initial commit

1 files changed, 519 insertions, 0 deletions

diff --git a/mbbmlib.py b/mbbmlib.py
new file mode 100644
index 0000000..68cbca2
--- /dev/null
+++ b/mbbmlib.py
@@ -0,0 +1,519 @@
+#!/usr/bin/env python3
+# File ffbm_lib.py
+# License: CC-BY-SA 4.0
+# Author: bgstack15@gmail.com
+# Startdate: 2020-12-12
+# Title: Export Firefox Bookmarks to Html
+# Purpose: Add favicons to bookmarks in exported html file
+# History:
+# Usage:
+#    main command is export_bookmarks_to_html.
+# Reference:
+#    pragma table_info(moz_bookmarks)
+#    https://stackoverflow.com/questions/464516/firefox-bookmarks-sqlite-structure?rq=1
+#    https://stackoverflow.com/questions/40408607/attach-database-in-sqlite3-with-python
+#    https://2.python-requests.org/en/latest/
+#    https://stackoverflow.com/questions/43446353/converting-png-images-to-base64
+#    future: https://stackoverflow.com/questions/5119041/how-can-i-get-a-web-sites-favicon
+#    https://rosettacode.org/wiki/Strip_control_codes_and_extended_characters_from_a_string#Python
+# Improve:
+#    add to pretty_print the option to select what format the output is in, like "title | url | icon" or similar.
+# Documentation:
+# Dependencies:
+#    devuan: python3
+import sqlite3, sys, os
+from urllib.parse import urlparse
+import requests, base64
+
+try:
+   if "PYTHONIOENCODING" not in os.environ:
+      print("You should really run this with env var PYTHONIOENCODING=UTF-8\n",sys.stderr)
+except:
+   pass
+
+class Bookmark:
+   def __init__(self, 
+         bid=None, btype=0, url="", title="", dateAdded=0, lastModified=0, bparent=0, position=0, folder_type=0, favicon="", current_bparent=0
+   ):
+      if bid == "root":
+         """
+         Make a root bookmark for populating with the contents of a places.sqlite 
+         Pass in just a single string with value "root" to bookmark:
+         I had to use this if wrapper because __init__ is apparently the one function of a python class that cannot be overloaded.
+         books = Bookmark("root")
+         """
+         self.bid = 0
+         self.btype = 0
+         self.url = "file:///self"
+         self.bparent = 0
+         self.position = 0
+         self.folder_type = 0
+         self.title = "root"
+         self.dateAdded = 0
+         self.lastModified = 0
+         self.favicon = ""
+         self.children = []
+         self.current_bparent = 0
+      else:
+         self.bid = bid
+         self.btype = btype
+         self.url = url
+         self.bparent = bparent #int
+         self.position = position # int
+         self.folder_type = folder_type
+         if title is not None:
+            #self.title = title.encode('utf-8','replace').decode('latin-1','replace')
+            stripped = lambda s: "".join(i for i in title if 31 < ord(i) < 1024*8)
+            self.title = stripped(title)
+         else:
+            self.title = ""
+         self.dateAdded = dateAdded # int
+         self.lastModified = lastModified # int
+         self.favicon = favicon
+         self.children = []
+         self.current_bparent = current_bparent
+
+   # make this bookmark object basically act like the list of its children
+   def count(self):
+      return len(self.children)
+   def append(self, newitem):
+      self.children.append(newitem)
+   def remove(self, olditem):
+      self.children.remove(olditem)
+   def __iter__(self):
+      return self.children.__iter__()
+   def __next__(self):
+      return self.children.__next__()
+   def __len__(self):
+      return len(self.children)
+   def __eq__(self, other):
+      return self.bid == other.bid
+
+   def pretty_print(self,i_count=0):
+      print(i_count*' |','+',self.title)
+      i_count+=1
+      for child in self.children:
+         child.pretty_print(i_count)
+
+   def to_html(self,i_count=0,file=sys.stdout,iconsize=32):
+
+      # i_count is used to determine nested level
+      if i_count == 0:
+         f = None
+         if file != sys.stdout:
+            print("Saving to file {0}".format(file),file=sys.stderr)
