sdf-parser - npm Package Compare versions

Comparing version 5.0.2 to 6.0.0

lib/index.js

@@ -6,14 +6,5 @@ 'use strict';
 var ensureString = require('ensure-string');
-var pipeline = require('pumpify');
-var split2 = require('split2');
-var through2 = require('through2');
-var filter = require('through2-filter');
+var readline = require('readline');
+var dynamicTyping = require('dynamic-typing');
 
-function _interopDefaultLegacy (e) { return e && typeof e === 'object' && 'default' in e ? e : { 'default': e }; }
-
-var pipeline__default = /*#__PURE__*/_interopDefaultLegacy(pipeline);
-var split2__default = /*#__PURE__*/_interopDefaultLegacy(split2);
-var through2__default = /*#__PURE__*/_interopDefaultLegacy(through2);
-var filter__default = /*#__PURE__*/_interopDefaultLegacy(filter);
-
 function getEntriesBoundaries(string, substring, eol) {
@@ -41,21 +32,75 @@ const res = [];
 
+function getMolecule$1(sdfPart, labels, currentLabels, options) {
+  let parts = sdfPart.split(`${options.eol}>`);
+  if (parts.length === 0 || parts[0].length <= 5) return;
+  let molecule = {};
+  molecule.molfile = parts[0] + options.eol;
+  for (let j = 1; j < parts.length; j++) {
+    let lines = parts[j].split(options.eol);
+    let from = lines[0].indexOf('<');
+    let to = lines[0].indexOf('>');
+    let label = lines[0].substring(from + 1, to);
+    currentLabels.push(label);
+    if (!labels[label]) {
+      labels[label] = {
+        counter: 0,
+        isNumeric: options.dynamicTyping,
+        keep: false,
+      };
+      if (
+        (!options.exclude || options.exclude.indexOf(label) === -1) &&
+        (!options.include || options.include.indexOf(label) > -1)
+      ) {
+        labels[label].keep = true;
+        if (options.modifiers[label]) {
+          labels[label].modifier = options.modifiers[label];
+        }
+        if (options.forEach[label]) {
+          labels[label].forEach = options.forEach[label];
+        }
+      }
+    }
+    if (labels[label].keep) {
+      for (let k = 1; k < lines.length - 1; k++) {
+        if (molecule[label]) {
+          molecule[label] += options.eol + lines[k];
+        } else {
+          molecule[label] = lines[k];
+        }
+      }
+      if (labels[label].modifier) {
+        let modifiedValue = labels[label].modifier(molecule[label]);
+        if (modifiedValue === undefined || modifiedValue === null) {
+          delete molecule[label];
+        } else {
+          molecule[label] = modifiedValue;
+        }
+      }
+      if (labels[label].isNumeric) {
+        if (!isFinite(molecule[label]) || molecule[label].match(/^0[0-9]/)) {
+          labels[label].isNumeric = false;
+        }
+      }
+    }
+  }
+  return molecule;
+}
+
 /**
  *  Parse a SDF file
  * @param {string|ArrayBuffer|Uint8Array} sdf SDF file to parse
- * @param {any} [options={}]
- * @param {array<string>} [options.include] List of fields to include
- * @param {array<string>} [options.exclude] List of fields to exclude
+ * @param {object} [options={}]
+ * @param {string[]} [options.include] List of fields to include
+ * @param {string[]} [options.exclude] List of fields to exclude
+ * @param {Function} [options.filter] Callback allowing to filter the molecules
  * @param {boolean} [options.dynamicTyping] Dynamically type the data
  * @param {object} [options.modifiers] Object containing callbacks to apply on some specific fields
  * @param {boolean} [options.mixedEOL=false] Set to true if you know there is a mixture between \r\n and \n
+ * @param {string} [options.eol] Specify the end of line character. Default will be the one found in the file
  */
 function parse(sdf, options = {}) {
-  const {
-    include,
-    exclude,
-    filter,
-    modifiers = {},
-    forEach = {},
-    dynamicTyping = true,
-  } = options;
+  options = { ...options };
+  if (options.modifiers === undefined) options.modifiers = {};
+  if (options.forEach === undefined) options.forEach = {};
+  if (options.dynamicTyping === undefined) options.dynamicTyping = true;
 
