From 261e96068a104051eb9fe56c51f5cc73eccf8b18 Mon Sep 17 00:00:00 2001
From: Alexander Rose <alex.rose@rcsb.org>
Date: Tue, 3 Oct 2017 22:35:42 -0700
Subject: [PATCH] Gro parser using computation
---
src/reader/common/text/tokenizer.ts | 16 ++++-
src/reader/gro/parser.ts | 15 ++--
src/reader/spec/gro.spec.ts | 10 +--
src/script.ts | 103 +++++++++++++++-------------
4 files changed, 84 insertions(+), 60 deletions(-)
diff --git a/src/reader/common/text/tokenizer.ts b/src/reader/common/text/tokenizer.ts
index 326497591..e4fac8678 100644
--- a/src/reader/common/text/tokenizer.ts
+++ b/src/reader/common/text/tokenizer.ts
@@ -6,6 +6,8 @@
* @author Alexander Rose <alexander.rose@weirdbyte.de>
*/
+import Computation from '../../../utils/computation'
+
export interface Tokenizer {
data: string
@@ -15,6 +17,8 @@ export interface Tokenizer {
currentLineNumber: number
currentTokenStart: number
currentTokenEnd: number
+
+ computation: Computation.Chunked
}
export interface Tokens {
@@ -23,14 +27,15 @@ export interface Tokens {
indices: ArrayLike<number>
}
-export function Tokenizer(data: string): Tokenizer {
+export function Tokenizer(data: string, ctx: Computation.Context): Tokenizer {
return {
data,
position: 0,
length: data.length,
currentLineNumber: 1,
currentTokenStart: 0,
- currentTokenEnd: 0
+ currentTokenEnd: 0,
+ computation: new Computation.Chunked(ctx, 1000000)
};
}
@@ -86,12 +91,17 @@ export namespace Tokenizer {
}
/** Advance the state by the given number of lines and return line starts/ends as tokens. */
- export function readLines(state: Tokenizer, count: number): Tokens {
+ export async function readLines(state: Tokenizer, count: number): Promise<Tokens> {
+ const { computation, position, length } = state
const lineTokens = TokenBuilder.create(state, count * 2);
for (let i = 0; i < count; i++) {
markLine(state);
TokenBuilder.addUnchecked(lineTokens, state.currentTokenStart, state.currentTokenEnd);
+
+ if (computation.requiresUpdate) {
+ await computation.updateProgress('Parsing...', void 0, position, length);
+ }
}
return { data: state.data, count, indices: lineTokens.indices };
diff --git a/src/reader/gro/parser.ts b/src/reader/gro/parser.ts
index 28a1e26ca..7573b6d4c 100644
--- a/src/reader/gro/parser.ts
+++ b/src/reader/gro/parser.ts
@@ -10,6 +10,7 @@ import FixedColumn from '../common/text/column/fixed'
import { ColumnType, UndefinedColumn } from '../common/column'
import * as Schema from './schema'
import Result from '../result'
+import Computation from '../../utils/computation'
interface State {
tokenizer: Tokenizer,
@@ -85,9 +86,9 @@ function handleNumberOfAtoms(state: State) {
* position (in nm, x y z in 3 columns, each 8 positions with 3 decimal places)
* velocity (in nm/ps (or km/s), x y z in 3 columns, each 8 positions with 4 decimal places)
*/
-function handleAtoms(state: State): Schema.Atoms {
+async function handleAtoms(state: State): Promise<Schema.Atoms> {
const { tokenizer, numberOfAtoms } = state;
- const lines = Tokenizer.readLines(tokenizer, numberOfAtoms);
+ const lines = await Tokenizer.readLines(tokenizer, numberOfAtoms);
const positionSample = tokenizer.data.substring(lines.indices[0], lines.indices[1]).substring(20);
const precisions = positionSample.match(/\.\d+/g)!;
@@ -134,15 +135,15 @@ function handleBoxVectors(state: State) {
state.header.box = [+values[0], +values[1], +values[2]];
