Merge pull request #138 from BCDA-APS/spec_newfile_128

BUG: SPEC filewriter scan numbering and newfile() when file exists

fixes #125 and #128

Author: prjemian

apstools/filewriters.py

@@ -338,7 +338,7 @@ def descriptor(self, doc):
         doc_hints_names = []
         for k, d in doc["hints"].items():
             doc_hints_names.append(k)
-            doc_hints_names += doc["hints"][k]["fields"]
+            doc_hints_names += d["fields"]
 
         # independent variable(s) first 
         # assumes start["motors"] was defined
@@ -528,22 +528,29 @@ def newfile(self, filename=None, scan_id=None, RE=None):
         self.clear()
         filename = filename or self.make_default_filename()
         if os.path.exists(filename):
-            ValueError(f"file {filename} exists")
+            from spec2nexus.spec import SpecDataFile
+            sdf = SpecDataFile(filename)
+            scan_list = sdf.getScanNumbers()
+            l = len(scan_list)
+            m = max(map(float, scan_list))
+            highest = int(max(l, m) + 0.9999)     # solves issue #128
+            scan_id = max(scan_id or 0, highest)
         self.spec_filename = filename
         self.spec_epoch = int(time.time())  # ! no roundup here!!!
         self.spec_host = socket.gethostname() or 'localhost'
         self.spec_user = getpass.getuser() or 'BlueSkyUser' 
         self.write_file_header = True       # don't write the file yet
 
         # backwards-compatibility
-        if scan_id == True:
-            scan_id = SCAN_ID_RESET_VALUE
-        elif scan_id == False:
-            scan_id = None
+        if isinstance(scan_id, bool):
+            # True means reset the scan ID to default
+            # False means do not modify it
+            scan_id = {True: SCAN_ID_RESET_VALUE, False: None}[scan_id]
         if scan_id is not None and RE is not None:
-            # assume isinstance(RE, bluesky.run_engine.RunEngine)
+            # RE is an instance of bluesky.run_engine.RunEngine
+            # (or duck type for testing)
             RE.md["scan_id"] = scan_id
-            print(f"scan ID set to {scan_id}")
+            self.scan_id = scan_id
         return self.spec_filename
 
     def usefile(self, filename):

conda-recipe/meta.yaml

@@ -43,6 +43,7 @@ requirements:
     - pyRestTable
     - pandas
     - xlrd
+    - spec2nexus
 
 test:
   imports:

packaging.md

@@ -0,0 +1,16 @@
+# Packaging Hints
+
+## PyPI upload
+
+Preceed the wildcard with tag text (`apstools-1.1.1*`)::
+
+	python setup.py sdist bdist_wheel
+	twine upload dist/*
+
+## Conda upload
+
+In the upload command below, use the text reported 
+at (near) the end of a successful conda build.
+
+	conda build ./conda-recipe/
+	anaconda upload /home/mintadmin/Apps/anaconda/conda-bld/noarch/apstools-1.1.1-py_0.tar.bz2

pypi.txt

@@ -5,3 +5,4 @@ ophyd
 pandas
 pyRestTable
 xlrd
+spec2nexus

requirements.txt

@@ -13,9 +13,11 @@
 def suite(*args, **kw):
 
     import test_simple
+    import test_filewriter
     # import test_excel
     test_list = [
         test_simple,
+        test_filewriter,
         # test_excel
         ]
 