@@ -67,17 +112,23 @@ sdf = ensureString.ensureString(sdf);
 
-  let eol = '\n';
-  if (options.mixedEOL) {
-    sdf = sdf.replace(/\r\n/g, '\n');
-    sdf = sdf.replace(/\r/g, '\n');
-  } else {
-    // we will find the delimiter in order to be much faster and not use regular expression
-    let header = sdf.substr(0, 1000);
-    if (header.indexOf('\r\n') > -1) {
-      eol = '\r\n';
-    } else if (header.indexOf('\r') > -1) {
-      eol = '\r';
+  if (options.eol === undefined) {
+    options.eol = '\n';
+    if (options.mixedEOL) {
+      sdf = sdf.replace(/\r\n/g, '\n');
+      sdf = sdf.replace(/\r/g, '\n');
+    } else {
+      // we will find the delimiter in order to be much faster and not use regular expression
+      let header = sdf.substr(0, 1000);
+      if (header.indexOf('\r\n') > -1) {
+        options.eol = '\r\n';
+      } else if (header.indexOf('\r') > -1) {
+        options.eol = '\r';
+      }
     }
   }
 
-  let entriesBoundaries = getEntriesBoundaries(sdf, `${eol}$$$$`, eol);
+  let entriesBoundaries = getEntriesBoundaries(
+    sdf,
+    `${options.eol}$$$$`,
+    options.eol,
+  );
   let molecules = [];
@@ -90,68 +141,14 @@ let labels = {};
     let sdfPart = sdf.substring(...entriesBoundaries[i]);
-    let parts = sdfPart.split(`${eol}>`);
-    if (parts.length > 0 && parts[0].length > 5) {
-      let molecule = {};
-      let currentLabels = [];
-      molecule.molfile = parts[0] + eol;
-      for (let j = 1; j < parts.length; j++) {
-        let lines = parts[j].split(eol);
-        let from = lines[0].indexOf('<');
-        let to = lines[0].indexOf('>');
-        let label = lines[0].substring(from + 1, to);
-        currentLabels.push(label);
-        if (!labels[label]) {
-          labels[label] = {
-            counter: 0,
-            isNumeric: dynamicTyping,
-            keep: false,
-          };
-          if (
-            (!exclude || exclude.indexOf(label) === -1) &&
-            (!include || include.indexOf(label) > -1)
-          ) {
-            labels[label].keep = true;
-            if (modifiers[label]) {
-              labels[label].modifier = modifiers[label];
-            }
-            if (forEach[label]) {
-              labels[label].forEach = forEach[label];
-            }
-          }
-        }
-        if (labels[label].keep) {
-          for (let k = 1; k < lines.length - 1; k++) {
-            if (molecule[label]) {
-              molecule[label] += eol + lines[k];
-            } else {
-              molecule[label] = lines[k];
-            }
-          }
-          if (labels[label].modifier) {
-            let modifiedValue = labels[label].modifier(molecule[label]);
-            if (modifiedValue === undefined || modifiedValue === null) {
-              delete molecule[label];
-            } else {
-              molecule[label] = modifiedValue;
-            }
-          }
-          if (labels[label].isNumeric) {
-            if (
-              !isFinite(molecule[label]) ||
-              molecule[label].match(/^0[0-9]/)
-            ) {
-              labels[label].isNumeric = false;
-            }
-          }
-        }
+
+    let currentLabels = [];
+    const molecule = getMolecule$1(sdfPart, labels, currentLabels, options);
+    if (!molecule) continue;
+    if (!options.filter || options.filter(molecule)) {
+      molecules.push(molecule);
+      // only now we can increase the counter
+      for (let j = 0; j < currentLabels.length; j++) {
+        labels[currentLabels[j]].counter++;
       }
-      if (!filter || filter(molecule)) {
-        molecules.push(molecule);
-        // only now we can increase the counter
-        for (let j = 0; j < currentLabels.length; j++) {
-          labels[currentLabels[j]].counter++;
-        }
-      }
     }
   }
-
   // all numeric fields should be converted to numbers
@@ -202,43 +199,55 @@ for (let label in labels) {
 
