Logo ROOT  
Reference Guide
 
Loading...
Searching...
No Matches
RRawFile.cxx
Go to the documentation of this file.
1// @(#)root/io:$Id$
2// Author: Jakob Blomer
3
4/*************************************************************************
5 * Copyright (C) 1995-2018, Rene Brun and Fons Rademakers. *
6 * All rights reserved. *
7 * *
8 * For the licensing terms see $ROOTSYS/LICENSE. *
9 * For the list of contributors see $ROOTSYS/README/CREDITS. *
10 *************************************************************************/
11
12#include <ROOT/RConfig.h>
13#include <ROOT/RRawFile.hxx>
14#ifdef _WIN32
15#include <ROOT/RRawFileWin.hxx>
16#else
17#include <ROOT/RRawFileUnix.hxx>
18#endif
19
20#include "TError.h"
21#include "TPluginManager.h"
22#include "TROOT.h"
23
24#include <algorithm>
25#include <cctype> // for towlower
26#include <cerrno>
27#include <cstddef>
28#include <cstdint>
29#include <cstring>
30#include <stdexcept>
31#include <string>
32
33namespace {
34const char *kTransportSeparator = "://";
35// Corresponds to ELineBreaks
36#ifdef _WIN32
37const char *kLineBreakTokens[] = {"", "\r\n", "\n", "\r\n"};
38constexpr unsigned int kLineBreakTokenSizes[] = {0, 2, 1, 2};
39#else
40const char *kLineBreakTokens[] = {"", "\n", "\n", "\r\n"};
41constexpr unsigned int kLineBreakTokenSizes[] = {0, 1, 1, 2};
42#endif
43constexpr unsigned int kLineBuffer = 128; // On Readln, look for line-breaks in chunks of 128 bytes
44} // anonymous namespace
45
46size_t ROOT::Internal::RRawFile::RBlockBuffer::CopyTo(void *buffer, size_t nbytes, std::uint64_t offset)
47{
49 return 0;
50
51 size_t copiedBytes = 0;
52 std::uint64_t offsetInBuffer = offset - fBufferOffset;
53 if (offsetInBuffer < static_cast<std::uint64_t>(fBufferSize)) {
54 size_t bytesInBuffer = std::min(nbytes, static_cast<size_t>(fBufferSize - offsetInBuffer));
55 memcpy(buffer, fBuffer + offsetInBuffer, bytesInBuffer);
56 copiedBytes = bytesInBuffer;
57 }
58 return copiedBytes;
59}
60
61ROOT::Internal::RRawFile::RRawFile(std::string_view url, ROptions options)
63 fOptions(options), fFilePos(0)
64{
65}
66
68{
69 delete[] fBufferSpace;
70}
71
72std::unique_ptr<ROOT::Internal::RRawFile>
73ROOT::Internal::RRawFile::Create(std::string_view url, ROptions options)
74{
75 std::string transport = GetTransport(url);
76 if (transport == "file") {
77#ifdef _WIN32
78 return std::unique_ptr<RRawFile>(new RRawFileWin(url, options));
79#else
80 return std::unique_ptr<RRawFile>(new RRawFileUnix(url, options));
81#endif
82 }
83 if (transport == "http" || transport == "https" ||
84 transport == "root" || transport == "roots" ) {
85 std::string plgclass = transport.compare( 0, 4, "http" ) == 0 ?
86 "RRawFileDavix" : "RRawFileNetXNG";
87 if (TPluginHandler *h = gROOT->GetPluginManager()->
88 FindHandler("ROOT::Internal::RRawFile", std::string(url).c_str())) {
89 if (h->LoadPlugin() == 0) {
90 return std::unique_ptr<RRawFile>(reinterpret_cast<RRawFile *>(h->ExecPlugin(2, &url, &options)));
91 }
92 throw std::runtime_error("Cannot load plugin handler for " + plgclass);
93 }
94 throw std::runtime_error("Cannot find plugin handler for " + plgclass);
95 }
96 throw std::runtime_error("Unsupported transport protocol: " + transport);
97}
98
100{
101 if (fIsOpen)
102 return;
103
104 OpenImpl();
105 fIsOpen = true;
106}
107
108void *ROOT::Internal::RRawFile::MapImpl(size_t /* nbytes */, std::uint64_t /* offset */,
109 std::uint64_t& /* mapdOffset */)
110{
111 throw std::runtime_error("Memory mapping unsupported");
112}
113
114void ROOT::Internal::RRawFile::ReadVImpl(RIOVec *ioVec, unsigned int nReq)
115{
116 for (unsigned i = 0; i < nReq; ++i) {
117 ioVec[i].fOutBytes = ReadAt(ioVec[i].fBuffer, ioVec[i].fSize, ioVec[i].fOffset);
118 }
119}
120
121void ROOT::Internal::RRawFile::UnmapImpl(void * /* region */, size_t /* nbytes */)
