servalcat 0.4.131__cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl

servalcat/refmac/refmac_keywords.py

@@ -0,0 +1,660 @@
+"""
+Author: "Keitaro Yamashita, Garib N. Murshudov"
+MRC Laboratory of Molecular Biology
+    
+This software is released under the
+Mozilla Public License, version 2.0; see LICENSE.
+"""
+from __future__ import absolute_import, division, print_function, generators
+from servalcat.utils import logger
+from servalcat.utils import model as model_util
+import gemmi
+b_to_u = model_util.b_to_u
+
+def parse_atom_spec(s, itk):
+    # s: list of keywords
+    ret = {}
+    while itk < len(s):
+        if s[itk].lower().startswith(("chai", "segm")):
+            ret["chain"] = s[itk+1]
+            itk += 2
+        elif s[itk].lower().startswith("resi"):
+            ret["resi"] = int(s[itk+1])
+            itk += 2
+        elif s[itk].lower().startswith("ins"):
+            ret["icode"] = s[itk+1] if s[itk+1] != "." else " "
+            itk += 2
+        elif s[itk].lower().startswith(("atom", "atna", "name")):
+            if s[itk+1] == "{":
+                idx_close = s[itk+1:].index("}") + itk + 1
+                ret["names"] = s[itk+2:idx_close]
+                itk = idx_close + 1
+            else:
+                ret["names"] = [s[itk+1]]
+                itk += 2
+        elif s[itk].lower().startswith("alt"):
+            ret["altloc"] = s[itk+1]
+            itk += 2
+        elif s[itk].lower().startswith("symm"):
+            ret["symm"] = s[itk+1][0].lower() == "y"
+            itk += 2
+        else:
+            break
+
+    return ret, itk
+# parse_atom_spec()
+
+def parse_from_to(s, itk):
+    # s: list of keywords
+    ret = {}
+    if not s[itk].lower().startswith("from"):
+        raise RuntimeError("invalid from_to instruction: {}".format(s))
+
+    if s[itk+1] == "*":
+        ret["ifirst"] = None # Refmac sets -9999
+    else:
+        ret["ifirst"] = int(s[itk+1])
+
+    if s[itk+2].lower() != "to":
+        ret["chain"] = s[itk+2]
+        assert s[itk+3].lower() == "to"
+        itk += 4
+    else:
+        itk += 3
+
+    if s[itk] == "*":
+        ret["ilast"] = None # Refmac sets 9999
+    else:
+        ret["ilast"] = int(s[itk])
+
+    if "chain" not in ret:
+        ret["chain"] = s[itk+1]
+        itk += 2
+    else:
+        itk += 2
+        
+    return ret, itk
+# parse_from_to()
+
+def read_exte(s, raise_unknown=True):
+    # using the same variable names as used in read_extra_restraints.f
+    ret = dict(defaults={})
+    if not s: return ret
+    defs = ret["defaults"]
+    rest_flag_old = rest_flag = False
+    if s[0].lower().startswith("exte"):
+        if s[1].lower().startswith("gene"): return ret
+        elif s[1].lower().startswith("file"): # XXX not supported
+            file_ext_now = s[2]
+        elif s[1].lower().startswith("syma"): # symall
+            # refmac sets "n" by default if "syma" given - but it is not good!
+            defs["symall_block"] = "y" if s[2][0].lower() == "y" else "n"
+            if len(s) > 3 and s[3].lower().startswith("excl"): # exclude
+                defs["exclude_self_block"] = s[4].lower().startswith("self")
+        elif s[1].lower().startswith("typa"): # typeall
+            #type_default = 2
+            defs["type_default"] = max(0, min(2, int(s[2])))
+        elif s[1].lower().startswith("alph"): # alphall
+            defs["alpha_default"] = float(s[2])
+        elif s[1].lower().startswith("verb"): # verbose
+            defs["ext_verbose"] = s[2][0].lower() != "n" and not s[2].lower().startswith("off")
+            # print("External verbose is on, i.e.") ...