+            f = open(file,'w')
+            old_stdout = sys.stdout
+            sys.stdout = f
+         print("<html>")
+         print("<head><title>{0}</title>".format("Bookmarks"))
+         print("<style>img #here {{ width: {0}px; height: {0}px; }}</style>".format(iconsize))
+         print("<style>img {{ width: 0px; height: 0px; }}</style>")
+         print("</head>")
+         print("<body>")
+
+      # print self
+      # btype == 2 is a folder
+      if self.btype == 2:
+         #print("<h2>{0}</h2>".format(self.title))
+         li_start = ""
+         li_stop = ""
+         if i_count >= 1:
+            li_start = "<li>"
+            li_stop = "</li>"
+         print("<h{0}>{1}{2}{3}</h{0}>".format(min(i_count,8),li_start,self.title,li_stop))
+      else:
+         #print("{0} {1} {2}".format(self.bid,self.btype,self.title))
+         title = self.title
+         if title == "":
+            title = "Untitled"
+         # specifically exclude a few internal items like "recently bookmarked"
+         if not self.url.startswith("place:") and not self.url.startswith("about:logopage"):
+            print(
+               '<li><a href="{0}" add_date="{1}" last_modified="{2}" ICON="{3}"><img {4}src="{3}"/>{5}</a></li>'.format(
+                  self.url,
+                  self.dateAdded,
+                  self.lastModified,
+                  self.favicon,
+                  'class="here" ' if self.favicon is not None else ' ',
+                  title,
+               )
+            )
+      # handle children
+      i_count+=1
+      if self.count() > 0:
+         print("<ul>")
+      for child in self.children:
+         child.to_html(i_count)
+      if self.count() > 0:
+         print("</ul>")
+
+      # close whole thing
+      if i_count == 1: # main body is now in i_count == 1
+         print("</body></html>")
+
+      # close file if it is open
+      try:
+         if f:
+            f.close()
+            sys.stdout = old_stdout
+      except:
+         pass
+
+   def sort_children(self):
+      self.new_children = sorted(self.children, key=lambda r: r.position)
+      self.children , self.new_children = self.new_children, self.children
+      self.new_children = None
+      for child in self.children:
+         child.sort_children()
+
+   def find(self,searchfield,value,i_count=0, debuglev = 0):
+      if searchfield != "id" and searchfield != "title":
+         #print("Error: Bookmark",self.bid," with title",self.title," 
+         print("Error: Bookmark object can only search on fields [\"id\",\"title\"]",file=sys.stderr)
+         return None
+      if searchfield == "id":
+         if self.bid == value:
+            return self
+         if debuglev >= 5:
+            print(i_count,"My id",self.bid,self.title,"is not searched value",value)
+         for child in self:
+            if child.bid == value:
+               #print("...but my child is!",child.bid,child)
+               return child
+            a = child.find(searchfield,value,i_count+1)
+            if a:
+               return a
+         # did not find it
+         return None
+      elif searchfield == "title":
+         print("Error: Gotta say unh! Bookmark search for title is not yet implemented.",file=sys.stderr)
+         return None
+      # catch-all
+      return None
+
+   def all_children(self):
+      # goal: return a flat list of all items underneath this one, and its real parent id
+      self.all_children_list = []
+      for child in self.children:
+         self.all_children_list.append(child)
+         for i in child.all_children():
+            self.all_children_list.append(i)
+         #if len(child.all_children()) > 0:
+         #   self.all_children_list.append(child.all_children())
+      return self.all_children_list
+
+   def prune(self, debuglev = 0):
+      prune_count = 0
+      i = 0
+      # using an index instead of "for child in self.children" allows 
+      # me to step backwards one so I don't miss the item right after the one I just deleted.
