Fix FASTG spec link; fix CI; small style fixes (#11)

* DOC: add internet archive link to FASTG spec

Fixes #9.

* DOC: tidy up link to archived specification

* BLD: pin CI Ubuntu version to 20.04

Fixes half of #10.

* STY/TST: fix flake8 line-length issues in test

this also made the code a lot cleaner, so probs for the best

* STY/BLD: lint/check setup.py; pin black < 22.10

.github/workflows/main.yml

@@ -4,16 +4,15 @@ name: pyfastg CI
 on: [push, pull_request]
 jobs:
   build:
-    runs-on: ubuntu-latest
+    # Stopgap solution -- "ubuntu-latest" (22.04, as of writing) no longer
+    # supports Python 3.6. See https://github.com/fedarko/pyfastg/issues/10.
+    runs-on: ubuntu-20.04
 
     strategy:
       matrix:
         # Gotta specify 3.10 as a string to avoid it being converted to 3.1:
         # https://github.com/actions/setup-python/issues/160
-        # The -dev after 3.11 is so that we can test on the latest dev version
-        # of python 3.11, since I believe no stable version of 3.11 exists yet
-        # (see https://github.com/actions/setup-python/issues/160 again)
-        python-version: [3.6, 3.7, 3.8, 3.9, "3.10", "3.11-dev"]
+        python-version: [3.6, 3.7, 3.8, 3.9, "3.10", "3.11"]
 
     steps:
 

Makefile

@@ -4,8 +4,8 @@ test:
 	python3 -B -m pytest pyfastg/tests/ --cov-report xml --cov-report term
 
 stylecheck:
-	flake8 pyfastg
-	black --check -l 79 pyfastg
+	flake8 pyfastg setup.py
+	black --check -l 79 pyfastg setup.py
 
 style:
-	black -l 79 pyfastg
+	black -l 79 pyfastg setup.py

README.md

@@ -14,7 +14,9 @@ accurately representing the ambiguity resulting from sequencing limitations, plo
 or other factors that complicate representation of a seqence as a simple string.
 
 The latest specification for the FASTG format is version 1.00, as of writing;
-this specification is located [here](http://fastg.sourceforge.net/FASTG_Spec_v1.00.pdf).
+the original FASTG website is down, but
+an archived version of the v1.00 specification is
+[accessible here](https://web.archive.org/web/20211209213905/http://fastg.sourceforge.net/FASTG_Spec_v1.00.pdf).
 Whenever the rest of this documentation mentions "the FASTG spec," this is in reference
 to this version of the specification.
 

pyfastg/tests/test_parse.py

@@ -3,6 +3,25 @@
 from pyfastg import parse_fastg
 
 
+# Represents the second line down of a simple FASTG file; we will add in the
+# "first line" of this file for a few tests later on below. (Adjusting this
+# first line programmatically is easier than writing out, like, five versions
+# of this one file.) For reference, a "correct" first line for this file is
+# >EDGE_1_length_9_cov_4.5:EDGE_3_length_5_cov_16.5',EDGE_2_length_3_cov_1;
+SIMPLE_LINES = """ATCGCCCAT
+>EDGE_1_length_9_cov_4.5':EDGE_2_length_3_cov_1';
+ATGGGCGAT
+>EDGE_2_length_3_cov_1:EDGE_1_length_9_cov_4.5;
+CGA
+>EDGE_2_length_3_cov_1';
+TCG
+>EDGE_3_length_5_cov_16.5:EDGE_1_length_9_cov_4.5',EDGE_2_length_3_cov_1';
+GGATC
+>EDGE_3_length_5_cov_16.5':EDGE_2_length_3_cov_1';
+GATCC
+"""
+
+
 def test_parse_medium_assembly_graph():
     digraph = parse_fastg("pyfastg/tests/input/assembly_graph.fastg")
     assert len(digraph.nodes) == 66
@@ -208,21 +227,11 @@ def write_tempfile(text):
 
 def test_parse_whitespace_btwn_adjs():
     # " , "
-    fh = write_tempfile(
-        """>EDGE_1_length_9_cov_4.5:EDGE_3_length_5_cov_16.5' , EDGE_2_length_3_cov_1;
-ATCGCCCAT
->EDGE_1_length_9_cov_4.5':EDGE_2_length_3_cov_1';
-ATGGGCGAT
->EDGE_2_length_3_cov_1:EDGE_1_length_9_cov_4.5;
-CGA
->EDGE_2_length_3_cov_1';
-TCG
->EDGE_3_length_5_cov_16.5:EDGE_1_length_9_cov_4.5',EDGE_2_length_3_cov_1';
-GGATC
->EDGE_3_length_5_cov_16.5':EDGE_2_length_3_cov_1';
-GATCC
-"""
-    )
+    t = (
+        ">EDGE_1_length_9_cov_4.5:EDGE_3_length_5_cov_16.5' , EDGE_2_length_"
+        "3_cov_1;\n"
+    ) + SIMPLE_LINES
+    fh = write_tempfile(t)
     with pytest.raises(ValueError) as ei:
         parse_fastg(fh.name)
     assert str(ei.value) == (
@@ -232,21 +241,11 @@ def test_parse_whitespace_btwn_adjs():
     )
 
