

class FlatPageForm(forms.Form):
    title = forms.CharField(widget=forms.TextInput(attrs={'size':'60'}))

    #html = forms.CharField(widget=TinyMCE(attrs={'cols': 80, 'rows': 20}))
    html = forms.CharField(widget=forms.Textarea(attrs={"cols":80, "rows":20}))