Read a sections file from file: text, Python text, Python module.

goatools/grouper/read_goids.py

@@ -5,6 +5,8 @@
 import os
 import sys
 import re
+import pkgutil
+import importlib
 from goatools.gosubdag.go_tasks import chk_goids
 from goatools.grouper.hdrgos import HdrgosSections
 from goatools.grouper.grprobj import Grouper
@@ -18,8 +20,18 @@ def read_sections(sections_file, exclude_ungrouped=False, prt=sys.stdout):
     """Get sections and GO grouping hdrgos from file, if sections exist."""
     if sections_file is None:
         return None
+    assert isinstance(sections_file, str), "BAD SECTIONS FILENAME({S})".format(
+        S=sections_file)
     if os.path.exists(sections_file):
         return ReadGoids().read_sections(sections_file, False, exclude_ungrouped)
+    # Is 'sections_file' a module string?
+    if '/' not in sections_file and r'\\' not in sections_file and \
+            pkgutil.find_loader(sections_file) is not None:
+        mod = importlib.import_module(sections_file)
+        var = getattr(mod, 'SECTIONS', None)
+        if var is not None:
+            return var
+        raise RuntimeError("NO 'SECTIONS' VARIABLE FOUND IN MODULE({M})".format(M=sections_file))
     if prt:
         prt.write("CANNOT READ: {SEC}\n".format(SEC=sections_file))
 
@@ -33,6 +45,13 @@ class ReadGoids(object):
     """Get user list of GO IDs either from a list or from GO IDs on the command-line"""
 
     srch_section = re.compile(r'^#?\s*SECTION:\s*(\S.*\S)\s*$', flags=re.IGNORECASE)
+    # For reading SECTIONS from a Python file as text (without importing)
+    #   ("cell death", [ # 6 GO-headers
+    #   ("viral/bacteria", [ # 4 GO-headers
+    srch_py_section = re.compile(r'^\s*(\(|\[)\s*(\'|")(.*)\s*(\'|")\s*,\s*\[')
+    #       "GO:0002376", # BP    564 L01 D01 M     immune system process
+    #       "GO:0002682", # BP  1,183 L02 D02 AB    regulation of immune system process
+    srch_py_goids = re.compile(r'^\s*(\'|")(GO:\d{7})(\'|")\s*,?')
 
     def __init__(self):
         self.goids_fin = []
@@ -50,7 +69,7 @@ def read_txt(self, fin_txt, get_goids_only, exclude_ungrouped, prt=sys.stdout):
 
     def read_py(self, fin_txt, get_goids_only, exclude_ungrouped, prt=sys.stdout):
         """Read GO IDs or sections data from a Python file."""
-        goids_fin = self._read_txt(fin_txt, get_goids_only, exclude_ungrouped)
+        goids_fin = self._read_py(fin_txt, get_goids_only, exclude_ungrouped)
         sections = self._read_finish(goids_fin, prt)
         # Print summary of GO IDs read
         if prt is not None:
@@ -59,9 +78,9 @@ def read_py(self, fin_txt, get_goids_only, exclude_ungrouped, prt=sys.stdout):
 
     def read_sections(self, sections_file, get_goids_only, exclude_ungrouped):
         """Read sections variable from a text file of from a Python file."""
-        ext = os.path.splitext(sections_file)
+        ext = os.path.splitext(sections_file)[1]
         file_contents = None
-        if ext and ext == ".py":
+        if ext == '.py':
             file_contents = self.read_py(sections_file, get_goids_only, exclude_ungrouped)
         else:
             file_contents = self.read_txt(sections_file, get_goids_only, exclude_ungrouped)
@@ -110,6 +129,31 @@ def _read_txt(self, fin_txt, get_goids_only, exclude_ungrouped):
                 self.section2goids[section_name] = goids_sec
         return goids_sec
 