tests/__main__.py

@@ -0,0 +1,282 @@
+
+"""
+unit tests for the SPEC filewriter
+"""
+
+import json
+import os
+import shutil
+import sys
+import tempfile
+import unittest
+import zipfile
+
+_test_path = os.path.dirname(__file__)
+_path = os.path.join(_test_path, '..')
+if _path not in sys.path:
+    sys.path.insert(0, _path)
+
+from apstools.filewriters import SpecWriterCallback
+
+
+ZIP_FILE = os.path.join(_test_path, "usaxs_docs.json.zip")
+JSON_FILE = "usaxs_docs.json.txt"
+
+
+def write_stream(specwriter, stream):
+    """write the doc stream to the file"""
+    for document in stream:
+        tag, doc = document
+        specwriter.receiver(tag, doc)
+
+
+def get_test_data():
+    """get document streams as dict from zip file"""
+    with zipfile.ZipFile(ZIP_FILE, "r") as fp:
+        buf = fp.read(JSON_FILE).decode("utf-8")
+        return json.loads(buf)
+
+
+class Test_Data_is_Readable(unittest.TestCase):
+    
+    def test_00_testdata_exist(self):
+        self.assertTrue(
+            os.path.exists(ZIP_FILE), 
+            "zip file with test data")
+        with zipfile.ZipFile(ZIP_FILE, "r") as fp:
+            self.assertIn(JSON_FILE, fp.namelist(), "JSON test data")
+        
+    def test_testfile_content(self):
+        # get our test document stream
+        datasets = get_test_data()
+        
+        census = {}
+        for document in datasets["tune_mr"]:
+            tag, _doc = document
+            if tag not in census:
+                census[tag] = 0
+            census[tag] += 1
+        
+        # test that tune_mr content arrived intact
+        keys = dict(start=1, descriptor=2, event=33, stop=1)
+        self.assertEqual(
+            len(census.keys()), 
+            len(keys), 
+            "four document types")
+        for k, v in keys.items():
+            self.assertIn(k, census, f"{k} document exists")
+            self.assertEqual(
+                census[k], 
+                v, 
+                f"expected {v} '{k}' document(s)")
+
+
+class Test_SpecWriterCallback(unittest.TestCase):
+
+    def setUp(self):
+        self.tempdir = tempfile.mkdtemp()
+        self.db = get_test_data()
+
+    def tearDown(self):
+        if os.path.exists(self.tempdir):
+            shutil.rmtree(self.tempdir, ignore_errors=True)
+    
+    def test_writer_default_name(self):
+        specwriter = SpecWriterCallback()
+        path = os.path.abspath(
+            os.path.dirname(
+                specwriter.spec_filename))
+        self.assertNotEqual(
+            path, 
+            self.tempdir, 
+            "default file not in tempdir")
+        self.assertEqual(
+            path, 
+            os.path.abspath(os.getcwd()), 
+            "default file to go in pwd")
+
+        # change the directory
+        specwriter.spec_filename = os.path.join(
+            self.tempdir, 
+            specwriter.spec_filename)
+
+        self.assertFalse(
+            os.path.exists(specwriter.spec_filename), 
+            "data file not created yet")
+        write_stream(specwriter, self.db["tune_mr"])
+        self.assertTrue(
+            os.path.exists(specwriter.spec_filename), 
+            "data file created")
+    
+    def test_writer_filename(self):
+        self.assertTrue(len(self.db) > 0, "test data ready")
+
+        testfile = os.path.join(self.tempdir, "tune_mr.dat")
+        if os.path.exists(testfile):
+            os.remove(testfile)
+        specwriter = SpecWriterCallback(filename=testfile)
+
+        self.assertIsInstance(
+            specwriter, SpecWriterCallback, 
+            "specwriter object")
+        self.assertEqual(
+            specwriter.spec_filename, 
+            testfile, 
+            "output data file")
+
+        self.assertFalse(
+            os.path.exists(testfile), 
+            "data file not created yet")
+        write_stream(specwriter, self.db["tune_mr"])
+        self.assertTrue(os.path.exists(testfile), "data file created")
+
+    def test_newfile_exists(self):
+        testfile = os.path.join(self.tempdir, "tune_mr.dat")
+        if os.path.exists(testfile):
+            os.remove(testfile)
+        specwriter = SpecWriterCallback(filename=testfile)
+
+        from apstools.filewriters import SCAN_ID_RESET_VALUE
+        self.assertEqual(SCAN_ID_RESET_VALUE, 0, "default reset scan id")
+
+        write_stream(specwriter, self.db["tune_mr"])
+        self.assertTrue(os.path.exists(testfile), "data file created")
+        
+        try:
+            specwriter.newfile(filename=testfile)
+            raised = False
+        except ValueError:
+            raised = True
+        finally:
+            self.assertFalse(raised, "file exists")
+        self.assertEqual(specwriter.reset_scan_id, 0, "check scan id")
+        
+        class my_RunEngine:
+            # dick type for testing _here_
+            md = dict(scan_id=SCAN_ID_RESET_VALUE)
+        RE = my_RunEngine()
+        
+        specwriter.scan_id = -5  # an unusual value for testing only
