lxyu / kindle-clippings Goto Github PK

Highlights have data like "Location 1513-1514" but notes can look like "Location 1514" so these are not picked up by the script. Maybe a regex like 'Location (\d+)' instead of '(\d+)-\d+'

I had a lot of books that had a ":" character in the title

Can be fixed by changing line 51 to:
filename = os.path.join(OUTPUT_DIR, u"%s.txt" % book.replace(":", ""))

The script works great, but there's no license for it. Any plan to provide it?

I know msgpack is great, but I don't see any significant benefits in this project. 😶
Isn't that great if this project comes with no external dependency? 😃

I have the same problem described here
http://stackoverflow.com/questions/22620965/ioerror-errno-22-invalid-mode-wb-or-filename

I was wondering whether the filename could be normalized, at least under windows, before creating the file with the filename of the title.

For the moment I'm importing unicodedata and using the following verbose code (and not immune to errors, as I allow the \ to be in the filename, as it is used for referring to the partial path) in export_txt to make things work:

        try:
            with open(filename, 'wb') as f:
                f.write("\n\n---\n\n".join(lines))
        except:
            oldfilename = filename
            filename = unicodedata.normalize('NFKD', filename).encode('ascii','ignore')
            valid_chars = '\\-_.() abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789'
            filename = ''.join(c for c in filename if c in valid_chars)
            print 'Converting', oldfilename, 'to', filename
            with open(filename, 'wb') as f:
                f.write("\n\n---\n\n".join(lines))

Thanks

For each book, a dict X is created using a string of the clip position as key.
Then clips are written for each pos in sorted(X), so that e.g. "3748" comes before "410".
For the moment I'm using an inefficient:

for pos in sorted([int(x) for x in clips[book].keys()]):

in export_txt.

Documents can have titles split with a newline, currently the script assumes the title is a single line so it does not find the Location in the expected line. Maybe get_sections can remove the newlines in the title.

It might be worth showing an error when a Location can't be found - this situation should not exist ?

Looks like this repo has fallen out of maintenance. Could you please hand over maintenance to someone else? If no one else is enthusiastic about doing it, I could.

The script doesn't process the whole file. Output folder has only few notes.
My Clippings.txt

To allow the script to be installed on a path it might be nice to create the output dir if it does not exist (rather than assuming the script is run from within the repository only). Thanks for making this available!

错误代码如下：
Traceback (most recent call last):
File "kindle.py", line 94, in
main()
File "kindle.py", line 90, in main
export_txt(clips)
File "kindle.py", line 50, in export_txt
with open(filename, 'w') as f:
IOError: [Errno 2] No such file or directory: u'output/Who Moved My Cheese? (Spencer Johnson).md'