+        elif s[1].lower().startswith("weig"): # weight
+            itk = 2
+            while itk < len(s): # FIXME check out-of-bounds in s[]
+                if s[itk].lower().startswith("scal"):
+                    itk += 1
+                    if itk >= len(s): break
+                    try:
+                        defs["scale_sigma_dist"] = float(s[itk]) # scale_sigma_loc
+                        itk += 1
+                        continue
+                    except ValueError:
+                        pass
+                    for k in ("angl", "tors", "chir", "plan", "dist", "inte"):
+                        if s[itk].lower().startswith(k):
+                            itk += 1
+                            if itk >= len(s): break
+                            try:
+                                defs["scale_sigma_{}".format(k)] = float(s[itk])
+                                itk += 1
+                                break
+                            except ValueError:
+                                pass
+                elif s[itk].lower().startswith("sgmn"):
+                    defs["sigma_min_loc"] = float(s[itk+1])
+                    itk += 2
+                elif s[itk].lower().startswith("sgmx"):
+                    defs["sigma_max_loc"] = float(s[itk+1])
+                    itk += 2
+                else:
+                    raise RuntimeError("Error==> EXTE keyword interpretation: {}".format(" ".join(s)))
+        elif s[1].lower().startswith(("miss", "unde")): # undefined
+            defs["ignore_undefined"] = s[2].lower().startswith("igno")
+        elif s[1].lower().startswith("hydr"): # hydrogen
+            defs["ignore_hydrogens"] = s[2].lower().startswith("igno")
+        elif s[1].lower().startswith("cut"): # TODO I think this should be parsed outside
+            ret["sd_ext_cut"] = float(s[2])  # as this affects everything (not only following blocks)
+        elif s[1].lower().startswith("dmax"):
+            defs["dist_max_external"] = float(s[2])
+        elif s[1].lower().startswith("dmin"):
+            defs["dist_min_external"] = float(s[2])
+        elif s[1].lower().startswith("use"):
+            defs["use_atoms"] = s[2][0].lower()
+            if defs["use_atoms"] not in ("a", "m", "h"):
+                logger.writeln("invalid exte use keyword: {}".format(s[2]))
+                defs["use_atoms"] = "a"
+        elif s[1].lower().startswith("conv"):
+            if s[2].lower().startswith("pref"):
+                defs["prefix_ch"] = s[3] # ???
+        elif s[1].lower().startswith(("dist", "plan", "chir", "angl", "inte", "tors")):
+            ret["rest_type"] = s[1][:4].lower()
+            itk = 2
+            iat = 0
+            ret["restr"] = {}
+            n_expect = dict(plan=0, dist=2, inte=2, angl=3).get(ret["rest_type"], 4)
+            ret["restr"]["specs"] = [None for _ in range(n_expect)]
+            while itk < len(s):
+                if s[itk].lower().startswith(("firs", "seco", "thir", "four", "next", "atre", "atin")):
+                    iat = dict(firs=0, seco=1, thir=2, four=3).get(s[itk][:4].lower(), iat+1)
+                    atoms, itk = parse_atom_spec(s, itk+1)
+                    if ret["rest_type"] == "plan":
+                        ret["restr"]["specs"].append(atoms)
+                    else:
+                        ret["restr"]["specs"][iat] = atoms
+                elif s[itk].lower().startswith("type"):
+                    try:
+                        ret["restr"]["itype_in"] = int(s[itk+1])
+                    except ValueError:
+                        ret["restr"]["itype_in"] = dict(o=0, f=2).get(s[itk+1][0].lower(), 1)
+                    if not (0 <= ret["restr"]["itype_in"] <= 2):
+                        logger.writeln("WARNING: wrong type is given. setting to 2.\n=> {}".format(" ".join(s)))
+                        ret["restr"]["itype_in"] = 2
+                    itk += 2
+                elif s[itk].lower().startswith("symm"): # only for distance and angle
+                    ret["restr"]["symm_in"] = s[itk+1][0].lower() == "y"
+                    itk += 2
+                else:
+                    d = dict(valu="value", dmin="dmin", dmax="dmax", smin="smin_value", smax="smax_value",
+                             sigm="sigma_value", alph="alpha_in", prob="prob_in")
+                    k = s[itk][:4].lower()
+                    if k in d:
+                        ret["restr"][d[k]] = float(s[itk+1])