+    def _read_py(self, fin_py, get_goids_only, exclude_ungrouped):
+        """Read Python sections file. Store: section2goids sections_seen. Return goids_fin."""
+        goids_sec = []
+        with open(fin_py) as istrm:
+            section_name = None
+            for line in istrm:
+                mgo = self.srch_py_goids.search(line)     # Matches GO IDs in sections
+                if mgo:
+                    goids_sec.append(mgo.group(2))
+                elif not get_goids_only and "[" in line:
+                    msec = self.srch_py_section.search(line)  # Matches sections
+                    if msec:
+                        secstr = msec.group(3)
+                        if section_name is not None and goids_sec:
+                            self.section2goids[section_name] = goids_sec
+                        if not exclude_ungrouped or secstr != HdrgosSections.secdflt:
+                            section_name = secstr
+                            self.sections_seen.append(section_name)
+                        else:
+                            section_name = None
+                        goids_sec = []
+            if section_name is not None and goids_sec:
+                self.section2goids[section_name] = goids_sec
+        return goids_sec
+
     def _rpt_unused_sections(self, prt):
         """Report unused sections."""
         sections_unused = set(self.sections_seen).difference(self.section2goids.keys())

tests/test_wr_sections_txt.py

@@ -23,33 +23,45 @@ def test_wr_sections_txt():
     # Print usrgos in txt (Do not use sections containing hdrgos)
     # ------------------------------------------------------------------
     # Show GO grouping hdrgos and usrgos to show how usrgos are grouped
-    _wr_sections_txt("a0_hdr1.txt", usrgos, sections_file=None, grprdflt=grprdflt)
+    _wr_sections_txt("a0_hdr1.txt", usrgos, sections=None, grprdflt=grprdflt)
 
     # ------------------------------------------------------------------
     # Print usrgos in txt using sections containing hdrgos
     # ------------------------------------------------------------------
     sec1 = _read_sections("./data/gjoneska/sections_in.txt")
-    # Print usrgos in sections, showing how they were grouped under hdrgos
     _wr_sections_txt("a_ec0_hdr1.txt", usrgos, sec1, grprdflt=grprdflt)
 
     # ------------------------------------------------------------------
-    # Print usrgos in txt using sections containing hdrgos
-    # ------------------------------------------------------------------
-    # sec2 = _read_sections("goatools/test_data/sections/gjoneska_pfenning.py")
-    # # Print usrgos in sections, showing how they were grouped under hdrgos
-    # _wr_sections_txt("b_sec0_hdr1.txt", usrgos, sec2, grprdflt=grprdflt)
+    sec2a = _read_sections("goatools/test_data/sections/gjoneska_pfenning.py")
+    _wr_sections_txt("b_sec0_hdr1.txt", usrgos, sec2a, grprdflt=grprdflt)
+
+    sec2b = _read_sections("goatools.test_data.sections.gjoneska_pfenning")
+    _wr_sections_txt("c_sec0_hdr1.txt", usrgos, sec2b, grprdflt=grprdflt)
+    # print("@@@@@@@@@ SECTIONS READ AS TEXT", sec1)
+    # print("@@@@@@@@@ SECTIONS READ AS TEXT", sec2)
+    # print("@@@@@@@@@ SECTIONS READ AS TEXT", sec3)
+    _chk_sections(sec2a, sec2b)
+
 
+def _chk_sections(sec_a, sec_b):
+    """Do the two sections variables contain the same data?"""
+    assert len(sec_a) == len(sec_b), "LENGTH MISMATCH: {A} != {B}".format(
+        A=len(sec_a), B=len(sec_b))
+    for (name_a, gos_a), (name_b, gos_b) in zip(sec_a, sec_b):
+        assert name_a == name_b, "NAME MISMATCH: {A} != {B}".format(A=name_a, B=name_b)
+        assert gos_a == gos_b, "GO IDs MISMATCH: {A} != {B}".format(A=gos_a, B=gos_b)
 
-def _read_sections(fin):
+def _read_sections(sec):
     """Get sections variable from file."""
-    sec = os.path.join(REPO, fin)
-    assert read_sections(sec), "EMPTY SECTIONS FILE({})".format(sec)
-    return sec
+    if '/' in sec:
+        sec = os.path.join(REPO, sec)
+    var = read_sections(sec)
+    assert var, "EMPTY SECTIONS FILE({})".format(sec)
+    return var
 
-def _wr_sections_txt(fout_txt, usrgos, sections_file, grprdflt):
+def _wr_sections_txt(fout_txt, usrgos, sections, grprdflt):
     """Given a list of usrgos and sections, write text file."""
     try:
-        sections = read_sections(sections_file)
         hdrobj = HdrgosSections(grprdflt.gosubdag, grprdflt.hdrgos_dflt, sections=sections)
         grprobj = Grouper(fout_txt, usrgos, hdrobj, grprdflt.gosubdag, go2nt=None)
         full_txt = os.path.join(REPO, fout_txt)