Child Item

Python TypeError using Dropbox API move_v2function

Hi, I am trying to use the API to batch rename file. I had specific questions are around the move_v2 function:

I’m running dropbox.files.move_v2(params), where the args are from_path and to_path – As I understand it, we can’t rename files specifically, rather we just use move/move_v2.

However, I keep getting the below error:

I know the function exists as a result, but clearly I’m not using it correctly.

Find more posts tagged with

API

Abuse

Comments

Greg-DB

Your image unfortunately is not loading. Can you try sharing that again?

ctubbs

Hi Greg, can you see this?

Greg-DB

Yes, thanks, that's visible now.

It looks like you're trying to call the Route object itself, but those aren't the methods for making API calls. To move or rename a file, you should use dropbox.Dropbox.files_move.

You can find some basic examples of making Dropbox API calls using the Dropbox Python SDK in this example, such as creating the client here, and making a call to start listing a folder here.

To use dropbox.Dropbox.files_move in particular, it would look like:

import dropbox

dbx = dropbox.Dropbox("ACCESS_TOKEN_HERE")

original_path = "/test.txt"
new_path = "/renamed test.txt"
result = dbx.files_move(from_path=original_path, to_path=new_path)

print(result)

LRomano

Hey Greg,

I tried to mod your code with some other snippets I've got so that I can rename all files within a root_file_path

file_path = '/root_path_here'

dbx.users_get_current_account()

result = dbx.files_list_folder(file_path, recursive=True)
file_list = []

def process_entries(entries):
    for entry in entries:
        if isinstance(entry, dropbox.files.FileMetadata):
            original = '/'+entry.name
            modified = '/test_'+entry.name
            result = dbx.files_move(from_path=original, to_path=modified)
            print(result)

process_entries(result.entries)


while result.has_more:
    result = dbx.files_list_folder_continue(result.cursor)

    process_entries(result.entries)

But am still ending up with errors.

Also, why do we assign the method to var: result and print?

Thanks

Luciano

Greg-DB

@LRomano What errors are you getting now?

Looking at your code, I expect the issue is that you're not specifying the correct file path for the file to move. The Metadata.name that you're using is just the name of the file, not the full path. To get the full path, including any parent path components, you would need to use Metadata.path_lower instead.

Also, I saved the result to a variable and printed it in my sample code just for demonstration purposes. You can do whatever you need in your actual code.

LRomano

I see, so you need the entire path for each file you iterate over.

In which case, if I'm looking to change all names of all files (which have a certain extension, say .mp4 or .png), is there a way to extract just the folder path so I can change the name of each file as a modification of its original?

Thanks

Edit:

Error was

process_entries(result.entries)

File "test.py", line 25, in process_entries

result = dbx.files_move(from_path=original, to_path=modified)

File "/opt/anaconda3/lib/python3.7/site-packages/dropbox/base.py", line 2398, in files_move

None,

File "/opt/anaconda3/lib/python3.7/site-packages/dropbox/dropbox.py", line 338, in request

user_message_locale)

dropbox.exceptions.ApiError: ApiError('ffb95a528b5ad351d54e7d1722f6baae', RelocationError('from_lookup', LookupError('not_found', None)))

Greg-DB

@LRomano Thanks for sharing the error. That 'from_lookup/not_found' error does indicate that the issue is that the 'from_path' value doesn't point to anything. That does seem to be because you specified just the name and not the path, as discussed.

The Metadata.path_lower value is just a string containing the path for the item, which might look something like: "/folder/another folder/file.ext".

Since that's just a string, you can operate on it however you need. For instance, you could split it on "/" and take the parts you want, modify them, and build a new path string. Or, you could use a library or module like pathlib to work with it.

LRomano

Fantastic. I used rsplit with '/' and to pull the lower_path apart giving me the file's directory and the file name.

Then I was able to add some text to the beginning of the file name, concat the whole thing together to create a new_path and it works perfectly!

Thanks so much for your help. Just FYI - code below if you want it (I had a spark DF with 1000s of rows of unique data that I needed to add to the beginning of each file):

result = dbx.files_list_folder(root_path, recursive=True)

index = 0

def process_entries(entries):
    for entry in entries:

      if isinstance(entry, dropbox.files.FileMetadata):
        path = entry.path_lower.rsplit('/', 1)[0]+'/'
        file = entry.path_lower.rsplit('/', 1)[1]
        path_original = path+file
        global index
        code = sparkDF.collect()[index][2]
        path_new = path+code+'_'+file
        print(path_original)
        print(path_new)
        move = dbx.files_move(from_path=path_original, to_path=path_new)
        index += 1


process_entries(result.entries)


while result.has_more:
    result = dbx.files_list_folder_continue(result.cursor)

    process_entries(result.entries)

LRomano

Hmmmm, actually, it seems to be running ok, but I've had to put in another if statement on has_more = TRUE

Even with the while statement, for some reason it was running the function twice. Any idea why?

result = dbx.files_list_folder(root_path, recursive=True)

full_list = []

index = 0

def process_entries(entries):
    for entry in entries:

      if result.has_more:
        if isinstance(entry, dropbox.files.FileMetadata):
          file = entry.path_lower.rsplit('/', 1)[1]
          if file.endswith(('.mp4','.png','.mov','.gif','.jpg')):
            path = entry.path_lower.rsplit('/', 1)[0]+'/'
            path_original = path+file
            global index
            code = sparkDF.collect()[index][2]
            file_new = code+'_'+file
            path_new = path+file_new
            sublist = [code,file_new,path_new]
            print(sublist)
            print(result.has_more)
            full_list.append(sublist)
            move = dbx.files_move(from_path=path_original, to_path=path_new)
            index += 1


process_entries(result.entries)


while result.has_more:
    result = dbx.files_list_folder_continue(result.cursor)

    process_entries(result.entries)

Greg-DB

I'm not sure off hand what the issue may be, but calling `if result.has_more` for each `entry` doesn't look like what you'd want to do. Once you're inspecting a particular entry, it shouldn't matter if there is another page of entries to retrieve (since you've already retrieved the entry you're operating on).

I recommend stepping through with a debugger to see where the code execution is going and why, and what the variables look like at any point in time.

Perhaps what you're seeing is the new entries created by your own move operations? You may need to implement some way to avoid operating on files that you've already moved, or maybe retrieve the full file listing before you start moving files, depending on exactly what you want your program to do and how.

LRomano

Hmmmmm, yes that makes sense actually, that the move operation would in a sense 'reset' the file in terms of where the cursor is, and thus might get included in a future operation.

Problem is the full file list is 160k total files across a number of directories, and even counting them all takes ~8 mins.

I think a workaround is to include an if statement to exclue any file that beginswith('str'), as all the strings that I'm appending to files begin with 'zas', but it's not exactly elegant.