diff --git a/README.md b/README.md index 5459e0e..f7a5742 100755 --- a/README.md +++ b/README.md @@ -9,7 +9,7 @@ If you want it to do other functions, grab the code and go for it. ## Usage ``` $ ./xbsjsonedit -h -usage: xbsjsonedit [-h] Infile +usage: xbsjsonedit [-h] [--dump] Infile xBrowserSync json backup editor @@ -18,6 +18,7 @@ positional arguments: optional arguments: -h, --help show this help message and exit + --dump Dump bookmark hierarchy (redirect to less or a file) ``` ## Internal command menu @@ -26,13 +27,20 @@ $ ./xbsjsonedit xBrowserSyncBackup_201811232150.json --------------------------------------------------------------------- Options: - s: Search term entry (currently <__NONE__>) - t: List search matches - T: Delete search matches + p: Print the bookmarks tree hierarchy + + s: Search term entry (currently ) + t: List bookmark search matches + T: Delete bookmark search matches + g: List folder names search matches + G: Delete folder names search matches d: List duplicate URLs D: Delete duplicate URLs + f: List duplicate folders + F: Delete duplicate folders + x: List tags on folders (only leaf nodes should have tags) X: Delete tags on folders @@ -42,17 +50,21 @@ Options: w: Write out the bookmarks data to a file q: Quit/exit (Do a Write first!) + Enter option: ``` ## Usage notes -Four modes are supported Listing/Deleting based on -- A search string +Five modes are supported Listing/Deleting based on: +- A search string in bookmarks or folders - Duplicate URLs in bookmarks - such as the same URL existing in two or more folders +- Duplicate folder names - Folders that have tags - there is no value for folders having tags - Empty folders - which may arise if all bookmarks were deleted from a folder -A lower case letter `t/d/x/y` will list the offenders, while an upper case letter -`T/D/X/Y` will allow for selective deletes of the offenders. The lower case List operation +The Print function prints the full list of bookmarks in folder hierarchical form, listing the ID number and Title of each folder and bookmark. The `--dump` switch may be used for getting a good visual dump of the bookmarks in less or an editor for reference while using the interactive mode. + +A lower case letter `t/g/d/f/x/y` will list the offenders, while an upper case letter +`T/G/D/F/X/Y` will allow for selective deletes of the offenders. The lower case List operation need not be run before running the upper case Delete operation. Delete functions support individual bookmark selection, or all that match the selection mode. You cannot do any damage to the original bookmark file during a session, so experiment and poke @@ -68,15 +80,11 @@ You may want to write out intermediate editing results to save work done up to t The output format is "pretty printed" json for readability, and is accepted by the xBrowserSync app Restore function via copy/paste. Note: xBrowserSync will allow you to restore to (blast) your current -syncID, so you may want to Disable Sync, then create a new sync before restoring the modified bookmarks. +syncID, so you may want to Disable Sync, then create a new sync before restoring the modified bookmarks. **NOTE** that tags are not restored if sync is disabled. To recover, simply enable sync and do another restore. ## Known issues: -- This code was developed on Python 2.7.10 on Windows, and exercised on Linux. It is known broken -on Python 3. -- The xBrowserSync Backup function appears to dump out Latin-1 encoding, not UTF-8. The -Write command from this tool outputs UTF-8, which seems to Restore without error in xBrowserSync, -but effectively garbages up the bookmark title text for non-ASCII bookmarks. +- This code only works with Python 3. Development and testing was done on Linux with Python 3.7.3. diff --git a/xbsjsonedit b/xbsjsonedit index d92c0a6..d2573a8 100755 --- a/xbsjsonedit +++ b/xbsjsonedit @@ -4,16 +4,13 @@ #========================================================== # desc = """xBrowserSync json backup editor""" -# Chris Nelson, November 2018 +# Chris Nelson 2018 - 2019 # +# 191205 Updated to xbrowsersync v1.5 and Python 3.x ONLY # 181128 New # # Issues, and changes pending -# This code was developed on Python 2.7.10 on Windows, and exercised on Linux. -# It is known broken on Python 3. -# The xBrowserSync Backup function appears to dump out Latin-1 encoding, not UTF-8. The Write command -# from this tool outputs UTF-8, which seems to Restore without error in xBrowserSync, but effectively -# garbages up the bookmark text for non-ASCII bookmarks. +# None # #========================================================== @@ -22,6 +19,7 @@ import os.path import json import codecs import sys +import io # Configs / Constants @@ -32,6 +30,7 @@ NONE_SEARCH = "__NONE__" # Global items url_dict = {} +folder_dict = {} def main(): @@ -40,15 +39,16 @@ def main(): global do_all global json_dict global url_dict + global folder_dict - # Using latin-1 encoding so that the raw bytes are accepted as they are. -## with codecs.open(args.Infile, encoding="latin-1") as json_data: -## json_dict = json.load(json_data) - with open(args.Infile) as json_data: - json_dict = json.load(json_data, encoding="latin-1") + with io.open(args.Infile, encoding='utf8', errors="replace") as json_data: + json_dict = json.load(json_data) term = NONE_SEARCH + if args.dump: + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term="", operation="printtree") + exit() while (1): match_cnt = 0 @@ -58,13 +58,20 @@ def main(): print (""" --------------------------------------------------------------------- Options: + p: Print the bookmarks tree hierarchy + s: Search term entry (currently <{}>) - t: List search matches - T: Delete search matches + t: List bookmark search matches + T: Delete bookmark search matches + g: List folder names search matches + G: Delete folder names search matches d: List duplicate URLs D: Delete duplicate URLs + f: List duplicate folders + F: Delete duplicate folders + x: List tags on folders (only leaf nodes should have tags) X: Delete tags on folders @@ -77,55 +84,74 @@ Options: select = prompt("Enter option: ") - if select == 's': + if select == 'p': + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term="", operation="printtree") + + elif select == 's': term = prompt("Search for (empty to clear search term): ").lower() if term == "": term = NONE_SEARCH elif select == 't': - digin(parent=json_dict["xBrowserSync"]["bookmarks"], parent_id="", path="/", search_term=term, operation="search") + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term=term, operation="search") print ("\nMatches: {}\n".format(match_cnt)) elif select == 'T': - digin(parent=json_dict["xBrowserSync"]["bookmarks"], parent_id="", path="/", search_term=term, operation="search", commit=True) + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term=term, operation="search", commit=True) + delete_items () # Do the queued deletes + print ("\nMatches: {} Deletes: {}\n".format(match_cnt, change_cnt)) + elif select == 'g': + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term=term, operation="SearchFolders") + print ("\nMatches: {}\n".format(match_cnt)) + elif select == 'G': + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term=term, operation="SearchFolders", commit=True) delete_items () # Do the queued deletes print ("\nMatches: {} Deletes: {}\n".format(match_cnt, change_cnt)) - elif select == 'd': - digin(parent=json_dict["xBrowserSync"]["bookmarks"], parent_id="", path="/", search_term="", operation="DupURLs") + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term="", operation="DupURLs") dup_urls() url_dict = {} print ("\nMatches: {}\n".format(match_cnt)) elif select == 'D': - digin(parent=json_dict["xBrowserSync"]["bookmarks"], parent_id="", path="/", search_term="", operation="DupURLs") + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term="", operation="DupURLs") dup_urls(commit=True) url_dict = {} delete_items () print ("\nMatches: {} Deletes: {}\n".format(match_cnt, change_cnt)) - - elif select == 'x': - digin(parent=json_dict["xBrowserSync"]["bookmarks"], parent_id="", path="/", search_term="", operation="FolderTags") + elif select == 'f': + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term="", operation="DupFolders") + dup_folders() + folder_dict = {} print ("\nMatches: {}\n".format(match_cnt)) - elif select == 'X': - digin(parent=json_dict["xBrowserSync"]["bookmarks"], parent_id="", path="/", search_term="", operation="FolderTags", commit=True) - print ("\nMatches: {} Deletes: {}\n".format(match_cnt, change_cnt)) - - - elif select == 'y': - digin(parent=json_dict["xBrowserSync"]["bookmarks"], parent_id="", path="/", search_term="", operation="EmptyFolders") - print ("\nMatches: {}\n".format(match_cnt)) - elif select == 'Y': - digin(parent=json_dict["xBrowserSync"]["bookmarks"], parent_id="", path="/", search_term="", operation="EmptyFolders", commit=True) + elif select == 'F': + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term="", operation="DupFolders") + dup_folders(commit=True) + folder_dict = {} delete_items () print ("\nMatches: {} Deletes: {}\n".format(match_cnt, change_cnt)) + elif select == 'x': + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term="", operation="FolderTags") + print ("\nMatches: {}\n".format(match_cnt)) + elif select == 'X': + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term="", operation="FolderTags", commit=True) + print ("\nMatches: {} Deletes: {}\n".format(match_cnt, change_cnt)) + + elif select == 'y': + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term="", operation="EmptyFolders") + print ("\nMatches: {}\n".format(match_cnt)) + elif select == 'Y': + digin(parent=json_dict["xbrowsersync"]["data"]["bookmarks"], parent_id="", path="/", search_term="", operation="EmptyFolders", commit=True) + delete_items () + print ("\nMatches: {} Deletes: {}\n".format(match_cnt, change_cnt)) elif select == 'w': ans = prompt("Output file name (default <{}>: ".format(args.Infile + OFILESUFFIX)) if ans == "": ans = args.Infile + OFILESUFFIX - with open(ans, 'w') as ofile: - json.dump(json_dict, ofile, indent=2) #, encoding="latin-1") + with io.open(ans, "w", encoding='utf8') as ofile: + # json.dump(json_dict, ofile, indent=2) #, encoding="latin-1") + json.dump(json_dict, ofile, ensure_ascii=False, indent=2) #, encoding="latin-1") elif select == 'q': exit() @@ -134,27 +160,52 @@ Options: -def digin(parent, parent_id, path, search_term, operation, commit=False): +def digin(parent, parent_id, path, search_term, operation, commit=False, indent=""): """Recurse through the json dictionary bookmark tree, with operation options.""" global match_cnt global change_cnt global do_all + global path_dict item_index = -1 ans = 'n' - item_list = [] - + for item in parent: item_index += 1 if "children" in item: # ***** Its a folder ***** + if operation == "printtree": + print ("{:<4} > {}{}".format(item["id"], indent, item["title"])) + + if operation == "SearchFolders": + if search_term in item["title"].lower(): + print ("{:<4} > {}{}".format(item["id"], indent, item["title"])) + match_cnt += 1 + if commit: + if do_all: + ans = 'y' + else: + ans = prompt("Confirm delete for this item ('y'es, 'a'll, or 'q'uit, default no) ").lower() + if ans == 'a': + do_all = True + ans = 'y' + if ans == 'y': + collect_items (path + parent_id, parent, item_index) + change_cnt += 1 + if ans == 'q': + if prompt("Discard pending deletes ('y'es, default no)? ").lower() == 'y': + path_dict = {} + return -1 + + + if operation == "FolderTags": if "tags" in item: print ("\n{:4} - {} >>> {}\n {}".format( item["id"], path[3:], - item["title"].encode('latin-1'), + item["title"], item["tags"])) if "url" in item: if item["url"] == None: @@ -181,7 +232,7 @@ def digin(parent, parent_id, path, search_term, operation, commit=False): print ("\n{:4} - {} >>> {}".format( item["id"], path, #[3:], - item["title"].encode('latin-1'))) + item["title"])) match_cnt += 1 if commit: if do_all: @@ -196,20 +247,28 @@ def digin(parent, parent_id, path, search_term, operation, commit=False): change_cnt += 1 if ans == 'q': if prompt("Discard pending deletes ('y'es, default no)? ").lower() == 'y': - path_dict = {} + path_dict = {} # TODO return -1 - + + if operation == "DupFolders": + log_folder (item["title"], path, item["id"], parent, parent_id, item_index) + + if digin ( # Recurse down to the next level parent = item["children"], - parent_id = str(item["id"]).encode("utf-8").decode("utf-8"), # WTF? https://stackoverflow.com/questions/17627834/convert-an-int-value-to-unicode + parent_id = str(item["id"]), path = path + " > " + item["title"], search_term = search_term, operation = operation, - commit = commit) == -1: + commit = commit, + indent = indent + " ") == -1: return -1 # recursion 'q'uit switch else: # ***** Its a Leaf node ***** + if operation == "printtree": + print ("{:<4} - {}{}".format(item["id"], indent, item["title"])) + if operation == "DupURLs": log_urls (item["url"], path, item["id"], parent, parent_id, item_index) @@ -228,7 +287,7 @@ def digin(parent, parent_id, path, search_term, operation, commit=False): print ("\n{:4} - {} >>> {}\n url: <{}>".format( item["id"], path[3:], - item["title"].encode('latin-1'), + item["title"], item["url"])) if "tags" in item: if item["tags"] == None: @@ -257,15 +316,15 @@ path_dict = {} def collect_items (path, parent, index): """Collect items for later deletion by delete_items.""" global path_dict -## print path -## print parent -## print index + # print (path) # "/ > [xbs] Toolbar > Chris' > Blogs166" + # print (parent) # List of dictionaries of bookmarks [{}, {}, {}] + # print (index) # Index within the list - 2 if path not in path_dict: path_dict[path] = {"parent":parent, "index":[int(index)]} else: + # else clause could happen for two identical bookmarks in the same folder, but normally they would have different ID#s path_dict[path]["index"].append(int(index)) - def delete_items (): """Delete queued-up items by collect_items. Note: Items in lists that are to be deleted must be deleted from the highest @@ -285,6 +344,44 @@ def delete_items (): path_dict = {} + +def log_folder (foldername, path, _id, parent, parent_id, item_index): + """Capture list of all instances of a folder name.""" + path_pid = path + " [" + parent_id + "]" + if foldername not in folder_dict: + folder_dict[foldername] = {"instance": [{"id":_id, "path":path_pid, "parent":parent, "item_index":item_index, "parent_id":parent_id}]} + else: + folder_dict[foldername]["instance"].append({"id":_id, "path":path_pid, "parent":parent, "item_index":item_index, "parent_id":parent_id}) + +def dup_folders (commit=False): + """List out, and optionally delete, duplicate folders.""" + global match_cnt + global change_cnt + global folder_dict + global path_dict + for folder in folder_dict: + if len(folder_dict[folder]["instance"]) > 1: + print ("-------------------------------------------------\n{}".format(folder)) + for instance in folder_dict[folder]["instance"]: + print (" {:>4} - {}".format(instance["id"], instance["path"])) + match_cnt += 1 + if commit: + print ("") + for instance in folder_dict[folder]["instance"]: + print (" {:>4} - {}".format(instance["id"], instance["path"])) + ans = prompt ("Confirm delete for this item ('y'es, 'q'uit, 's'kip to next URL - default no) ").lower() + if ans == 'y': + collect_items (instance["path"], instance["parent"], instance["item_index"]) + change_cnt += 1 + if ans == 's': + break + if ans == 'q': + if prompt("Discard pending deletes ('y'es, default no)? ").lower() == 'y': + path_dict = {} + return -1 + + + def log_urls (url, path, _id, parent, parent_id, item_index): """Capture a list of all instances of each URL.""" path_pid = path + " [" + parent_id + "]" @@ -293,7 +390,6 @@ def log_urls (url, path, _id, parent, parent_id, item_index): else: url_dict[url]["instance"].append({"id":_id, "path":path_pid, "parent":parent, "item_index":item_index, "parent_id":parent_id}) - def dup_urls (commit=False): """List out, and optionally delete, bookmarks with the same URL.""" global match_cnt @@ -329,59 +425,60 @@ def prompt (prompt_text): -def get_item (path): - """Lookup leaf node data given hierarchical path info. NOT USED, BUT POTENTIALLY INTERESTING. -Example passed-in path list structure: - [ "xBrowserSync", "bookmarks", 2, "ATV", 7, "Honda parts", 4 ] - A named item translates to "children" in the json_dict hierarchy. -Corresponds to this path in json_dict: - json_dict["xBrowserSync"]["bookmarks"][2]["children"][7]["children"][4] +# def get_item (path): +# """Lookup leaf node data given hierarchical path info. NOT USED, BUT POTENTIALLY INTERESTING. +# Example passed-in path list structure: +# [ "xbrowsersync", "data", "bookmarks", 2, "ATV", 7, "Honda parts", 4 ] +# A named item translates to "children" in the json_dict hierarchy. +# Corresponds to this path in json_dict: +# json_dict["xbrowsersync"]["data"]["bookmarks"][2]["children"][7]["children"][4] -Returns dictionary: - {"title" : "", - "url" : "<url text>", - "id" : id (int) - "tags" : [ "<tags>", "<list"> ], - "text_path: "<string concat of titles and list indexes>" - } -""" - global json_dict +# Returns dictionary: +# {"title" : "<title text>", +# "url" : "<url text>", +# "id" : id (int) +# "tags" : [ "<tags>", "<list"> ], +# "text_path: "<string concat of titles and list indexes>" +# } +# """ +# global json_dict - JOINTEXT = " > " +# JOINTEXT = " > " - xxx = json_dict["xBrowserSync"]["bookmarks"] - text_full_path = "" +# xxx = json_dict["xbrowsersync"]["data"]["bookmarks"] +# text_full_path = "" - for item in path: - print (item) - if type(item) is str: # Its the title of a folder (child) level - xxx = xxx["children"] - else: # Its a list index (integer) - text_full_path += JOINTEXT + xxx[item]["title"] - xxx = xxx[item] +# for item in path: +# print (item) +# if type(item) is str: # Its the title of a folder (child) level +# xxx = xxx["children"] +# else: # Its a list index (integer) +# text_full_path += JOINTEXT + xxx[item]["title"] +# xxx = xxx[item] - tags = "" - if "tags" in xxx: - tags = xxx["tags"] +# tags = "" +# if "tags" in xxx: +# tags = xxx["tags"] - return { - "title" : xxx["title"], - "url" : xxx["url"], - "id" : xxx["id"], - "tags" : tags, - "text_full_path": text_full_path - } +# return { +# "title" : xxx["title"], +# "url" : xxx["url"], +# "id" : xxx["id"], +# "tags" : tags, +# "text_full_path": text_full_path +# } if __name__ == '__main__': parser = argparse.ArgumentParser(description=desc, formatter_class=argparse.RawTextHelpFormatter) parser.add_argument('Infile', help="json backup file") + parser.add_argument('--dump', action='store_true', + help="Dump bookmark hierarchy (redirect to less or a file)") args = parser.parse_args() if not os.path.exists(args.Infile): print ("Can't find the input file <{}>".format(args.Infile)) exit() - main()