diff --git a/src/mol-io/reader/cif/data-model.ts b/src/mol-io/reader/cif/data-model.ts
index c5778c7b55a847616c5c36ff57dcf300f9af1926..db35b9419e9e38c74f50ad7ed7c9be6b9cabcb51 100644
--- a/src/mol-io/reader/cif/data-model.ts
+++ b/src/mol-io/reader/cif/data-model.ts
@@ -101,7 +101,7 @@ export namespace CifField {
return ofStrings([value]);
}
- export function ofStrings(values: string[]): CifField {
+ export function ofStrings(values: ArrayLike<string>): CifField {
const rowCount = values.length;
const str: CifField['str'] = row => { const ret = values[row]; if (!ret || ret === '.' || ret === '?') return ''; return ret; };
const int: CifField['int'] = row => { const v = values[row]; return fastParseInt(v, 0, v.length) || 0; };
diff --git a/src/mol-model-formats/structure/_spec/pdb.spec.ts b/src/mol-model-formats/structure/_spec/pdb.spec.ts
index 8e365ed4c9c8f2fc1b01e447e32e5314eeb6bd45..90a453277857d128415367de3ab0b63113f1a9a4 100644
--- a/src/mol-model-formats/structure/_spec/pdb.spec.ts
+++ b/src/mol-model-formats/structure/_spec/pdb.spec.ts
@@ -4,8 +4,8 @@
* @author Alexander Rose <alexander.rose@weirdbyte.de>
*/
-import { guessElementSymbol } from '../pdb/to-cif';
import { TokenBuilder } from 'mol-io/reader/common/text/tokenizer';
+import { guessElementSymbolTokens } from '../util';
const records = [
['ATOM 19 HD23 LEU A 1 151.940 143.340 155.670 0.00 0.00', 'H'],
@@ -19,7 +19,7 @@ describe('PDB to-cif', () => {
for (let i = 0, il = records.length; i < il; ++i) {
const [ data, element ] = records[i]
const tokens = TokenBuilder.create(data, 2)
- guessElementSymbol(tokens, data, 12, 16)
+ guessElementSymbolTokens(tokens, data, 12, 16)
expect(data.substring(tokens.indices[0], tokens.indices[1])).toBe(element)
}
});
diff --git a/src/mol-model-formats/structure/gro.ts b/src/mol-model-formats/structure/gro.ts
index bd75839a3008eba2d73005d55155c0f40d210e08..2b7d95817cd4c73a9b0d9ce797cdd2e7910f7af5 100644
--- a/src/mol-model-formats/structure/gro.ts
+++ b/src/mol-model-formats/structure/gro.ts
@@ -12,6 +12,7 @@ import { GroFile, GroAtoms } from 'mol-io/reader/gro/schema';
import { CifCategory, CifField } from 'mol-io/reader/cif';
import { Column } from 'mol-data/db';
import { mmCIF_Schema } from 'mol-io/reader/cif/schema/mmcif';
+import { guessElementSymbolString } from './util';
// TODO multi model files
// TODO seperate chains
@@ -37,7 +38,7 @@ function _atom_site(atoms: GroAtoms): { [K in keyof mmCIF_Schema['atom_site']]?:
auth_comp_id,
auth_seq_id,
B_iso_or_equiv: CifField.ofColumn(Column.Undefined(atoms.count, Column.Schema.float)),
- Cartn_x: CifField.ofNumbers(Column.mapToArray(atoms.x, x => x * 10, Float32Array)),
+ Cartn_x: CifField.ofNumbers(Column.mapToArray(atoms.x, x => x * 10, Float32Array)),
Cartn_y: CifField.ofNumbers(Column.mapToArray(atoms.y, y => y * 10, Float32Array)),
Cartn_z: CifField.ofNumbers(Column.mapToArray(atoms.z, z => z * 10, Float32Array)),
group_PDB: CifField.ofColumn(Column.Undefined(atoms.count, Column.Schema.str)),