+                        itk += 2
+                    else:
+                        logger.writeln("unrecognised key: {}\n=> {}".format(s[itk], " ".join(s)))
+                        break
+        elif s[1].lower().startswith("stac"):
+            ret["rest_type"] = "stac"
+            ret["restr"] = {}
+            ret["restr"]["specs"] = [[] for _ in range(2)]
+            itk = 2
+            #if s[itk].lower().startswith("dist"):
+            ip = 0
+            while itk < len(s):
+                if s[itk].lower().startswith("plan"):
+                    ip = int(s[itk+1])
+                    itk += 2
+                    if ip not in (1, 2):
+                        raise RuntimeError("Problem with stacking instructions. Plane number can be 1 or 2.\n=> {}".format(" ".join(s)))
+                elif s[itk].lower().startswith(("firs", "next")):
+                    atoms, itk = parse_atom_spec(s, itk+1)
+                    ret["restr"]["specs"][ip-1] = atoms
+                elif s[itk].lower().startswith(("dist", "sddi", "angl", "sdan", "type")):
+                    k = dict(dist="dist_id", sddi="dist_sd", angl="angle_id", sdan="angle_sd", type="type_r")[s[itk][:4].lower()]
+                    ret["restr"][k] = float(s[itk+1]) if k != "type_r" else int(s[itk+1])
+                    itk += 2
+                else:
+                    logger.writeln("WARNING: unrecognised keyword: {}\n=> {}".format(s[itk], " ".join(s)))
+                    itk += 1
+        elif s[1].lower().startswith(("harm", "spec")):
+            ret["rest_type"] = s[1][:4].lower() # in Refmac, irest_type = 1 if harm else 2
+            ret["restr"] = dict(rectype="", toler=0.5, sigma_t=0.5, sigma_u=2.0 * b_to_u, u_val_incl=False)
+            itk = 2
+            while itk < len(s):
+                if s[itk].lower().startswith("auto"):
+                    ret["restr"]["rectype"] = "auto"
+                    itk += 1
+                elif s[itk].lower().startswith("atin"):
+                    ret["restr"]["rectype"] = "atom"
+                    atoms, itk = parse_atom_spec(s, itk+1)
+                    ret["restr"]["specs"] = [atoms]
+                elif s[itk].lower().startswith("resi"):
+                    ret["restr"]["rectype"] = "resi"
+                    fromto, itk = parse_from_to(s, itk+1)
+                    ret["restr"]["specs"] = [fromto]
+                    if s[itk].lower().startswith("atom"):
+                        ret["restr"]["specs"][0]["atom"] = s[itk+1] # called atom_resi in Refmac
+                        itk += 2
+                elif s[itk].lower().startswith("sigm"):
+                    ret["restr"]["sigma_t"] = float(s[itk+1])
+                    itk += 2
+                elif s[itk].lower().startswith("tole"):
+                    ret["restr"]["toler"] = float(s[itk+1])
+                    itk += 2
+                elif s[itk].lower().startswith(("uval", "bval")):
+                    if len(s) > itk+1 and s[itk+1].lower().startswith("incl"):
+                        ret["restr"]["u_val_incl"] = True
+                        itk += 2
+                    else:
+                        ret["restr"]["u_val_incl"] = False
+                        itk += 1
+                elif s[itk].lower().startswith(("sigb", "sigu")):
+                    ret["restr"]["sigma_u"] = float(s[itk+1]) * b_to_u
+                    itk += 2
+                else:
+                    logger.writeln("WARNING: unrecognised keyword: {}\n=> {}".format(s[itk], " ".join(s)))
+                    itk += 1
+
+        else:
+            msg = "unrecognized exte keyword: " + " ".join(s)
+            if raise_unknown:
+                raise RuntimeError(msg)
+            else:
+                logger.writeln(f"WARNING: {msg}")
+    return ret
+# read_exte()
+
+def read_ridge_params(l, r, raise_unknown=True):
+    s = l.split()
+    assert s[0].lower().startswith("ridg")
+    ntok = len(s)
+    if s[1].lower().startswith("dist") and ntok > 2:
+        if s[2].lower().startswith("with"):
+            r.setdefault("groups", []).append({})
+            #r["groups"][-1]["sigma"] = sigma_dist_r
+            #r["groups"][-1]["dmax"] = dmax_dist_r
+            itk = 3
+            while itk < ntok:
+                if s[itk].lower().startswith("chai"):
+                    r["groups"][-1]["chain"] = s[itk+1]
+                    itk += 2
+                elif s[itk].lower().startswith("resi"):
