...

README.md

@@ -4,8 +4,9 @@ pppadump
 
 Bunch of scripts to extract documents from an [etherpad](http://etherpad.org/) instance and produce templated static files.
 
-Forked from [constant](https://constantvzw.org/site/)'s [active archives](http://activearchives.org/wiki/Main_Page) [etherdump](https://gitlab.constantvzw.org/aa/etherdump) mixed with logic devised for [varia](https://varia.zone)'s [etherpump](https://git.vvvvvvaria.org/varia/etherpump) project
+Forked from [constant](https://constantvzw.org/site/)'s [active archives](http://activearchives.org/wiki/Main_Page) [etherdump](https://gitlab.constantvzw.org/aa/etherdump) mixed with logic devised for [varia](https://varia.zone)'s [etherpump](https://git.vvvvvvaria.org/varia/etherpump) project.
 
+It's all about \_\_MAGICWRORDS__ 🪄✨ 
 
 
 Requirements / dependencies

pppadump/commands/index.py

@@ -98,6 +98,8 @@ def main (args):
 
     p.add_argument("--files", default=False, action="store_true", help="include files (experimental)")
 
+    p.add_argument("--filebased", default=False, action="store_true", help="uses (reconstructed) urls instead of file paths")
+
     pg = p.add_argument_group('template variables')
     pg.add_argument("--feedurl", default="feed.xml", help="rss: to use as feeds own (self) link, default: feed.xml")
     pg.add_argument("--siteurl", default=None, help="rss: to use as channel's site link, default: the etherpad url")
@@ -275,7 +277,7 @@ def main (args):
 
         if "text" in versions_by_type:
             try:
-                with open (versions_by_type["text"]["path"]) as f:
+                with open (versions_by_type["text"]["path"], encoding="utf-8") as f:
                     p["text"] = f.read()
             except FileNotFoundError:
                 p['text'] = ''
@@ -292,35 +294,57 @@ def main (args):
                     pass
 
         
-        # Not sure this goes here but fixing relpaths with output is quite nice...         
-        if args.output:
-            outpath = Path(args.output).parent
-            for v in p["versions"]:
-                if v["type"] != "pad":
-                    vpath = Path(v["path"])
-                    try:
-                        # 2 types (pathlib or os.path)
-                        # v["path"] = vpath.parent.relative_to(outpath).joinpath(vpath.name)
-                        v["path"] = os.path.join(os.path.relpath(vpath.parent, outpath), vpath.name)
+        # Below is somehow a mess since one the one hand I'm trying to keep the file based publiching
+        # ethos while my setup is not really (reverse proxy, etc.) 
+        # Fix (1): settings.json has indexurl which stands for the base URL 
+        # Fix (2): added argument <p.add_argument("--filebased", default=False, action="store_true", help="uses (reconstructed) urls instead of file paths")>      
 
-                        # path or url??? if it's a file based publishing ethos, it should be path non?
-                        # puttting this here since templates are using url... 
-                        v["url"] = v["path"]
+        if args.output:
+
+            if args.filebased:
+
+                # work with paths
+                
+                outpath = Path(args.output).parent
+                for v in p["versions"]:
+
+                    if v["type"] != "pad":
+                        vpath = Path(v["path"])
+
+                        try:
+                            v["path"] = os.path.join(os.path.relpath(vpath.parent, outpath), vpath.name)
+                            v["url"] = v["path"]
+                        except Exception as e:
+                            pass
+
+                if not p["link"].startswith("http"):
+                    vpath = Path(p["link"])
+                    try:
+                        p["link"] = os.path.join(os.path.relpath(vpath.parent, outpath), vpath.name)
                     except Exception as e:
                         pass
 
-            if not p["link"].startswith("http"):
-                vpath = Path(p["link"])
-                try:
-                    p["link"] = os.path.join(os.path.relpath(vpath.parent, outpath), vpath.name)
-                except Exception as e:
-                    pass
+            else:
 
+                # work with URLs
 
+                indexurl = info["indexurl"]
+                outpath = Path(args.output).parent
+
+                for v in p["versions"]:
+
+                    if v["type"] != "pad":
+
+                        vpath = Path(v["url"])
+
+                        # strip output dir....
+                        relpath = Path(*vpath.parts[1:])
+
+                        v["url"] = os.path.join(indexurl, relpath.as_posix())
 
 
     if args.output:
-        with open(args.output, "w") as f:
+        with open(args.output, "w", encoding="utf-8") as f:
             print (template.render(vars(args)), file=f)
     else:
         print (template.render(vars(args)))

pppadump/commands/init.py

@@ -73,6 +73,7 @@ def main(args):
     p.add_argument("arg", nargs="*", default=[], help="optional positional args: path etherpadurl")
     p.add_argument("--path", default=None, help="path to initialize")
     p.add_argument("--padurl", default=None, help="")
+    p.add_argument("--indexurl", default=None, help="")
     p.add_argument("--apikey", default=None, help="")
     p.add_argument("--verbose", default=False, action="store_true", help="")
     p.add_argument("--reinit", default=False, action="store_true", help="")
@@ -114,6 +115,15 @@ def main(args):
             break
         apiurl = input("Please type the URL of the etherpad: ").strip()
     padinfo["apiurl"] = apiurl
+
+
+    indexurl = args.indexurl
+    while True:
+        if indexurl:
+            break
+        indexurl = input("Please type the URL of the index to be generated (i.e. should differ from padurl): ").strip()
+    padinfo["indexurl"] = indexurl
+
     apikey = args.apikey
     while True:
         if apikey: