save compressed data to MongoDB, can't be decompressed again about lz-string HOT 6 CLOSED

Just want to bump here that I had the same issue with MongoDB here and maybe it's related to those "invalid" UTF-16 characters. Switching compress to compressToUTF16 and same for decompress fixed this for me

from lz-string.

First, have a look at this (very similar) issue which could probably solve your problem: [https://github.com//issues/57]

lz-string is a hack that consists in storing binary data in a String. I only produce valid UTF-16 but maybe a text field in MongoDB is not accepting ALL characters.

The best way to troubleshoot is to compare the string you're storing to the string you're retrieving. My guess is that they are different in some way. Some UTF processing replaces characters with others deemed "equivalent". Of course, since lz-strnig uses it as a bitstream, it destroys the input.

from lz-string.

@pieroxy thanks

from lz-string.

It might be worth filing an issue with the mongo driver you're using, it's likely that the strings are being normalized (as @pieroxy mentioned) many characters can have multiple representations, and this may be a path taken to improve indexability.

Also, mongodb as of 3.0 supports compression, including zlib at the database level, so you should probably just store it from the buffer as a blob/binary... IIRC the node mongo interface should handle Buffer's

from lz-string.

thanks for your info @tracker1

from lz-string.

As the documentation states, compress and decompress produces invalid UTF-16 strings, so there is no guarantee a storage system will output its input unchanged.

from lz-string.