+                    r["groups"][-1]["resi"] = (int(s[itk+1]), int(s[itk+2]))
+                    itk += 3
+                elif s[itk].lower().startswith("sigm"):
+                    v = float(s[itk+1])
+                    if v < 0: v = 0.01
+                    r["groups"][-1]["sigma"] = v
+                    itk += 2
+                elif s[itk].lower().startswith("dmax"):
+                    v = float(s[itk+1])
+                    if v < 0: v = 4.2
+                    r["groups"][-1]["dmax"] = v
+                    itk += 2
+        elif s[2].lower().startswith("incl") and ntok > 3:
+            # a: ridge_dist_include_all
+            # h: ridge_dist_include_hbond
+            # m: ridge_dist_include_main
+            v = s[3][0].lower()
+            if v in ("a", "h", "m"): r["include"] = v
+        elif s[2].lower().startswith("sigm"):
+            v = float(s[3])
+            r["sigma"] = v if v > 0 else 0.01
+        elif s[2].lower().startswith("dmax"):
+            v = float(s[3])
+            r["dmax"] = v if v > 0 else 4.2
+        elif s[2].lower().startswith("inte") and ntok > 3:
+            r["interchain"] = s[3][0].lower() == "y"
+        elif s[2].lower().startswith("symm") and ntok > 3:
+            r["intersym"] = s[3][0].lower() == "y"
+        elif s[2].lower().startswith("long") and ntok > 3:
+            r["long_range"] = max(0, int(s[3])) # long_range_residue_gap
+        elif s[2].lower().startswith("shor") and ntok > 3:
+            r["short_range"] = max(0, int(s[3])) # short_range_residue_gap
+        elif s[2].lower().startswith("hydr"):
+            r["hydrogen"] = ntok < 4 or s[3][0].lower() == "i" # hydrogens_include
+        elif s[2].lower().startswith("side") and ntok > 3:
+            r["sidechain"] = s[3][0].lower() == "i" # rb_side_chain_include
+        elif s[2].lower().startswith("filt"):
+            r["bvalue_filter"] = True
+            if s[3].lower().startswith("bran"):
+                v = float(s[4])
+                r["bvalue_filter_range"] = v if v > 0 else 2.0
+        else:
+            msg = "unrecognised keyword: {}\n=> {}".format(s[2], l)
+            if raise_unknown:
+                raise RuntimeError(msg)
+            else:
+                logger.writeln(f"WARNING: {msg}")
+    elif s[1].lower().startswith(("atom", "posi")): # not used
+        r["sigma_pos"] = float(s[2]) if ntok > 2 else 0.1
+    elif s[1].lower().startswith(("bval", "uval")) and ntok > 2:
+        if s[2].lower().startswith("diff"):
+            itk = 3
+            while itk < ntok:
+                if s[itk].lower().startswith("sigm"):
+                    if ntok > itk + 1:
+                        r["sigma_uval_diff"] = float(s[itk+1])
+                        itk += 2
+                    else:
+                        r["sigma_uval_diff"] = 0.025
+                        itk += 1
+                elif s[itk].lower().startswith("dmax"): 
+                    if ntok > itk + 1:
+                        r["dmax_uval_diff"] = float(s[itk+1])
+                        itk += 2
+                    else:
+                        r["dmax_uval_diff"] = 4.2
+                        itk += 1
+                elif s[itk].lower().startswith("dmwe"): # not used
+                    if ntok > itk + 1:
+                        r["dmax_uval_weight"] = float(s[itk+1]) * b_to_u
+                        itk += 2
+                    else:
+                        r["dmax_uval_weight"] = 3.0
+                        itk += 1
+                else:
+                    itk += 1
+        else:
+            r["sigma_b"] = float(s[2])
+            r["sigma_u"] = float(s[2])
+    else:
+        msg = "unrecognised keyword: {}\n=> {}".format(s[1], l)
+        if raise_unknown:
+            raise RuntimeError(msg)
+        else:
+            logger.writeln(f"WARNING: {msg}")
+
+    return r
+# read_ridge_params()
+
+def read_occupancy_params(l, r, raise_unknown=True):
+    s = l.split()
+    if not s[0].lower().startswith("occu"):
+        return r
+    ntok = len(s)
+    r.setdefault("groups", {}) # {igr: [{selection}, {selection}, ..]}
+    r.setdefault("const", []) # [[is_comp, group_ids]]
+    r.setdefault("ncycle", 0) # 0 means no refine
+    if (ntok > 4 and
+        s[1].lower().startswith("grou") and
+        s[2].lower().startswith("id")):
+        igr = s[3]
+        gr = r["groups"].setdefault(igr, [])
+        gr.append({})
+        itk = 4
+        while itk < ntok:
+            if s[itk].lower().startswith(("chai", "segm")):
+                gr[-1]["chains"] = []
+                itk += 1
+                while itk < ntok and not s[itk].lower().startswith(("resi","atom","alt")):
+                    gr[-1]["chains"].append(s[itk])
+                    itk += 1
+            elif s[itk].lower().startswith("resi"):
+                if s[itk+1].lower().startswith("from"):
+                    gr[-1]["resi_from"] = gemmi.SeqId(s[itk+2])
+                    if s[itk+3].lower().startswith("to"):
+                        gr[-1]["resi_to"] = gemmi.SeqId(s[itk+4])
+                    itk += 5
+                else:
+                    gr[-1]["resi"] = gemmi.SeqId(s[itk+1])
+                    itk += 2
+            elif s[itk].lower().startswith("atom"):
+                gr[-1]["atom"] = s[itk+1]
+                itk += 2
+            elif s[itk].lower().startswith("alt"):
+                gr[-1]["alt"] = s[itk+1]
+                itk += 2
+    elif (ntok > 4 and
+          s[1].lower().startswith("grou") and
+          s[2].lower().startswith("alts")):
+        r["const"].append((s[3].lower().startswith("comp"), s[4:]))
+    elif ntok > 1 and s[1].lower().startswith("refi"):
+        if ntok > 3 and s[2].lower().startswith("ncyc"):
+            r["ncycle"] = max(int(s[3]), r["ncycle"])
+        elif r["ncycle"] == 0:
+            r["ncycle"] = 1 # default
+    elif raise_unknown:
+        raise RuntimeError("unrecognized keyword: " + l)
+
+    return r
+# read_occupancy_params()
+
+def read_restr_params(l, r, raise_unknown=True):
+    s = l.split()
+
+    def read_tors_params(itk):
+        ret = {"flag": True}
+        if s[itk].lower().startswith("none"): # remove all
+            ret["flag"] = False
+            itk += 1
+        elif s[itk].lower().startswith("resi"):
+            ret["residue"] = s[itk+1]
+            itk += 2
+        elif s[itk].lower().startswith("grou"):
+            ret["group"] = s[itk+1] # group_name_tors_restr_o
+            itk += 2
+        elif s[itk].lower().startswith("link"):
+            ret["link"] = s[itk+1] #link_name_tors_restr_o
+            itk += 2
+        else:
+            pass # raise error?
+        while itk < len(s):
+            if s[itk].lower().startswith("name"):
+                itk += 1
+                ret["tors_name"] = s[itk] # RES_NAME_TORS_NAME_O
+            elif s[itk].lower().startswith("valu"):
+                itk += 1
+                ret["tors_value"] = float(s[itk]) # RES_NAME_TORS_VALUE_O
+            elif s[itk].lower().startswith("sigm"):
+                itk += 1
+                ret["tors_sigma"] = float(s[itk]) # RES_NAME_TORS_SIGMA_O
+            elif s[itk].lower().startswith("peri"):
+                itk += 1
+                ret["tors_period"] = int(s[itk]) # RES_NAME_TORS_PERIOD_O
+            itk += 1
+        return ret, itk
+    # read_tors_params()
+    
+    if not s[0].lower().startswith("rest"):
+        return r
+    if s[1].lower().startswith("excl"):
+        # restr_params.f90 subroutine exclude_restraints
+        #r["exclude"] = True
+        pass
+    elif s[1].lower().startswith("conf"): # not implemented in Refmac
+        pass
+    elif s[1].lower().startswith(("bp", "pair", "base")):
+        if s[2].lower().startswith("dist"):
+            r["plane_dist"] = True
+        else:
+            r["basepair"] = True
+            # TODO read dnarna_params.txt
+    elif s[1].lower().startswith("tors") and len(s) > 2:
+        itk = 2
+        if s[itk].lower().startswith("hydr"):
+            itk += 1
+            if s[itk].lower().startswith("incl"):
+                itk += 1
+                r["htors_restraint"] = True
+                if s[itk].lower().startswith("all"): # this is servalcat default for now
+                    r["htors_restraint_type"] = "all"
+                elif s[itk].lower().startswith("sele"):
+                    r["htors_restraint_type"] = "sele"
+                    r["htors_restraint_list"] = s[itk+1:]
+            else:
+                itk += 1
+                r["htors_restraint"] = False
+        elif s[itk].lower().startswith("fbas"):
+            pass # set user_basepair_file
+        elif s[itk].lower().startswith("incl"):
+            tmp, itk = read_tors_params(itk+1)
+            r.setdefault("torsion_include", []).append(tmp)
+        elif s[itk].lower().startswith("excl"):
+            # Should warn if value/sigma/period given?