+      while i < len(self.children):
+         child = self.children[i]
+         prune_count += child.prune(debuglev=debuglev)
+         if child.bparent != self.bid:
+            if debuglev >= 5:
+               print("Removing",child.bid,child.title,"from parent",self.bid,self.title)
+            prune_count += 1
+            self.children.remove(child)
+            i -= 1
+         i += 1
+      return prune_count
+
+   def unflatten(self,debuglev=0,i_count=0, root = None):
+      """
+      Main task of moving bookmarks to their intended parent bookmark objects.
+      """
+      if debuglev >= 7:
+         print("{0} {1} has {2} children.".format(self.bid,self.title,self.count()))
+      safety_limit = 2000
+      x=0
+      if i_count == 0:
+         root = self
+      
+      for child in self.children:
+         x += 1
+         if x < safety_limit:
+            try:
+               if debuglev >= 7:
+                  print("{0} {1} is looking for parent {2}".format(child.bid,child.title,child.bparent))
+               if child.bparent > 0:
+                  try:
+                     thisparent=root.find('id',child.bparent)
+                     try:
+                        if self.bid != child.bparent:
+                           thisparent.append(child)
+                           # the old child will still exist until we run prune() which will remove
+                           # any child child whose bparent is not the same as its parent bid.
+                           if debuglev >= 5:
+                              print("SUCCESS: move",child.bid,child,"to parent",child.bparent,thisparent)
+                        else:
+                           if debuglev >= 6:
+                              print("info: {0} {1} is already underneath parent {2} {3}".format(child.bid,child.title,thisparent.bid,thisparent.title))
+                     except:
+                        if debuglev >= 4:
+                           print("FAIL: move",child.bid,child,"to parent",child.bparent,thisparent)
+                  except:
+                     if debuglev >= 4:
+                        print("Unable to find parent item which should be bid",item.bparent)
+            except:
+               if debuglev >= 4:
+                  print("Unable to list bparent for child",child)
+               pass
+         else:
+            if debuglev >= 1:
+               print("{0} {1} has to stop after {2} children".format(self.bid,self.title,safety_limit),file=sys.stderr)
+            break
+         # still within for child in self.children
+         child.unflatten(debuglev=debuglev, i_count=i_count+1, root=root)
+
+      self.prune(debuglev=debuglev)
+      self.sort_children()
+      if self.count() == 1:
+         return self.children[0]
+      else:
+         return self
+
+   def __str__(self):
+      encoded = "Encoding error!"
+      try:
+         encoded = self.title.encode('utf-8','replace').decode('utf-8','replace')
+      except:
+         pass
+      return encoded
+
+   def load_from_places(self, places_file, icons_file = None, limit = 50, fetch_icons=True, debuglev = 0, iconsize = 32, browser = "autodetect"):
+      # initialize
+      db_bookmarks = []
+      db_places = []
+      db_icons = []
+      if limit >= 8000:
+         print("WARNING! Limit is really big, but we will proceed.",file=sys.stderr)
+      elif limit < 1:
+         limit = 1
+
+      if icons_file is None:
+         icons_file = places_file.replace("places.","favicons.")