-const filterStream = filter__default["default"].bind(null, { objectMode: true });
+/**
+ *  Parse a SDF file
+ * @param {string|ArrayBuffer|Uint8Array} sdf SDF file to parse
+ * @param {object} [options={}]
+ * @param {Function} [options.filter] Callback allowing to filter the molecules
+ * @param {boolean} [options.dynamicTyping] Dynamically type the data
+ */
 
-function filterCb(chunk) {
-  return chunk.length > 1 && chunk.trim().length > 1;
-}
+async function* iterator(readStream, options = {}) {
+  const lines = readline.createInterface(readStream);
+  const currentLines = [];
+  options = { ...options };
+  if (options.dynamicTyping === undefined) options.dynamicTyping = true;
 
-function entries() {
-  return pipeline__default["default"].obj(
-    split2__default["default"](/\r?\n\${4}.*\r?\n/),
-    filterStream(filterCb),
-    through2__default["default"]({ objectMode: true }, function process(value, encoding, callback) {
-      const eol = value.includes('\r\n') ? '\r\n' : '\n';
-      this.push(`${value + eol}$$$$${eol}`);
-      callback();
-    }),
-  );
+  options.eol = '\n';
+  for await (let line of lines) {
+    if (line.startsWith('$$$$')) {
+      const molecule = getMolecule(currentLines.join(options.eol), options);
+      if (!options.filter || options.filter(molecule)) {
+        yield molecule;
+      }
+      currentLines.length = 0;
+    } else {
+      currentLines.push(line);
+    }
+  }
 }
 
-function molecules(options) {
-  return pipeline__default["default"].obj(
-    entries(),
-    through2__default["default"]({ objectMode: true }, function process(value, encoding, callback) {
-      try {
-        const parsed = parse(value, options);
-        if (parsed.molecules.length === 1) {
-          if (options && options.fullResult) {
-            this.push(parsed);
-          } else {
-            this.push(parsed.molecules[0]);
-          }
-        }
-        callback();
-      } catch (e) {
-        callback(e);
+function getMolecule(sdfPart, options) {
+  let parts = sdfPart.split(`${options.eol}>`);
+  if (parts.length === 0 || parts[0].length <= 5) return;
+  let molecule = {};
+  molecule.molfile = parts[0] + options.eol;
+  for (let j = 1; j < parts.length; j++) {
+    let lines = parts[j].split(options.eol);
+    let from = lines[0].indexOf('<');
+    let to = lines[0].indexOf('>');
+    let label = lines[0].substring(from + 1, to);
+    for (let k = 1; k < lines.length - 1; k++) {
+      if (molecule[label]) {
+        molecule[label] += options.eol + lines[k];
+      } else {
+        molecule[label] = lines[k];
       }
-    }),
-  );
+    }
+    if (options.dynamicTyping) {
+      molecule[label] = dynamicTyping.parseString(molecule[label]);
+    }
+  }
+  return molecule;
 }
 
-exports.entries = entries;
-exports.molecules = molecules;
+exports.iterator = iterator;
 exports.parse = parse;

package.json

 {
   "name": "sdf-parser",
-  "version": "5.0.2",
+  "version": "6.0.0",
   "description": "SDF parser",
@@ -52,2 +52,3 @@ "main": "lib/index.js",
     "eslint-config-cheminfo": "^8.0.2",
+    "filelist-utils": "^0.6.0",
     "jest": "^28.1.3",
@@ -58,8 +59,5 @@ "openchemlib": "^8.0.1",
   "dependencies": {
-    "ensure-string": "^1.2.0",
-    "pumpify": "^2.0.1",
-    "split2": "^4.1.0",
-    "through2": "^4.0.2",
-    "through2-filter": "^3.0.0"
+    "dynamic-typing": "^1.0.0",
+    "ensure-string": "^1.2.0"
   }
 }

README.md

@@ -60,37 +60,16 @@ # sdf-parser
 
-## Streams
+## Iterator
 
 This API is only available on Node.js.
 