+            tmp, itk = read_tors_params(itk+1)
+            r.setdefault("torsion_exclude", []).append(tmp)
+    elif s[1].lower().startswith("resi"):
+        pass
+    elif s[1].lower().startswith("chir"):
+        pass
+    elif raise_unknown:
+        raise RuntimeError("unrecognized keyword: " + l)
+# read_restr_params()
+
+def read_make_params(l, r, raise_unknown=True):
+    # TODO: hout,ribo,valu,spec,form,sdmi,segi
+    s = l.split()
+    assert s[0].lower().startswith("make")
+    itk = 1
+    ntok = len(s)
+    # keyword, key, func, possible, default
+    keys = (("hydr", "hydr", lambda x: x[0].lower(), set("aynf"), "a"),
+            ("hout", "hout", lambda x: x[0].lower() in ("y", "p"), (True, False), True),
+            ("chec", "check", lambda x: "0" if x.lower().startswith(("none", "0"))
+             else ("n" if x.lower().startswith(("liga", "n"))
+                   else ("y" if x.lower().startswith(("all", "y")) else None)),
+             set("0ny"), None), # no default
+            ("newl", "newligand", lambda x: x.lower().startswith(("c", "y", "noex")), (True, False), False),
+            ("buil", "build", lambda x: x[0].lower(), set("ny"), "n"),
+            ("pept", "pept", lambda x: x[0].lower(), set("yn"), "y"),
+            ("link", "link", lambda x: x[0].lower(), set("ynd0"), "y"),
+            ("suga", "sugar", lambda x: x[0].lower(), set("ynds"), "y"),
+            #("conn", "conn", lambda x: x[0].lower(), set("nyd0"), "n"), # TODO read conn_tolerance? (make conn tole val)
+            ("symm", "symm", lambda x: x[0].lower(), set("ync"), "y"),
+            ("chai", "chain", lambda x: x[0].lower(), set("yn"), "y"),
+            ("cisp", "cispept", lambda x: x[0].lower(), set("yn"), "y"),
+            ("ss", "ss", lambda x: x[0].lower(), set("ydn"), "y"),
+            ("exit", "exit", lambda x: x[0].lower() == "y", (True, False), True),
+            )
+    while itk < ntok:
+        found = False
+        for k, t, f, p, d in keys:
+            if s[itk].lower().startswith(k):
+                if itk + 1 < len(s):
+                    r[t] = f(s[itk+1])
+                    if r[t] not in p:
+                        raise SystemExit("Invalid make instruction: {}".format(l))
+                    itk += 2
+                elif d is not None:
+                    r[t] = d # set default
+                    itk += 1
+                else:
+                    raise SystemExit("Invalid make instruction: {}".format(l))
+                break
+        else:
+            if raise_unknown:
+                raise RuntimeError("unrecognized keyword: " + l)
+            else:
+                itk += 1
+    return r
+# read_make_params()
+
+def parse_line(l, ret, raise_unknown=True):
+    s = l.split()
+    ntok = len(s)
+    if ntok == 0: return
+    if s[0].lower().startswith("exte"):
+        ret.setdefault("exte", []).append(read_exte(s, raise_unknown))
+    elif s[0].lower().startswith("make"):
+        read_make_params(l, ret.setdefault("make", {}), raise_unknown)
+    elif s[0].lower().startswith(("sour", "scat")):
+        k = s[1].lower()
+        if k.startswith("em"):
+            ret["source"] = "em"
+        elif k.startswith("e"):
+            ret["source"] = "ec"
+        elif k.startswith("n"):
+            ret["source"] = "ne"
+        else:
+            ret["source"] = "xr"
+        # TODO check mb, lamb
