From 3dae72ae5a77d0bbd9b549b0024c2c83e6a11aec Mon Sep 17 00:00:00 2001
From: Zhiyi Wu <zwu@exscientia.ai>
Date: Thu, 22 Jun 2023 13:28:45 +0100
Subject: [PATCH] ValueError will be raised if concatenated amber output file
 has been ingested (#326)

* update

* update

* update

---------

Co-authored-by: Zhiyi Wu <zwu@exscientia.co.uk>
---
 CHANGES                                   |  2 ++
 src/alchemlyb/parsing/amber.py            |  9 ++++++-
 src/alchemlyb/tests/parsing/test_amber.py | 33 +++++++++++++++++++++++
 3 files changed, 43 insertions(+), 1 deletion(-)

diff --git a/CHANGES b/CHANGES
index c3ed0b3e..37d88468 100644
--- a/CHANGES
+++ b/CHANGES
@@ -18,6 +18,8 @@ The rules for this file:
   * 2.1.0
 
 Changes
+  - ValueError raised if concatenated amber output file is passed to amber
+    parser (issue #315, PR #326).
   - Change the % based string formatting to {} based string formatting (issue #323, PR #324).
   - Use loguru instead of logging for log (issue #301, PR #303).
 
diff --git a/src/alchemlyb/parsing/amber.py b/src/alchemlyb/parsing/amber.py
index 543d4d54..6bb5441f 100644
--- a/src/alchemlyb/parsing/amber.py
+++ b/src/alchemlyb/parsing/amber.py
@@ -332,10 +332,18 @@ def extract(outfile, T):
                     "^ NSTEP", "^ ---", ["NSTEP", "DV/DL"], extra=line
                 )
                 if nstep != old_nstep and dvdl is not None and nstep is not None:
+                    if finished:
+                        raise ValueError(
+                            "TI Energy detected after the TIMINGS section. Did you concatenate the output file?"
+                        )
                     file_datum.gradients.append(dvdl)
                     nensec += 1
                     old_nstep = nstep
             elif line.startswith("MBAR Energy analysis") and file_datum.have_mbar:
+                if finished:
+                    raise ValueError(
+                        "MBAR Energy detected after the TIMINGS section. Did you concatenate the output file?"
+                    )
                 mbar = secp.extract_section(
                     "^MBAR", "^ ---", file_datum.mbar_lambdas, extra=line
                 )
@@ -356,7 +364,6 @@ def extract(outfile, T):
                     )
             elif line == "   5.  TIMINGS\n":
                 finished = True
-                break
 
         if high_E_cnt:
             logger.warning(
diff --git a/src/alchemlyb/tests/parsing/test_amber.py b/src/alchemlyb/tests/parsing/test_amber.py
index 0d186cc7..21b6064e 100644
--- a/src/alchemlyb/tests/parsing/test_amber.py
+++ b/src/alchemlyb/tests/parsing/test_amber.py
@@ -1,6 +1,7 @@
 """Amber parser tests.
 
 """
+import bz2
 import logging
 
 import pandas as pd
@@ -250,3 +251,35 @@ def test_u_nk_improper(improper_filename, names=("time", "lambdas")):
         assert u_nk.index.names == names
     except Exception:
         assert "0.5626" in improper_filename
+
+
+def test_concatenated_amber_u_nk(tmp_path):
+    with bz2.open(load_bace_example()["data"]["complex"]["decharge"][0], "rt") as file:
+        content = file.read()
+
+    with open(tmp_path / "amber.out", "w") as f:
+        f.write(content)
+        f.write("\n")
+        f.write(content)
+
+    with pytest.raises(
+        ValueError,
+        match="MBAR Energy detected after the TIMINGS section.",
+    ):
+        extract(tmp_path / "amber.out", 298)
+
+
+def test_concatenated_amber_dhdl(tmp_path):
+    with bz2.open(load_bace_example()["data"]["complex"]["decharge"][0], "rt") as file:
+        content = file.read().replace("MBAR Energy analysis", "")
+
+    with open(tmp_path / "amber.out", "w") as f:
+        f.write(content)
+        f.write("\n")
+        f.write(content)
+
+    with pytest.raises(
+        ValueError,
+        match="TI Energy detected after the TIMINGS section.",
+    ):
+        extract(tmp_path / "amber.out", 298)