Update to Python3; Remove references to UF Sparse; Add Jupyter notebook support

Author: Sudarshan Raghunathan

PyUFGet/__main__.py

@@ -1,5 +1,5 @@
 import unittest
-import PyUFGet as p
+import ssget as p
 
 class TestPyUFGet(unittest.TestCase):
  def test_search_by_id(self):

PyUFGet/config.py

@@ -1,10 +1,16 @@
-#!/usr/bin/env python
-from distutils.core import setup
+from setuptools import setup
 
-setup(name="PyUFGet",
+setup(name="ssget",
  version="0.92",
- description="A Python interface to the University of Florida Sparse Matrix Collection",
+ description="A Python interface to the SparseSuite Matrix Collection",
  author="Sudarshan Raghunathan",
  author_email="rdarshan@gmail.com",
- url="http://www.github.com/drdarshan/PyUFGet",
- packages=["PyUFGet"])
+ url="http://www.github.com/drdarshan/ssget",
+ packages=["ssget"],
+ entry_points={
+ "console_scripts": [
+ "ssget = ssget.query:cli",
+ ],
+ },
+ install_requires=['tqdm']
+)

PyUFGet/query.py

@@ -1,23 +1,23 @@
 '''
-The PyUFGet module provides interfaces to search and download matrices from the
+The `ssget` module provides interfaces to search and download matrices from the
 University of Florida Sparse Matrix Collection.
 
-There are two ways to use PyUFGet:
-* By importing the PyUFGet module in Python, or
+There are two ways to use `ssget`:
+* By importing the `ssget` module in Python, or
 * As a standalone command-line tool
 
-To search for sparse matrices that match a given criterion, use `PyUFGet.search` ::
+To search for sparse matrices that match a given criterion, use `ssget.search` ::
 
- PyUFGet.search(name_or_id, **kwargs)
+ `ssget.search(name_or_id, **kwargs)`
 
-`PyUFGet.search` only returns a list of `Matrix` objects that match
+`ssget.search` only returns a list of `Matrix` objects that match
 the selection criterion. To download the matrices themselves, use the
-`download` method in the `Matrix object` or use `PyUFGet.fetch` ::
+`download` method in the `Matrix object` or use `ssget.fetch` ::
 
- PyUFGet.fetch(name_or_id, format, location, **kwargs)
+ `ssget.fetch(name_or_id, format, location, **kwargs)`
 
-The rules for specifying the search criteria in `PyUFGet.search` and
- `PyUFGet.fetch` are as follows:
+The rules for specifying the search criteria in `ssget.search` and
+ `ssget.fetch` are as follows:
 
 1. `name_or_id` can be either the numerical ID of the matrix such as
  `42` or a pattern such as `"HB/ash*"` or `"c-"`. This field is
@@ -39,17 +39,17 @@
 
 If `name_or_id` is specified, it overrides any conflicting key-value settings in `**kwargs`.
 
-In `PyUFGet.fetch`, `format` can be one of 'MM', 'MAT' or 'RB'; 'MM'
+In `ssget.fetch`, `format` can be one of 'MM', 'MAT' or 'RB'; 'MM'
 is the default if `format` is omitted. Finally, `location` refers
 to the directory where the matrices will be downloaded on the local
-machine. It defaults to `%APPDATA%/PyUFGet` on Windows and
-`~/.PyUFGet` on Unix-like platforms.
+machine. It defaults to `%APPDATA%/`ssget`` on Windows and
+`~/.ssget` on Unix-like platforms.
 
-In addition to its usage as a Python library, PyUFGet can be run from
+In addition to its usage as a Python library, `ssget` can be run from
 the command line as follows ::
 
- python PyUFGet
- Usage: PyUFGet [NameOrID] [options]
+ python `ssget`
+ Usage: `ssget` [NameOrID] [options]
 
  NameOrID A numerical matrix ID or a pattern matching the name of
  the matrix.
@@ -76,7 +76,7 @@
  -o LOCATION, --outdir=LOCATION
  The directory in the local machine where matrices will
  be downloaded to. Defaults to
- %AppData%\PyUFGet on Windows and ~/.PyUFGet on Unix.
+ `%AppData%\ssget` on Windows and `~/.ssget` on Unix.
  