+
+      # retrieve contents of sqlite
+      if debuglev > 4:
+         print("Using places file {0} and icons_file {1}".format(places_file,icons_file))
+      with sqlite3.connect(places_file) as conn:
+
+         # detect browser automatically
+         if browser == 'autodetect':
+            try:
+               conn.execute("SELECT syncStatus FROM moz_bookmarks limit 1")
+               browser = "firefox"
+            except sqlite3.OperationalError as e:
+               if str(e).startswith("no such column"):
+                  browser = "palemoon"
+               else:
+                  print("Other error,",e)
+                  e
+
+         if debuglev > 3:
+            print("Using browser value {0}".format(browser))
+
+         # now conduct regular operations
+         if browser == 'firefox':
+            raw_bookmarks= conn.execute("SELECT * FROM moz_bookmarks")
+            raw_places = conn.execute("SELECT * FROM moz_places")
+         elif browser == 'palemoon':
+            raw_bookmarks= conn.execute("SELECT id,type,fk,parent,position,title,keyword_id,folder_type,dateAdded,lastModified,guid FROM moz_bookmarks")
+            raw_places = conn.execute("SELECT id,url,title,rev_host,visit_count,hidden,typed,favicon_id,frecency,last_visit_date,guid,foreign_count,url_hash FROM moz_places")
+
+      with sqlite3.connect(icons_file) as conn:
+         raw_icons = conn.execute("SELECT * FROM moz_icons")
+
+      if browser == "firefox":
+         for row in raw_bookmarks:
+            bid, btype, parent, fk, position, title, keyword_id, folder_type, dateAdded, lastModified, guid, syncStatus, syncChangeCounter = row
+            db_bookmarks.append(db_bookmark(bid, btype, parent, fk, position, title, keyword_id, folder_type, dateAdded, lastModified, guid, syncStatus, syncChangeCounter))
+         for row in raw_places:
+            pid, url, title, rev_host, visit_count, hidden, typed, favicon_id, frecency, last_visit_date, guid, foreign_count, url_hash, description, preview_image_url, origin_id = row
+            db_places.append(db_place(pid, url, title, rev_host, visit_count, hidden, typed, favicon_id, frecency, last_visit_date, guid, foreign_count, url_hash, description, preview_image_url, origin_id))
+
+      elif browser == "palemoon":
+         for row in raw_bookmarks:
+            bid, btype, fk, parent, position, title, keyword_id, folder_type, dateAdded, lastModified, guid = row
+            db_bookmarks.append(db_bookmark(bid, btype, fk, parent, position, title, keyword_id, folder_type, dateAdded, lastModified, guid))
+         for row in raw_places:
+            pid, url, title, rev_host, visit_count, hidden, typed, favicon_id, frecency, last_visit_date, guid, foreign_count, url_hash = row
+            db_places.append(db_place(pid, url, title, rev_host, visit_count, hidden, typed, favicon_id, frecency, last_visit_date, guid, foreign_count, url_hash, "", "", 0))
+
+      if fetch_icons:
+         for row in raw_icons:
+            iid, icon_url, fixed_icon_url, width, root, color, expire_ms, data = row
+            db_icons.append(db_icon(iid, icon_url, fixed_icon_url, width, root, color, expire_ms, data))
+
+      ## print, to validate
+      #for bm in db_bookmarks:
+      #   print(bm)
+      if debuglev >= 1:
+         print("Found this many bookmarks, places, and icons:")
+         print(len(db_bookmarks))
+         print(len(db_places))
+         if fetch_icons:
+            print(len(db_icons))
+
+      # now all items are loaded.
+      # flow:
+      # 1. make new list with Bookmark objects with desired info from all 3 old lists
+      # 2. move bookmarks with "parent" value to be a child of that parent
+
+      # make new list with all info
+
+      count = 0
+      for item in db_bookmarks:
+         count += 1
+         if count >= limit:
+            continue
+         place = None
+         icon = None
+         url = ""
+         favicon = ""
+
+         # print this, so I have something to catch so I can interrupt this process like with | head -n40
+         if debuglev >= 1:
+            print("{0}: {1}".format(count,item))
+
+         # attempt to find matching place
+         try:
+            for tmp_place in db_places:
+               if tmp_place.pid == item.fk:
+                  place = tmp_place
+                  continue
+         except:
+            pass
+         try:
+            url = place.url
+         except:
+            pass
+
+         # attempt to find icon for that place
+         if fetch_icons:
+            try:
+               if place.favicon_id is not None:
+                  for tmp_icon in db_icons:
+                     if tmp_icon.iid == place.favicon_id:
+                        icon = tmp_icon
