Skip to content

Commit

Permalink
wip new msgspec decoder
Browse files Browse the repository at this point in the history
  • Loading branch information
pwrose committed Jul 6, 2022
1 parent fd7bf43 commit afa4799
Showing 1 changed file with 20 additions and 4 deletions.
24 changes: 20 additions & 4 deletions mmtfPyspark/io/mmtfReader.py
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,10 @@
import urllib
import urllib.request as urllib2
import pandas as pd
# -pr
# -pr import pandas as pd
import msgspec
# -pr

text = "org.apache.hadoop.io.Text"
byteWritable = "org.apache.hadoop.io.BytesWritable"
Expand Down Expand Up @@ -230,7 +234,10 @@ def _get_structure(pdbId, reduced, first_model):
data = gzip.decompress(response.read())
else:
data = response.read()
unpack = pd.read_msgpack(data)
# -pr unpack = pd.read_msgpack(data)
dc_msg = msgspec.json.Decoder()
unpack = dc_msg.decode(data)
# -pr
decoder = MmtfStructure(unpack, first_model)
return (pdbId, decoder)
except urllib.error.HTTPError:
Expand All @@ -245,7 +252,10 @@ def _call_sequence_file(t, first_model):
# decoder = MmtfStructure(unpack)
# return (str(t[0]), decoder)
data = gzip.decompress(t[1])
unpack = pd.read_msgpack(data)
# -pr unpack = pd.read_msgpack(data)
dc_msg = msgspec.json.Decoder()
unpack = dc_msg.decode(data)
# -pr
decoder = MmtfStructure(unpack, first_model)
return (t[0], decoder)

Expand All @@ -257,7 +267,10 @@ def _call_mmtf(f, first_model=False):
name = f.split('/')[-1].split('.')[0].upper()
data = gzip.open(f, 'rb')
#unpack = msgpack.unpack(data, raw=False)
unpack = pd.read_msgpack(data)
# -pr unpack = pd.read_msgpack(data)
dc_msg = msgspec.json.Decoder()
unpack = dc_msg.decode(data)
# -pr
decoder = MmtfStructure(unpack, first_model)
return (name, decoder)

Expand All @@ -266,7 +279,10 @@ def _call_mmtf(f, first_model=False):
#unpack = msgpack.unpack(open(f, "rb"), raw=False)
#decoder = MmtfStructure(unpack)
name = f.split('/')[-1].split('.')[0].upper()
unpack = pd.read_msgpack(f)
# -prunpack = pd.read_msgpack(f)
dc_msg = msgspec.json.Decoder()
unpack = dc_msg.decode(f)
# -pr
decoder = MmtfStructure(unpack, first_model)
return (name, decoder)

Expand Down

0 comments on commit afa4799

Please sign in to comment.