-### molecules(options)
-
-Transform an input text stream to a stream of molecule objects.
-
-#### options
-
-- `fullResult`: true to emit the full result of `parse` instead of just the molecules.
-- All other options from the `parse` function.
-
 ```js
-const { stream } = require('sdf-parser');
-fs.createReadStream('test.sdf')
-  .pipe(stream.molecules())
-  .on('data', (molecule) => {
-    console.log(molecule.molfile);
-  });
+const { iterator } = require('sdf-parser');
+const readStream = createReadStream(join(__dirname, 'test.sdf.gz'));
+const stream = readStream.pipe(createGunzip());
+const results = [];
+for await (const entry of iterator(stream)) {
+  results.push(entry);
+}
 ```
 
-### entries()
-
-Transform an input text stream to a stream of sdf entries.
-
-```js
-const { stream } = require('sdf-parser');
-fs.createReadStream('test.sdf')
-  .pipe(stream.entries())
-  .on('data', (entry) => {
-    // sdf entry as a string
-  });
-```
-
 ## License
@@ -102,5 +81,5 @@
 [npm-url]: https://www.npmjs.com/package/sdf-parser
-[travis-image]: https://img.shields.io/travis/cheminfo/sdf-parser/master.svg?style=flat-square
+[travis-image]: https://img.shields.io/travis/cheminfo/sdf-parser/main.svg?style=flat-square
 [travis-url]: https://travis-ci.org/cheminfo/sdf-parser
 [download-image]: https://img.shields.io/npm/dm/sdf-parser.svg?style=flat-square
 [download-url]: https://www.npmjs.com/package/sdf-parser

src/index.js

 export * from './parse';
-export * from './stream';
+export * from './iterator';

src/parse.js

 import { ensureString } from 'ensure-string';
 
 import { getEntriesBoundaries } from './getEntriesBoundaries';
+import { getMolecule } from './util/getMolecule';
 /**
  *  Parse a SDF file
  * @param {string|ArrayBuffer|Uint8Array} sdf SDF file to parse
- * @param {any} [options={}]
- * @param {array<string>} [options.include] List of fields to include
- * @param {array<string>} [options.exclude] List of fields to exclude
+ * @param {object} [options={}]
+ * @param {string[]} [options.include] List of fields to include
+ * @param {string[]} [options.exclude] List of fields to exclude
+ * @param {Function} [options.filter] Callback allowing to filter the molecules
  * @param {boolean} [options.dynamicTyping] Dynamically type the data
  * @param {object} [options.modifiers] Object containing callbacks to apply on some specific fields
  * @param {boolean} [options.mixedEOL=false] Set to true if you know there is a mixture between \r\n and \n
+ * @param {string} [options.eol] Specify the end of line character. Default will be the one found in the file
  */
 export function parse(sdf, options = {}) {
-  const {
-    include,
-    exclude,
-    filter,
-    modifiers = {},
-    forEach = {},
-    dynamicTyping = true,
-  } = options;
+  options = { ...options };
+  if (options.modifiers === undefined) options.modifiers = {};
+  if (options.forEach === undefined) options.forEach = {};
+  if (options.dynamicTyping === undefined) options.dynamicTyping = true;
 
@@ -29,17 +28,23 @@ sdf = ensureString(sdf);
 
-  let eol = '\n';
-  if (options.mixedEOL) {
-    sdf = sdf.replace(/\r\n/g, '\n');
-    sdf = sdf.replace(/\r/g, '\n');
-  } else {
-    // we will find the delimiter in order to be much faster and not use regular expression
-    let header = sdf.substr(0, 1000);
-    if (header.indexOf('\r\n') > -1) {
-      eol = '\r\n';
-    } else if (header.indexOf('\r') > -1) {
-      eol = '\r';
+  if (options.eol === undefined) {
+    options.eol = '\n';
+    if (options.mixedEOL) {
+      sdf = sdf.replace(/\r\n/g, '\n');
+      sdf = sdf.replace(/\r/g, '\n');
+    } else {
+      // we will find the delimiter in order to be much faster and not use regular expression
+      let header = sdf.substr(0, 1000);
+      if (header.indexOf('\r\n') > -1) {
+        options.eol = '\r\n';
+      } else if (header.indexOf('\r') > -1) {
+        options.eol = '\r';
+      }
     }
   }
 
