Python 3, read/write compressed json objects from/to gzip file

Question

For Python3, I followed @Martijn Pieters's code with this:

import gzip import json  # writing with gzip.GzipFile(jsonfilename, 'w') as fout:     for i in range(N):         uid = "whatever%i" % i         dv = [1, 2, 3]         data = json.dumps({             'what': uid,             'where': dv})          fout.write(data + '\n') 

but this results in an error:

Traceback (most recent call last):     ...   File "C:\Users\Think\my_json.py", line 118, in write_json     fout.write(data + '\n')   File "C:\Users\Think\Anaconda3\lib\gzip.py", line 258, in write     data = memoryview(data) TypeError: memoryview: a bytes-like object is required, not 'str' 

Any thoughts about what is going on?

Answer 1

You have four steps of transformation here.

1. a Python data structure (nested dicts, lists, strings, numbers, booleans)
2. a Python string containing a serialized representation of that data structure ("JSON")
3. a list of bytes containing a representation of that string ("UTF-8")
4. a list of bytes containing a - shorter - representation of that previous byte list ("gzip")

So let's take these steps one by one.

import gzip import json  data = [] for i in range(N):     uid = "whatever%i" % i     dv = [1, 2, 3]     data.append({         'what': uid,         'where': dv     })                                           # 1. data  json_str = json.dumps(data) + "\n"               # 2. string (i.e. JSON) json_bytes = json_str.encode('utf-8')            # 3. bytes (i.e. UTF-8)  with gzip.open(jsonfilename, 'w') as fout:       # 4. fewer bytes (i.e. gzip)     fout.write(json_bytes)                        

Note that adding "\n" is completely superfluous here. It does not break anything, but beyond that it has no use. I've added that only because you have it in your code sample.

Reading works exactly the other way around:

with gzip.open(jsonfilename, 'r') as fin:        # 4. gzip     json_bytes = fin.read()                      # 3. bytes (i.e. UTF-8)  json_str = json_bytes.decode('utf-8')            # 2. string (i.e. JSON) data = json.loads(json_str)                      # 1. data  print(data) 

Of course the steps can be combined:

with gzip.open(jsonfilename, 'w') as fout:     fout.write(json.dumps(data).encode('utf-8'))                        

and

with gzip.open(jsonfilename, 'r') as fin:     data = json.loads(fin.read().decode('utf-8')) 

Answer 2

The solution mentioned here (thanks, @Rafe) has a big advantage: as encoding is done on-the-fly, you don't create two complete, intermediate string objects of the generated json. With big objects, this saves memory.

with gzip.open(jsonfilename, 'wt', encoding='UTF-8') as zipfile:     json.dump(data, zipfile) 

In addition, reading and decoding is simple as well:

with gzip.open(jsonfilename, 'rt', encoding='UTF-8') as zipfile:     my_object = json.load(zipfile)