From 502a734794efca4acca4cc3025a4455d5ed8355b Mon Sep 17 00:00:00 2001
From: "T.Tian" <alchem0x2a@gmail.com>
Date: Sun, 14 Jan 2024 16:56:42 +0800
Subject: [PATCH] add hard-coded parser items

---
 sparc/docparser.py          | 19 ++++++++++++++++++-
 sparc/sparc_parsers/aimd.py |  2 +-
 2 files changed, 19 insertions(+), 2 deletions(-)

diff --git a/sparc/docparser.py b/sparc/docparser.py
index ff1e1de8..d8df8c5b 100644
--- a/sparc/docparser.py
+++ b/sparc/docparser.py
@@ -16,6 +16,15 @@
 
 import numpy as np
 
+# Some fields in master SPARC doc may cause auto type detection
+# to fail, need hard-coded post-processing for now
+postprocess_items = {
+    "RELAX_FLAG": {"allow_bool_input": False},
+    "NPT_SCALE_CONSTRAINTS": {"type": "string"},
+    "NPT_SCALE_VECS": {"type": "integer array"},
+    "TOL_POISSON": {"type": "double"},
+}
+
 
 class SPARCDocParser(object):
     """Use regex to parse LaTeX doc to python API"""
@@ -55,6 +64,7 @@ def __init__(
         self.params_from_intro = params_from_intro
         self.parse_version(parse_version)
         self.parse_parameters()
+        self.postprocess()
 
     def find_main_file(self, main_file_pattern):
         """Find the matching name for the main-file, e.g. Manual.tex or Manual_cyclix.tex"""
@@ -251,6 +261,13 @@ def parse_parameters(self):
 
         return
 
+    def postprocess(self):
+        """Use the hardcoded parameter correction dict to fix some issues"""
+        for param, fix in postprocess_items.items():
+            if param in self.parameters:
+                self.parameters[param].update(**fix)
+        return
+
     def to_dict(self):
         """Output a json string from current document parser
 
@@ -264,7 +281,7 @@ def to_dict(self):
         doc["other_parameters"] = {
             k: v for k, v in sorted(self.other_parameters.items())
         }
-        doc["data_types"] = list(set([p["type"] for p in self.parameters.values()]))
+        doc["data_types"] = sorted(set([p["type"] for p in self.parameters.values()]))
         # json_string = json.dumps(doc, indent=indent)
         return doc
 
diff --git a/sparc/sparc_parsers/aimd.py b/sparc/sparc_parsers/aimd.py
index 78529ca6..d604320f 100644
--- a/sparc/sparc_parsers/aimd.py
+++ b/sparc/sparc_parsers/aimd.py
@@ -137,7 +137,7 @@ def _read_aimd_step(raw_aimd_text):
                 value = raw_value * GPa
             elif dim == 2:
                 name = "stress_2d"
-                value = raw_value * Hartree / Bohr ** 2
+                value = raw_value * Hartree / Bohr**2
             elif dim == 1:
                 name = "stress_1d"
                 value = raw_value * Hartree / Bohr