RDKit
Open-source cheminformatics and machine learning.
Loading...
Searching...
No Matches
MultithreadedSDMolSupplier.h
Go to the documentation of this file.
1//
2// Copyright (C) 2020 Shrey Aryan
3//
4// @@ All Rights Reserved @@
5// This file is part of the RDKit.
6// The contents are covered by the terms of the BSD license
7// which is included in the file license.txt, found at the root
8// of the RDKit source tree.
9//
10#ifdef RDK_BUILD_THREADSAFE_SSS
11#ifndef MULTITHREADED_SD_MOL_SUPPLIER
12#define MULTITHREADED_SD_MOL_SUPPLIER
14namespace RDKit {
15namespace v2 {
16namespace FileParsers {
17
18//! This class is still a bit experimental and the public API may change
19//! in future releases.
22 public:
23 explicit MultithreadedSDMolSupplier(
24 const std::string &fileName, const Parameters &params = Parameters(),
25 const MolFileParserParams &parseParams = MolFileParserParams());
26
27 explicit MultithreadedSDMolSupplier(
28 std::istream *inStream, bool takeOwnership = true,
29 const Parameters &params = Parameters(),
30 const MolFileParserParams &parseParams = MolFileParserParams());
31
32 MultithreadedSDMolSupplier();
33 ~MultithreadedSDMolSupplier() override;
34 void init() override {}
35
36 void checkForEnd();
37 bool getEnd() const override;
38 void setProcessPropertyLists(bool val) { df_processPropertyLists = val; }
39 bool getProcessPropertyLists() const { return df_processPropertyLists; }
40 bool getEOFHitOnRead() const { return df_eofHitOnRead; }
41
42 //! reads next record and returns whether or not EOF was hit
43 bool extractNextRecord(std::string &record, unsigned int &lineNum,
44 unsigned int &index) override;
45 void readMolProps(RWMol &mol, std::istringstream &inStream);
46 //! parses the record and returns the resulting molecule
47 RWMol *processMoleculeRecord(const std::string &record,
48 unsigned int lineNum) override;
49
50 private:
51 void initFromSettings(bool takeOwnership, const Parameters &params,
52 const MolFileParserParams &parseParams);
53
54 bool df_end = false; //!< have we reached the end of the file?
55 int d_line = 0; //!< line number we are currently on
56 bool df_processPropertyLists = true;
57 bool df_eofHitOnRead = false;
58 unsigned int d_currentRecordId = 1; //!< current record id
59 MolFileParserParams d_parseParams;
60};
61} // namespace FileParsers
62} // namespace v2
63
64inline namespace v1 {
66 //! this is an abstract base class to concurrently supply molecules one at a
67 //! time
68 public:
69 using ContainedType = v2::FileParsers::MultithreadedSDMolSupplier;
70 MultithreadedSDMolSupplier() {}
71 explicit MultithreadedSDMolSupplier(
72 const std::string &fileName, bool sanitize = true, bool removeHs = true,
73 bool strictParsing = true, unsigned int numWriterThreads = 1,
74 size_t sizeInputQueue = 5, size_t sizeOutputQueue = 5) {
75 v2::FileParsers::MultithreadedSDMolSupplier::Parameters params;
76 params.numWriterThreads = numWriterThreads;
77 params.sizeInputQueue = sizeInputQueue;
78 params.sizeOutputQueue = sizeOutputQueue;
79 v2::FileParsers::MolFileParserParams parseParams;
80 parseParams.sanitize = sanitize;
81 parseParams.removeHs = removeHs;
82 parseParams.strictParsing = strictParsing;
83
84 dp_supplier.reset(new v2::FileParsers::MultithreadedSDMolSupplier(
85 fileName, params, parseParams));
86 }
87
88 explicit MultithreadedSDMolSupplier(
89 std::istream *inStream, bool takeOwnership = true, bool sanitize = true,
90 bool removeHs = true, bool strictParsing = true,
91 unsigned int numWriterThreads = 1, size_t sizeInputQueue = 5,
92 size_t sizeOutputQueue = 5) {
93 v2::FileParsers::MultithreadedSDMolSupplier::Parameters params;
94 params.numWriterThreads = numWriterThreads;
95 params.sizeInputQueue = sizeInputQueue;
96 params.sizeOutputQueue = sizeOutputQueue;
97 v2::FileParsers::MolFileParserParams parseParams;
98 parseParams.sanitize = sanitize;
99 parseParams.removeHs = removeHs;
100 parseParams.strictParsing = strictParsing;
101
102 dp_supplier.reset(new v2::FileParsers::MultithreadedSDMolSupplier(
103 inStream, takeOwnership, params, parseParams));
104 }
105
106 //! included for the interface, always returns false
107 bool getEOFHitOnRead() const {
108 if (dp_supplier) {
109 return static_cast<ContainedType *>(dp_supplier.get())->getEOFHitOnRead();
110 }
111 return false;
112 }
113
114 //! returns the record id of the last extracted item
115 //! Note: d_LastRecordId = 0, initially therefore the value 0 is returned
116 //! if and only if the function is called before extracting the first
117 //! record
118 unsigned int getLastRecordId() const {
119 PRECONDITION(dp_supplier, "no supplier");
120 return static_cast<ContainedType *>(dp_supplier.get())->getLastRecordId();
121 }
122 //! returns the text block for the last extracted item
123 std::string getLastItemText() const {
124 PRECONDITION(dp_supplier, "no supplier");
125 return static_cast<ContainedType *>(dp_supplier.get())->getLastItemText();
126 }
127 void setProcessPropertyLists(bool val) {
128 PRECONDITION(dp_supplier, "no supplier");
129 static_cast<ContainedType *>(dp_supplier.get())
130 ->setProcessPropertyLists(val);
131 }
132 bool getProcessPropertyLists() const {
133 PRECONDITION(dp_supplier, "no supplier");
134 return static_cast<ContainedType *>(dp_supplier.get())
135 ->getProcessPropertyLists();
136 }
137};
138} // namespace v1
139} // namespace RDKit
140#endif
141#endif
#define PRECONDITION(expr, mess)
Definition Invariant.h:109
#define RDKIT_FILEPARSERS_EXPORT
Definition export.h:161
RDKIT_GRAPHMOL_EXPORT ROMol * removeHs(const ROMol &mol, bool implicitOnly=false, bool updateExplicitCount=false, bool sanitize=true)
returns a copy of a molecule with hydrogens removed
Std stuff.
bool rdvalue_is(const RDValue_cast_t)