  Size and Non-zero filters:
  These options may be used to restrict the shape or number of non-zero
@@ -97,5 +97,5 @@
  The maximum number of non-zero values in the
  matrix/matrices.
 '''
-from query import search, fetch, cli
+from .query import search, fetch, cli
 __all__ = ["query"]

Tests/PyUFGet_test.py renamed to Tests/ssget_test.py

@@ -0,0 +1,2 @@
+from .query import cli
+cli()

setup.py

@@ -1,12 +1,13 @@
+import gzip
+import os
+import tarfile
+import shutil
+
 def extract(bundle):
- import gzip
- import os
- import tarfile
  basedir, filename = os.path.split(bundle)
  tarfilename = os.path.join(basedir, '.'.join((filename.split('.')[0], 'tar')))
  gzfile = gzip.open(bundle, 'rb')
  with open(tarfilename, 'wb') as outtarfile:
- import shutil
  shutil.copyfileobj(gzfile, outtarfile)
  gzfile.close() 
  tarfile.open(tarfilename).extractall(basedir)

PyUFGet/__init__.py renamed to ssget/__init__.py

@@ -0,0 +1,26 @@
+import os
+import sys
+import logging
+
+logger = logging.getLogger(__name__)
+
+SS_DIR = None
+SS_DB = "index.db"
+SS_TABLE = "MATRICES"
+SS_ROOT_URL = "https://sparse.tamu.edu"
+SS_INDEX_URL = "/".join((SS_ROOT_URL, "files", "ssstats.csv"))
+
+if sys.platform == "win32":
+ SS_DIR = os.path.join(os.environ["APPDATA"], "ssget")
+else:
+ SS_DIR = os.path.join(os.environ["HOME"], ".ssget")
+
+SS_DB = os.path.join(SS_DIR, SS_DB)
+
+os.makedirs(SS_DIR, exist_ok=True)
+
+def dump():
+ logger.debug(dict(SS_DIR=SS_DIR, SS_DB=SS_DB, SS_TABLE=SS_TABLE, SS_ROOT_URL=SS_ROOT_URL, SS_INDEX_URL=SS_INDEX_URL))
+
+
+

ssget/__main__.py

@@ -1,19 +1,25 @@
 '''
-The `csvindex` module parses the UFStats.csv file and generate entries for each row in a Matrix database
+The `csvindex` module parses the SSStats.csv file and generate entries for each row in a Matrix database
 '''
 
+import csv
+import logging
+import requests
+from .config import SS_INDEX_URL
+
+logger = logging.getLogger(__name__)
+
 def getdtype(real, logical):
  '''
  Converts a (real, logical) pair into one of the three types, 'real', 'complex' and 'binary'
  '''
  return 'binary' if logical else ('real' if real else 'complex')
 
-def gen_rows(csvfile):
+def gen_rows(csvrows):
  '''
  Creates a generator that returns a single row in the matrix database.
  '''
- import csv
- reader = csv.reader(csvfile)
+ reader = csv.reader(csvrows)
  matid = 0
  for line in reader:
  matid += 1
@@ -32,14 +38,12 @@ def gen_rows(csvfile):
  yield matid, group, name, rows, cols, nnz, getdtype(real, logical), is2d3d, isspd, psym, nsym, kind
 
 def generate():
- from config import UF_INDEX_URL
- import urllib2
- csvfile = urllib2.urlopen(UF_INDEX_URL)
-
- import logging
+ response = requests.get(SS_INDEX_URL)
+ lines = response.iter_lines()
+ 
  # Read the number of entries
- logging.info("Number of entries in the CSV file: " + csvfile.readline())
+ logger.info(f'Number of entries in the CSV file: {next(lines)}')
  # Read the last modified date
- logging.info("Last modified date: " + csvfile.readline())
+ logger.info(f'Last modified date: {next(lines)}')
 
- return gen_rows(csvfile)
+ return gen_rows(line.decode('utf-8') for line in lines)