-  let entriesBoundaries = getEntriesBoundaries(sdf, `${eol}$$$$`, eol);
+  let entriesBoundaries = getEntriesBoundaries(
+    sdf,
+    `${options.eol}$$$$`,
+    options.eol,
+  );
   let molecules = [];
@@ -52,68 +57,14 @@ let labels = {};
     let sdfPart = sdf.substring(...entriesBoundaries[i]);
-    let parts = sdfPart.split(`${eol}>`);
-    if (parts.length > 0 && parts[0].length > 5) {
-      let molecule = {};
-      let currentLabels = [];
-      molecule.molfile = parts[0] + eol;
-      for (let j = 1; j < parts.length; j++) {
-        let lines = parts[j].split(eol);
-        let from = lines[0].indexOf('<');
-        let to = lines[0].indexOf('>');
-        let label = lines[0].substring(from + 1, to);
-        currentLabels.push(label);
-        if (!labels[label]) {
-          labels[label] = {
-            counter: 0,
-            isNumeric: dynamicTyping,
-            keep: false,
-          };
-          if (
-            (!exclude || exclude.indexOf(label) === -1) &&
-            (!include || include.indexOf(label) > -1)
-          ) {
-            labels[label].keep = true;
-            if (modifiers[label]) {
-              labels[label].modifier = modifiers[label];
-            }
-            if (forEach[label]) {
-              labels[label].forEach = forEach[label];
-            }
-          }
-        }
-        if (labels[label].keep) {
-          for (let k = 1; k < lines.length - 1; k++) {
-            if (molecule[label]) {
-              molecule[label] += eol + lines[k];
-            } else {
-              molecule[label] = lines[k];
-            }
-          }
-          if (labels[label].modifier) {
-            let modifiedValue = labels[label].modifier(molecule[label]);
-            if (modifiedValue === undefined || modifiedValue === null) {
-              delete molecule[label];
-            } else {
-              molecule[label] = modifiedValue;
-            }
-          }
-          if (labels[label].isNumeric) {
-            if (
-              !isFinite(molecule[label]) ||
-              molecule[label].match(/^0[0-9]/)
-            ) {
-              labels[label].isNumeric = false;
-            }
-          }
-        }
+
+    let currentLabels = [];
+    const molecule = getMolecule(sdfPart, labels, currentLabels, options);
+    if (!molecule) continue;
+    if (!options.filter || options.filter(molecule)) {
+      molecules.push(molecule);
+      // only now we can increase the counter
+      for (let j = 0; j < currentLabels.length; j++) {
+        labels[currentLabels[j]].counter++;
       }
-      if (!filter || filter(molecule)) {
-        molecules.push(molecule);
-        // only now we can increase the counter
-        for (let j = 0; j < currentLabels.length; j++) {
-          labels[currentLabels[j]].counter++;
-        }
-      }
     }
   }
-
   // all numeric fields should be converted to numbers
@@ -120,0 +71,0 @@ for (let label in labels) {

New alerts

Native code

Supply chain risk

Contains native code (e.g., compiled binaries or shared libraries). Including native code can obscure malicious behavior.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

393703

increased by9.78%

Dependency count

2

decreased by-60%

Number of package files

22

increased by10%

Lines of code

748

increased by7.16%

Worsened metrics

Dev dependency count

10

increased by11.11%

Number of lines in readme file

84

decreased by-20%

Number of medium supply chain risk alerts

1

increased byInfinity%

Dependency changes

• + Addeddynamic-typing@^1.0.0

• + Addeddynamic-typing@1.0.1(transitive)

• - Removedpumpify@^2.0.1

• - Removedsplit2@^4.1.0

• - Removedthrough2@^4.0.2

• - Removedthrough2-filter@^3.0.0

• - Removedduplexify@4.1.3(transitive)

• - Removedend-of-stream@1.4.4(transitive)

• - Removedinherits@2.0.4(transitive)

• - Removedonce@1.4.0(transitive)

• - Removedpump@3.0.2(transitive)

• - Removedpumpify@2.0.1(transitive)

• - Removedreadable-stream@3.6.2(transitive)

• - Removedsafe-buffer@5.2.1(transitive)

• - Removedsplit2@4.2.0(transitive)

• - Removedstream-shift@1.0.3(transitive)

• - Removedstring_decoder@1.3.0(transitive)

• - Removedthrough2@4.0.2(transitive)

• - Removedthrough2-filter@3.1.0(transitive)

• - Removedutil-deprecate@1.0.2(transitive)

• - Removedwrappy@1.0.2(transitive)