@@ -52,7 +53,8 @@ function _atom_site(atoms: GroAtoms): { [K in keyof mmCIF_Schema['atom_site']]?:
label_entity_id: CifField.ofColumn(Column.ofConst('1', atoms.count, Column.Schema.str)),
occupancy: CifField.ofColumn(Column.ofConst(1, atoms.count, Column.Schema.float)),
- type_symbol: CifField.ofColumn(Column.Undefined(atoms.count, Column.Schema.str)),
+ type_symbol: CifField.ofStrings(Column.mapToArray(atoms.atomName, s => guessElementSymbolString(s))),
+ // type_symbol: CifField.ofColumn(Column.Undefined(atoms.count, Column.Schema.str)),
pdbx_PDB_ins_code: CifField.ofColumn(Column.Undefined(atoms.count, Column.Schema.str)),
pdbx_PDB_model_num: CifField.ofColumn(Column.ofConst('1', atoms.count, Column.Schema.str)),
diff --git a/src/mol-model-formats/structure/pdb/to-cif.ts b/src/mol-model-formats/structure/pdb/to-cif.ts
index 0f699f297a4c3f03095cd1514bfb8d7db7c129b8..d280d1d371d8aa7c87410dfc9da5c884a517c4c9 100644
--- a/src/mol-model-formats/structure/pdb/to-cif.ts
+++ b/src/mol-model-formats/structure/pdb/to-cif.ts
@@ -8,11 +8,12 @@
import { substringStartsWith } from 'mol-util/string';
import { CifField, CifCategory, CifFrame } from 'mol-io/reader/cif';
import { mmCIF_Schema } from 'mol-io/reader/cif/schema/mmcif';
-import { TokenBuilder, Tokenizer, Tokens } from 'mol-io/reader/common/text/tokenizer';
+import { TokenBuilder, Tokenizer } from 'mol-io/reader/common/text/tokenizer';
import { PdbFile } from 'mol-io/reader/pdb/schema';
import { parseCryst1, parseRemark350, parseMtrix } from './assembly';
import { WaterNames } from 'mol-model/structure/model/types';
import { parseHelix, parseSheet } from './secondary-structure';
+import { guessElementSymbolTokens } from '../util';
function _entity(): { [K in keyof mmCIF_Schema['entity']]?: CifField } {
return {
@@ -89,43 +90,6 @@ function getEntityId(residueName: string, isHet: boolean) {
return '1';
}
-export function guessElementSymbol(tokens: Tokens, str: string, start: number, end: number) {
- let s = start, e = end - 1
-
- // trim spaces and numbers
- let c = str.charCodeAt(s)
- while ((c === 32 || (c >= 48 && c <= 57)) && s <= e) c = str.charCodeAt(++s)
- c = str.charCodeAt(e)
- while ((c === 32 || (c >= 48 && c <= 57)) && e >= s) c = str.charCodeAt(--e)
-
- ++e
-
- if (s === e) return TokenBuilder.add(tokens, s, e) // empty
- if (s + 1 === e) return TokenBuilder.add(tokens, s, e) // one char
-
- c = str.charCodeAt(s)
-
- if (s + 2 === e) { // two chars
- const c2 = str.charCodeAt(s + 1)
- if (
- ((c === 78 || c === 110) && (c2 === 65 || c2 === 97)) || // NA na Na nA
- ((c === 67 || c === 99) && (c2 === 76 || c2 === 108)) || // CL
- ((c === 70 || c === 102) && (c2 === 69 || c2 === 101)) // FE
- ) return TokenBuilder.add(tokens, s, s + 2)
- }
-
- if (
- c === 67 || c === 99 || // C c
- c === 72 || c === 104 || // H h
- c === 78 || c === 110 || // N n
- c === 79 || c === 111 || // O o
- c === 80 || c === 112 || // P p
- c === 83 || c === 115 // S s
- ) return TokenBuilder.add(tokens, s, s + 1)
-