}
-function parseInternal(data: string): Result<Schema.File> {
- const tokenizer = Tokenizer(data);
+async function parseInternal(data: string, ctx: Computation.Context): Promise<Result<Schema.File>> {
+ const tokenizer = Tokenizer(data, ctx);
const structures: Schema.Structure[] = [];
while (tokenizer.position < data.length) {
const state = State(tokenizer);
handleTitleString(state);
handleNumberOfAtoms(state);
- const atoms = handleAtoms(state);
+ const atoms = await handleAtoms(state);
handleBoxVectors(state);
structures.push({ header: state.header, atoms });
}
@@ -152,7 +153,9 @@ function parseInternal(data: string): Result<Schema.File> {
}
export function parse(data: string) {
- return parseInternal(data);
+ return new Computation<Result<Schema.File>>(async ctx => {
+ return await parseInternal(data, ctx);
+ });
}
export default parse;
\ No newline at end of file
diff --git a/src/reader/spec/gro.spec.ts b/src/reader/spec/gro.spec.ts
index 67dd4b21d..157688fa4 100644
--- a/src/reader/spec/gro.spec.ts
+++ b/src/reader/spec/gro.spec.ts
@@ -25,8 +25,9 @@ const groStringHighPrecision = `Generated by trjconv : 2168 system t= 15.00000
1.82060 2.82060 3.82060`
describe('gro reader', () => {
- it('basic', () => {
- const parsed = Gro(groString)
+ it('basic', async () => {
+ const comp = Gro(groString)
+ const parsed = await comp.run()
if (parsed.isError) {
console.log(parsed)
@@ -57,8 +58,9 @@ describe('gro reader', () => {
expect(atoms.vz.value(5)).toBeCloseTo(-0.0244, 0.0001);
});
- it('high precision', () => {
- const parsed = Gro(groStringHighPrecision);
+ it('high precision', async () => {
+ const comp = Gro(groStringHighPrecision)
+ const parsed = await comp.run()
if (parsed.isError) {
console.log(parsed)
diff --git a/src/script.ts b/src/script.ts
index 9db88b748..3a040acf6 100644
--- a/src/script.ts
+++ b/src/script.ts
@@ -2,6 +2,7 @@
* Copyright (c) 2017 molio contributors, licensed under MIT, See LICENSE file for more info.
*
* @author Alexander Rose <alexander.rose@weirdbyte.de>
+ * @author David Sehnal <david.sehnal@gmail.com>
*/
// import * as util from 'util'
@@ -15,69 +16,77 @@ import CIF from './reader/cif/index'
// const file = 'test.gro'
const file = 'md_1u19_trj.gro'
-export function _gro() {
- fs.readFile(`./examples/${file}`, 'utf8', function (err, input) {
- if (err) {
- return console.log(err);
- }
- // console.log(data);
-
- console.time('parse')
- const parsed = Gro(input)
- console.timeEnd('parse')
- if (parsed.isError) {
- console.log(parsed)
- return;
- }
+async function runGro(input: string) {
+ console.time('parseGro');
+ const comp = Gro(input);
+ const running = comp.runWithContext(new Computation.ObservableContext({ updateRateMs: 250 }));
+ running.subscribe(p => console.log(`[Gro] ${(p.current / p.max * 100).toFixed(2)} (${p.elapsedMs | 0}ms)`));
+ const parsed = await running.result;
+ console.timeEnd('parseGro');
+
+ if (parsed.isError) {
+ console.log(parsed);
+ return;
+ }
- const groFile = parsed.result
+ const groFile = parsed.result
- console.log('structure count: ', groFile.structures.length);
+ console.log('structure count: ', groFile.structures.length);
- const data = groFile.structures[0];
+ const data = groFile.structures[0];
- // const header = groFile.blocks[0].getCategory('header')
- const { header, atoms } = data;
- console.log(JSON.stringify(header, null, 2));
