Support parsing for CP2K 2025 output format (#95)

* Fix regex pattern for SCF run convergence

* Update atomic forces regex patterns for parsing

* Fix regex pattern for energy extraction

* Add files via upload

* add test case for CP2K 2025 support

Author: yckbz

cp2kdata/block_parser/cells.py

@@ -95,7 +95,7 @@ def parse_all_md_cells(output_file: List[str],
     # notice that the cell of step 0 is excluded from MD| block
 
     # choose parser according to cp2k_info.version
-    if cp2k_info.version in ['9.1', '2022.2', '2023.1', '2023.2', '2024.1']:
+    if cp2k_info.version in ['9.1', '2022.2', '2023.1', '2023.2', '2024.1', '2024.2', '2025.1', '2025.2']:
         ALL_MD_CELL_RE = ALL_MD_CELL_RE_V2023
     elif cp2k_info.version in ['7.1']:
         ALL_MD_CELL_RE = ALL_MD_CELL_RE_V7

cp2kdata/block_parser/energies.py

@@ -3,9 +3,11 @@
 
 ENERGIES_RE = re.compile(
     r"""
-    \sENERGY\|\sTotal\sFORCE_EVAL\s\(\sQS\s\)\senergy\s\S{6}:\s+(?P<energy>[\s-]\d+\.\d+)
+    ^\s*ENERGY\|\s+Total\s+FORCE_EVAL\s+\(\s*QS\s*\)\s+energy
+    (?:\s+[\[\(][^\]\)]+[\]\)])?
+    \s*:?\s+(?P<energy>[-+]?\d+\.\d+(?:[Ee][+-]?\d+)?)
     """,
-    re.VERBOSE
+    re.VERBOSE | re.MULTILINE
 )
 
 

cp2kdata/block_parser/forces.py

@@ -19,6 +19,11 @@
     re.VERBOSE
 )
 
+FORCES_HEADER_RE = re.compile(
+    r"^\s*FORCES\|\s+Atomic\s+forces\s+\[.*\]\s*$",
+    re.MULTILINE
+)
+
 
 def parse_atomic_forces_list(output_file):
     atomic_forces_list = []
@@ -29,5 +34,33 @@ def parse_atomic_forces_list(output_file):
         atomic_forces_list.append(atomic_forces)
     if atomic_forces_list:
         return np.array(atomic_forces_list, dtype=float)
-    else:
-        return None
+
+    atomic_forces_list = []
+    lines = output_file.splitlines()
+    in_block = False
+    current = []
+    for line in lines:
+        if FORCES_HEADER_RE.match(line):
+            in_block = True
+            current = []
+            continue
+        if in_block:
+            if line.lstrip().startswith("FORCES|"):
+                parts = line.split()
+                if len(parts) >= 5 and parts[1].isdigit():
+                    current.append([parts[2], parts[3], parts[4]])
+                    continue
+                if parts[1] in ("Sum", "Total"):
+                    if current:
+                        atomic_forces_list.append(current)
+                    in_block = False
+            else:
+                if current:
+                    atomic_forces_list.append(current)
+                in_block = False
+
+    if in_block and current:
+        atomic_forces_list.append(current)
+    if atomic_forces_list:
+        return np.array(atomic_forces_list, dtype=float)
+    return None

tests/test_energy_force/v2025.1/normal/answer/answer.json

@@ -0,0 +1,42 @@
+{
+    "run_type": "ENERGY_FORCE",
+    "version": 2025.2,
+    "pot_energy": [
+        -75.63501801634357
+    ],
+    "chemical_symbols": [
+        "Si",
+        "Si",
+        "Si",
+        "O",
+        "O",
+        "O",
+        "O"
+    ],
+    "chemical_symbols_fake": [
+        "Si",
+        "Si",
+        "Si",
+        "O",
+        "O",
+        "O",
+        "O"
+    ],
+    "atomic_kind": [
+        "Si",
+        "O"
+    ],
+    "atom_num": [
+        3,
+        4
+    ],
+    "atom_kinds_list": [
+        1,
+        1,
+        1,
+        2,
+        2,
+        2,
+        2
+    ]
+}