     # " ,"
-    fh = write_tempfile(
-        """>EDGE_1_length_9_cov_4.5:EDGE_3_length_5_cov_16.5' ,EDGE_2_length_3_cov_1;
-ATCGCCCAT
->EDGE_1_length_9_cov_4.5':EDGE_2_length_3_cov_1';
-ATGGGCGAT
->EDGE_2_length_3_cov_1:EDGE_1_length_9_cov_4.5;
-CGA
->EDGE_2_length_3_cov_1';
-TCG
->EDGE_3_length_5_cov_16.5:EDGE_1_length_9_cov_4.5',EDGE_2_length_3_cov_1';
-GGATC
->EDGE_3_length_5_cov_16.5':EDGE_2_length_3_cov_1';
-GATCC
-"""
-    )
+    t = (
+        ">EDGE_1_length_9_cov_4.5:EDGE_3_length_5_cov_16.5' ,EDGE_2_length_"
+        "3_cov_1;\n"
+    ) + SIMPLE_LINES
+    fh = write_tempfile(t)
     with pytest.raises(ValueError) as ei:
         parse_fastg(fh.name)
     assert str(ei.value) == (
@@ -256,21 +255,11 @@ def test_parse_whitespace_btwn_adjs():
     )
 
     # ", "
-    fh = write_tempfile(
-        """>EDGE_1_length_9_cov_4.5:EDGE_3_length_5_cov_16.5', EDGE_2_length_3_cov_1;
-ATCGCCCAT
->EDGE_1_length_9_cov_4.5':EDGE_2_length_3_cov_1';
-ATGGGCGAT
->EDGE_2_length_3_cov_1:EDGE_1_length_9_cov_4.5;
-CGA
->EDGE_2_length_3_cov_1';
-TCG
->EDGE_3_length_5_cov_16.5:EDGE_1_length_9_cov_4.5',EDGE_2_length_3_cov_1';
-GGATC
->EDGE_3_length_5_cov_16.5':EDGE_2_length_3_cov_1';
-GATCC
-"""
-    )
+    t = (
+        ">EDGE_1_length_9_cov_4.5:EDGE_3_length_5_cov_16.5', EDGE_2_length_"
+        "3_cov_1;\n"
+    ) + SIMPLE_LINES
+    fh = write_tempfile(t)
     with pytest.raises(ValueError) as ei:
         parse_fastg(fh.name)
     assert str(ei.value) == (
@@ -282,21 +271,11 @@ def test_parse_whitespace_btwn_adjs():
 
 def test_parse_tilde_before_outgoing_adj_name():
     """These are technically supported in the spec, but we don't allow them."""
-    fh = write_tempfile(
-        """>EDGE_1_length_9_cov_4.5:EDGE_3_length_5_cov_16.5',~EDGE_2_length_3_cov_1;
-ATCGCCCAT
->EDGE_1_length_9_cov_4.5':EDGE_2_length_3_cov_1';
-ATGGGCGAT
->EDGE_2_length_3_cov_1:EDGE_1_length_9_cov_4.5;
-CGA
->EDGE_2_length_3_cov_1';
-TCG
->EDGE_3_length_5_cov_16.5:EDGE_1_length_9_cov_4.5',EDGE_2_length_3_cov_1';
-GGATC
->EDGE_3_length_5_cov_16.5':EDGE_2_length_3_cov_1';
-GATCC
-"""
-    )
+    t = (
+        ">EDGE_1_length_9_cov_4.5:EDGE_3_length_5_cov_16.5',~EDGE_2_length_"
+        "3_cov_1;\n"
+    ) + SIMPLE_LINES
+    fh = write_tempfile(t)
     with pytest.raises(ValueError) as ei:
         parse_fastg(fh.name)
     assert str(ei.value) == (
@@ -308,21 +287,11 @@ def test_parse_brackets_after_outgoing_adj_name():
     """Again, technically supported in the spec, but we don't allow this.
 
     This is a way of encoding adjacency-specific properties."""
-    fh = write_tempfile(
-        """>EDGE_1_length_9_cov_4.5:EDGE_3_length_5_cov_16.5'[prop=val],EDGE_2_length_3_cov_1;
-ATCGCCCAT
->EDGE_1_length_9_cov_4.5':EDGE_2_length_3_cov_1';
-ATGGGCGAT
->EDGE_2_length_3_cov_1:EDGE_1_length_9_cov_4.5;
-CGA
->EDGE_2_length_3_cov_1';
-TCG
->EDGE_3_length_5_cov_16.5:EDGE_1_length_9_cov_4.5',EDGE_2_length_3_cov_1';
-GGATC
->EDGE_3_length_5_cov_16.5':EDGE_2_length_3_cov_1';
-GATCC
-"""
-    )
+    t = (
+        ">EDGE_1_length_9_cov_4.5:EDGE_3_length_5_cov_16.5'[prop=val],EDGE_2_"
+        "length_3_cov_1;\n"
+    ) + SIMPLE_LINES
+    fh = write_tempfile(t)
     with pytest.raises(ValueError) as ei:
         parse_fastg(fh.name)
     assert str(ei.value) == (

setup.py

@@ -31,5 +31,9 @@
     classifiers=classifiers,
     packages=find_packages(),
     install_requires=["networkx >= 2"],
-    extras_require={"dev": ["pytest", "pytest-cov", "flake8", "black"]},
+    extras_require={
+        # The black < 22.10 pin is because black 22.10 stopped supporting
+        # python 3.6
+        "dev": ["pytest", "pytest-cov", "flake8", "black < 22.10"]
+    },
 )