+        RE.md["scan_id"] = -10   # an unusual value for testing only
+        specwriter.newfile(filename=testfile, scan_id=None, RE=RE)
+        self.assertEqual(specwriter.scan_id, 108, "scan_id unchanged")
+        self.assertEqual(RE.md["scan_id"], 108, "RE.md['scan_id'] unchanged")
+
+        specwriter.scan_id = -5  # an unusual value for testing only
+        RE.md["scan_id"] = -10   # an unusual value for testing only
+        specwriter.newfile(filename=testfile, scan_id=False, RE=RE)
+        self.assertEqual(specwriter.scan_id, 108, "scan_id unchanged")
+        self.assertEqual(RE.md["scan_id"], 108, "RE.md['scan_id'] unchanged")
+
+        specwriter.scan_id = -5  # an unusual value for testing only
+        RE.md["scan_id"] = -10   # an unusual value for testing only
+        specwriter.newfile(filename=testfile, scan_id=True, RE=RE)
+        self.assertEqual(specwriter.scan_id, 108, "scan_id reset")
+        self.assertEqual(RE.md["scan_id"], 108, "RE.md['scan_id'] reset")
+
+        for n, s in {'0': 108, '108': 108, '110': 110}.items():
+            specwriter.scan_id = -5  # an unusual value for testing only
+            RE.md["scan_id"] = -10   # an unusual value for testing only
+            specwriter.newfile(filename=testfile, scan_id=int(n), RE=RE)
+            self.assertEqual(specwriter.scan_id, s, f"scan_id set to {n}, actually {s}")
+            self.assertEqual(RE.md["scan_id"], s, f"RE.md['scan_id'] set to {n}, actually {s}")
+    
+    def test__rebuild_scan_command(self):
+        from apstools.filewriters import _rebuild_scan_command
+
+        self.assertTrue(len(self.db) > 0, "test data ready")
+
+        start_docs = []
+        for header in self.db["tune_mr"]:
+            tag, doc = header
+            if tag == "start":
+                start_docs.append(doc)
+        self.assertEqual(len(start_docs), 1, "unique start doc found")
+
+        doc = start_docs[0]
+        expected = "108  tune_mr()"
+        result = _rebuild_scan_command(doc)
+        self.assertEqual(result, expected, "rebuilt #S line")
+    
+    def test_spec_comment(self):
+        from apstools.filewriters import spec_comment
+        
+        # spec_comment(comment, doc=None, writer=None)
+        testfile = os.path.join(self.tempdir, "spec_comment.dat")
+        if os.path.exists(testfile):
+            os.remove(testfile)
+        specwriter = SpecWriterCallback(filename=testfile)
+
+        for category in "buffered_comments comments".split():
+            for k in "start stop descriptor event".split():
+                o = getattr(specwriter, category)
+                self.assertEqual(len(o[k]), 0, f"no '{k}' {category}")
+        
+        # insert comments with every document
+        spec_comment(
+            "TESTING: Should appear within start doc", 
+            doc=None, 
+            writer=specwriter)
+
+        for idx, document in enumerate(self.db["tune_mr"]):
+            tag, doc = document
+            msg = f"TESTING: document {idx+1}: '{tag}' %s specwriter.receiver"
+            spec_comment(
+                msg % "before", 
+                doc=tag, 
+                writer=specwriter)
+            specwriter.receiver(tag, doc)
+            if tag == "stop":
+                # since stop doc was received, this appears in the next scan
+                spec_comment(
+                    str(msg % "before") + " (appears at END of next scan)", 
+                    doc=tag, 
+                    writer=specwriter)
+            else:
+                spec_comment(
+                    msg % "after", 
+                    doc=tag, 
+                    writer=specwriter)
+
+        self.assertEqual(
+            len(specwriter.buffered_comments['stop']), 
+            1, 
+            "last 'stop' comment buffered")
+        
+        # since stop doc was received, this appears in the next scan
+        spec_comment(
+            "TESTING: Appears at END of next scan", 
+            doc="stop", 
+            writer=specwriter)
+
+        self.assertEqual(
+            len(specwriter.buffered_comments['stop']), 
+            2, 
+            "last end of scan comment buffered")
+        write_stream(specwriter, self.db["tune_ar"])
+
+        for k in "start descriptor event".split():
+            o = specwriter.buffered_comments
+            self.assertEqual(len(o[k]), 0, f"no '{k}' {category}")
+        expected = dict(start=2, stop=5, event=0, descriptor=0)
+        for k, v in expected.items():
+            self.assertEqual(
+                len(specwriter.comments[k]), 
+                v, 
+                f"'{k}' comments")
+
+
+def suite(*args, **kw):
+    test_list = [
+        Test_Data_is_Readable,
+        Test_SpecWriterCallback,
+        ]
+    test_suite = unittest.TestSuite()
+    for test_case in test_list:
+        test_suite.addTest(unittest.makeSuite(test_case))
+    return test_suite
+
+
+if __name__ == "__main__":
+    runner=unittest.TextTestRunner()
+    runner.run(suite())