- TokenBuilder.add(tokens, s, s) // no reasonable guess, add empty token
-}
-
function addAtom(sites: AtomSiteTemplate, model: string, data: Tokenizer, s: number, e: number, isHet: boolean) {
const { data: str } = data;
const length = e - s;
@@ -199,10 +163,10 @@ function addAtom(sites: AtomSiteTemplate, model: string, data: Tokenizer, s: num
if (data.tokenStart < data.tokenEnd) {
TokenBuilder.addToken(sites.type_symbol, data);
} else {
- guessElementSymbol(sites.type_symbol, str, s + 12, s + 16)
+ guessElementSymbolTokens(sites.type_symbol, str, s + 12, s + 16)
}
} else {
- guessElementSymbol(sites.type_symbol, str, s + 12, s + 16)
+ guessElementSymbolTokens(sites.type_symbol, str, s + 12, s + 16)
}
sites.label_entity_id[sites.index] = getEntityId(residueName, isHet);
diff --git a/src/mol-model-formats/structure/util.ts b/src/mol-model-formats/structure/util.ts
new file mode 100644
index 0000000000000000000000000000000000000000..3fc4b147cd52913d580339b6e5d41ca3e3dcec04
--- /dev/null
+++ b/src/mol-model-formats/structure/util.ts
@@ -0,0 +1,62 @@
+/**
+ * Copyright (c) 2019 mol* contributors, licensed under MIT, See LICENSE file for more info.
+ *
+ * @author Alexander Rose <alexander.rose@weirdbyte.de>
+ */
+
+import { TokenBuilder, Tokens } from 'mol-io/reader/common/text/tokenizer';
+
+export function guessElementSymbolTokens(tokens: Tokens, str: string, start: number, end: number) {
+ let s = start, e = end - 1
+
+ // trim spaces and numbers
+ let c = str.charCodeAt(s)
+ while ((c === 32 || (c >= 48 && c <= 57)) && s <= e) c = str.charCodeAt(++s)
+ c = str.charCodeAt(e)
+ while ((c === 32 || (c >= 48 && c <= 57)) && e >= s) c = str.charCodeAt(--e)
+
+ ++e
+
+ if (s === e) return TokenBuilder.add(tokens, s, e) // empty
+ if (s + 1 === e) return TokenBuilder.add(tokens, s, e) // one char
+
+ c = str.charCodeAt(s)
+
+ if (s + 2 === e) { // two chars
+ const c2 = str.charCodeAt(s + 1)
+ if (
+ ((c === 78 || c === 110) && (c2 === 65 || c2 === 97)) || // NA na Na nA
+ ((c === 67 || c === 99) && (c2 === 76 || c2 === 108)) || // CL
+ ((c === 70 || c === 102) && (c2 === 69 || c2 === 101)) // FE
+ ) return TokenBuilder.add(tokens, s, s + 2)
+ }
+
+ if (
+ c === 67 || c === 99 || // C c
+ c === 72 || c === 104 || // H h
+ c === 78 || c === 110 || // N n
+ c === 79 || c === 111 || // O o
+ c === 80 || c === 112 || // P p
+ c === 83 || c === 115 // S s
+ ) return TokenBuilder.add(tokens, s, s + 1)
+
+ TokenBuilder.add(tokens, s, s) // no reasonable guess, add empty token
+}
+
+export function guessElementSymbolString(str: string) {
+ // trim spaces and numbers, convert to upper case
+ str = str.trim().toUpperCase()
+ const l = str.length
+
+ if (l === 0) return str // empty
+ if (l === 1) return str // one char
+
+ if (l === 2) { // two chars
+ if (str === 'NA' || str === 'CL' || str === 'FE') return str
+ }
+
+ const c = str[0]
+ if (c === 'C' || c === 'H' || c === 'N' || c === 'O' || c === 'P' || c === 'S') return c
+
+ return '' // no reasonable guess, return empty string
+}
\ No newline at end of file