+    elif s[0].lower().startswith("refi"): # TODO support this. Note that only valid with hklin
+        ret.setdefault("refi", {})
+        itk = 1
+        while itk < ntok:
+            if s[itk].lower().startswith("type"):
+                if itk+1 < ntok and s[itk+1].lower().startswith("unre"):
+                    ret["refi"]["type"] = "unre"
+                itk += 2
+            else:
+                itk += 1
+    elif s[0].lower().startswith("dist"):
+        try:
+            ret["wbond"] = float(s[1])
+        except:
+            pass
+        # TODO read sdex, excu, dele, dmxe, dmne
+    elif s[0].lower().startswith("ridg"):
+        read_ridge_params(l, ret.setdefault("ridge", {}), raise_unknown)
+    elif s[0].lower().startswith("occu"):
+        read_occupancy_params(l, ret.setdefault("occu", {}), raise_unknown)
+    elif s[0].lower().startswith("rest"):
+        read_restr_params(l, ret.setdefault("restr", {}), raise_unknown)
+    elif s[0].lower().startswith("angl") and ntok > 1:
+        ret["wangle"] = float(s[1])
+    elif s[0].lower().startswith("tors") and ntok > 1:
+        ret["wtors"] = float(s[1])
+    elif s[0].lower().startswith(("bfac", "temp", "bval")):
+        pass # TODO
+    elif s[0].lower().startswith("plan") and ntok > 1:
+        ret["wplane"] = float(s[1])
+    elif s[0].lower().startswith("chir") and ntok > 1:
+        ret["wchir"] = float(s[1])
+        # TODO read calp
+    elif s[0].lower().startswith(("vdwr", "vand", "nonb")) and ntok > 1:
+        itk = 1
+        try:
+            ret["wvdw"] = float(s[itk])
+            itk += 1
+        except ValueError:
+            pass
+    elif raise_unknown:
+        raise RuntimeError(f"unrecognized keyword: {l}")
+        # TODO read maxr, over, sigm, incr, chan, vdwc, excl
+# parse_line()
+
+def get_lines(lines, depth=0):
+    ret = []
+    cont = ""
+    for l in lines:
+        if "!" in l: l = l[:l.index("!")]
+        if "#" in l: l = l[:l.index("#")]
+        l = l.strip()
+        if not l: continue
+        if l[0] == "@":
+            f = l[1:]
+            try:
+                yield from get_lines(open(f).readlines(), depth+1)
+            except RuntimeError:
+                return
+            continue
+        if l.split()[-1] == "-":
+            cont += l[:l.rfind("-")] + " "
+            continue
+        if cont:
+            l = cont + l
+            cont = ""
+        if l.split()[0].lower().startswith("end"):
+            # refmac stops reading keywords when "exit" is seen in stdin or a file
+            # but won't do this from nested files
+            if depth == 1:
+                raise RuntimeError
+            break
+        yield l
+# get_lines()
+            
+def update_params(ret, inputs, raise_unknown=True):
+    if not inputs:
+        return
+    for l in get_lines(inputs):
+        parse_line(l, ret, raise_unknown=raise_unknown)
+# update_keywords()
+
+def parse_keywords(inputs, raise_unknown=True):
+    ret = {"make":{}, "ridge":{}, "refi":{}}
+    update_params(ret, inputs, raise_unknown)
+    return ret
+# parse_keywords()
+
+if __name__ == "__main__":
+    import sys
+    import json
+    print("waiting for input")
+    ret = {} #{"make":{}, "ridge":{}, "refi":{}}
+    for l in get_lines(sys.stdin):
+        parse_line(l, ret)
+    print()
+    print("Parsed:")
+    print(json.dumps(ret, indent=1))