- console.log('number of atoms:', atoms.count);
+ // const header = groFile.blocks[0].getCategory('header')
+ const { header, atoms } = data;
+ console.log(JSON.stringify(header, null, 2));
+ console.log('number of atoms:', atoms.count);
- console.log(`'${atoms.residueNumber.value(1)}'`)
- console.log(`'${atoms.residueName.value(1)}'`)
- console.log(`'${atoms.atomName.value(1)}'`)
- console.log(atoms.z.value(1))
- console.log(`'${atoms.z.value(1)}'`)
+ console.log(`'${atoms.residueNumber.value(1)}'`)
+ console.log(`'${atoms.residueName.value(1)}'`)
+ console.log(`'${atoms.atomName.value(1)}'`)
+ console.log(atoms.z.value(1))
+ console.log(`'${atoms.z.value(1)}'`)
- const n = atoms.count;
- console.log('rowCount', n)
+ const n = atoms.count;
+ console.log('rowCount', n)
- console.time('getFloatArray x')
- const x = atoms.x.toArray({ array: Float32Array })
- console.timeEnd('getFloatArray x')
- console.log(x.length, x[0], x[x.length - 1])
+ console.time('getFloatArray x')
+ const x = atoms.x.toArray({ array: Float32Array })
+ console.timeEnd('getFloatArray x')
+ console.log(x.length, x[0], x[x.length - 1])
- console.time('getFloatArray y')
- const y = atoms.y.toArray({ array: Float32Array })
- console.timeEnd('getFloatArray y')
- console.log(y.length, y[0], y[y.length - 1])
+ console.time('getFloatArray y')
+ const y = atoms.y.toArray({ array: Float32Array })
+ console.timeEnd('getFloatArray y')
+ console.log(y.length, y[0], y[y.length - 1])
- console.time('getFloatArray z')
- const z = atoms.z.toArray({ array: Float32Array })
- console.timeEnd('getFloatArray z')
- console.log(z.length, z[0], z[z.length - 1])
+ console.time('getFloatArray z')
+ const z = atoms.z.toArray({ array: Float32Array })
+ console.timeEnd('getFloatArray z')
+ console.log(z.length, z[0], z[z.length - 1])
- console.time('getIntArray residueNumber')
- const residueNumber = atoms.residueNumber.toArray({ array: Int32Array })
- console.timeEnd('getIntArray residueNumber')
- console.log(residueNumber.length, residueNumber[0], residueNumber[residueNumber.length - 1])
+ console.time('getIntArray residueNumber')
+ const residueNumber = atoms.residueNumber.toArray({ array: Int32Array })
+ console.timeEnd('getIntArray residueNumber')
+ console.log(residueNumber.length, residueNumber[0], residueNumber[residueNumber.length - 1])
+}
+
+async function _gro() {
+ fs.readFile(`./examples/${file}`, 'utf8', function (err, input) {
+ if (err) {
+ return console.log(err);
+ }
+ runGro(input)
});
}
+_gro()
+
async function runCIF(input: string | Uint8Array) {
console.time('parseCIF');
const comp = typeof input === 'string' ? CIF.parseText(input) : CIF.parseBinary(input);
-
+
const running = comp.runWithContext(new Computation.ObservableContext({ updateRateMs: 250 }));
- running.subscribe(p => console.log(`${(p.current / p.max * 100).toFixed(2)} (${p.elapsedMs | 0}ms)`));
+ running.subscribe(p => console.log(`[CIF] ${(p.current / p.max * 100).toFixed(2)} (${p.elapsedMs | 0}ms)`));
const parsed = await running.result;
console.timeEnd('parseCIF');
if (parsed.isError) {
@@ -98,7 +107,7 @@ async function runCIF(input: string | Uint8Array) {
export function _cif() {
let path = `./examples/1cbs_updated.cif`;
- path = 'c:/test/quick/3j3q.cif';
+ path = '../test/3j3q.cif';
fs.readFile(path, 'utf8', function (err, input) {
if (err) {
return console.log(err);
--
GitLab