+                        continue
+            except:
+               pass
+            try:
+               favicon = icon.icon_url
+            except:
+               pass
+            # use google service to get a favicon for this domain
+            try:
+               if favicon == "" and url != "":
+                  domain = urlparse(url).netloc
+                  if domain != "":
+                     #print("Will try to download favicon for domain", domain)
+                     # fetch it from google
+                     # parameter sz is undocumented but works
+                     r = requests.get("http://www.google.com/s2/favicons?sz={0}&domain={1}".format(iconsize,domain))
+                     if r.status_code == 200:
+                        # r.content is the useful stuff
+                        a = base64.b64encode(r.content)
+                        favicon = "data:image/png;base64," + a.decode('utf-8')
+                     else:
+                        print("unable to fetch icon for",domain," because of:",file=sys.stderr)
+                        print(r,file=sys.stderr)
+            except:
+               pass
+
+         #print("{0} | {1} | {2}".format(item,url,favicon))
+         self.append(
+            Bookmark(
+               item.bid, item.btype, url, item.title, item.dateAdded, item.lastModified, item.bparent, item.position, item.folder_type, favicon
+            )
+         )
+
+class db_bookmark:
+   def __init__(self,bid,btype,fk,bparent,position,title,keyword_id,folder_type,dateAdded,lastModified,guid,syncStatus="",syncChangeCounter=""):
+      self.bid = bid
+      self.btype = btype
+      self.bparent = bparent
+      self.fk = fk
+      self.position = position
+      self.title = title
+      self.keyword_id = keyword_id
+      self.folder_type = folder_type
+      self.dateAdded = dateAdded
+      self.lastModified = lastModified
+      self.guid = guid
+      self.syncStatus = syncStatus
+      self.syncChangeCounter = syncChangeCounter
+
+   def __str__(self):
+      encoded = "Encoding error on bookmark " + str(self.bid)
+
+      try:
+         encoded = self.title.encode('utf-8','replace') if self.title is not None else "(untitled)"
+         #encoded = self.title if self.title is not None else "(untitled)"
+      except:
+         pass
+      try:
+         return encoded.decode('utf-8','replace')
+      except:
+         return encoded
+
+class db_place:
+   def __init__(self,pid,url,title,rev_host,visit_count,hidden,typed,favicon_id,frecency,last_visit_date,guid,foreign_count,url_hash,description,preview_image_url,origin_id):
+      self.pid = pid
+      self.url = url
+      self.title = title
+      self.rev_host = rev_host
+      self.visit_count = visit_count
+      self.hidden = hidden
+      self.typed = typed
+      self.favicon_id = favicon_id
+      self.frecency = frecency
+      self.last_visit_date = last_visit_date
+      self.guid = guid
+      self.foreign_count = foreign_count
+      self.url_hash = url_hash
+      self.description = description
+      self.preview_image_url = preview_image_url
+      self.origin_id = origin_id
+
+class db_icon:
+   def __init__(self,iid,icon_url,fixed_icon_url,width,root,color,expire_ms,data):
+      self.iid = iid
+      self.icon_url = icon_url
+      self.fixed_icon_url = fixed_icon_url
+      self.width = width
+      self.root = root
+      self.color = color
+      self.expire_ms = expire_ms
+      self.data = data
+
+def export_bookmarks_to_html(places_file, output_file, limit=2000, fetch_icons=True, iconsize=32, debuglev=0):
+   """
+   Export to output_file the html representation of the bookmarks in the input places_file.
+   This is the main interface from the library.
+   """
+
+   # prepare the places_file in case we were only handed a profile directory name
+   if "places.sqlite" not in places_file:
+      orig_places_file = places_file
+      places_file += "/places.sqlite"
+      if not os.path.isfile(places_file):
+         print("Provided path {0} does not contain a places.sqlite file. Cannot export these bookmarks to html!".format(orig_places_file),file=sys.stderr)
+         return 1
+
+   bm = Bookmark("root")
+   favicons_file = places_file.replace("places.","favicons.")
+   bm.load_from_places(places_file, limit=limit, fetch_icons=fetch_icons, iconsize=iconsize, debuglev=debuglev)
+   b2 = bm.unflatten(debuglev=debuglev)
+   b2.to_html(file=output_file,iconsize=iconsize)
+
+def find_available_places(basedir=os.path.expanduser("~")):
+   # return list of places.sqlite that meet the criteria
+   import pathlib
+   results = []
+   for path in pathlib.Path(basedir).rglob('places.sqlite'):
+      results.append(str(path.parent))
+   return results