122{
123 throw std::runtime_error("Memory mapping unsupported");
124}
125
126std::string ROOT::Internal::RRawFile::GetLocation(std::string_view url)
127{
128 auto idx = url.find(kTransportSeparator);
129 if (idx == std::string_view::npos)
130 return std::string(url);
131 return std::string(url.substr(idx + strlen(kTransportSeparator)));
132}
133
135{
136 if (fFileSize != kUnknownFileSize)
137 return fFileSize;
138
139 EnsureOpen();
140 fFileSize = GetSizeImpl();
141 return fFileSize;
142}
143
145 return fUrl;
146}
147
148std::string ROOT::Internal::RRawFile::GetTransport(std::string_view url)
149{
150 auto idx = url.find(kTransportSeparator);
151 if (idx == std::string_view::npos)
152 return "file";
153 std::string transport(url.substr(0, idx));
154 std::transform(transport.begin(), transport.end(), transport.begin(), ::tolower);
155 return transport;
156}
157
158void *ROOT::Internal::RRawFile::Map(size_t nbytes, std::uint64_t offset, std::uint64_t &mapdOffset)
159{
160 EnsureOpen();
161 return MapImpl(nbytes, offset, mapdOffset);
162}
163
164size_t ROOT::Internal::RRawFile::Read(void *buffer, size_t nbytes)
165{
166 size_t res = ReadAt(buffer, nbytes, fFilePos);
167 fFilePos += res;
168 return res;
169}
170
171size_t ROOT::Internal::RRawFile::ReadAt(void *buffer, size_t nbytes, std::uint64_t offset)
172{
173 EnsureOpen();
174 R__ASSERT(fOptions.fBlockSize >= 0);
175
176 // "Large" reads are served directly, bypassing the cache
177 if (nbytes > static_cast<unsigned int>(fOptions.fBlockSize))
178 return ReadAtImpl(buffer, nbytes, offset);
179
180 if (fBufferSpace == nullptr) {
181 fBufferSpace = new unsigned char[kNumBlockBuffers * fOptions.fBlockSize];
182 for (unsigned int i = 0; i < kNumBlockBuffers; ++i)
183 fBlockBuffers[i].fBuffer = fBufferSpace + i * fOptions.fBlockSize;
184 }
185
186 size_t totalBytes = 0;
187 size_t copiedBytes = 0;
188 /// Try to serve as many bytes as possible from the block buffers
189 for (unsigned int idx = fBlockBufferIdx; idx < fBlockBufferIdx + kNumBlockBuffers; ++idx) {
190 copiedBytes = fBlockBuffers[idx % kNumBlockBuffers].CopyTo(buffer, nbytes, offset);
191 buffer = reinterpret_cast<unsigned char *>(buffer) + copiedBytes;
192 nbytes -= copiedBytes;
193 offset += copiedBytes;
194 totalBytes += copiedBytes;
195 if (copiedBytes > 0)
196 fBlockBufferIdx = idx;
197 if (nbytes == 0)
198 return totalBytes;
199 }
200 fBlockBufferIdx++;
201
202 /// The request was not fully satisfied and fBlockBufferIdx now points to the previous shadow buffer
203
204 /// The remaining bytes populate the newly promoted main buffer
205 RBlockBuffer *thisBuffer = &fBlockBuffers[fBlockBufferIdx % kNumBlockBuffers];
206 size_t res = ReadAtImpl(thisBuffer->fBuffer, fOptions.fBlockSize, offset);
207 thisBuffer->fBufferOffset = offset;
208 thisBuffer->fBufferSize = res;
209 size_t remainingBytes = std::min(res, nbytes);
210 memcpy(buffer, thisBuffer->fBuffer, remainingBytes);
211 totalBytes += remainingBytes;
212 return totalBytes;
213}
214
215void ROOT::Internal::RRawFile::ReadV(RIOVec *ioVec, unsigned int nReq)
216{
217 EnsureOpen();
218 ReadVImpl(ioVec, nReq);
219}
220
222{
223 if (fOptions.fLineBreak == ELineBreaks::kAuto) {
224 // Auto-detect line breaks according to the break discovered in the first line
225 fOptions.fLineBreak = ELineBreaks::kUnix;
226 bool res = Readln(line);
227 if ((line.length() > 0) && (*line.rbegin() == '\r')) {
228 fOptions.fLineBreak = ELineBreaks::kWindows;
229 line.resize(line.length() - 1);
230 }
231 return res;
232 }
233
234 line.clear();
235 char buffer[kLineBuffer];
236 size_t nbytes;
237 do {
238 nbytes = Read(buffer, sizeof(buffer));
239 std::string_view bufferView(buffer, nbytes);
240 auto idx = bufferView.find(kLineBreakTokens[static_cast<int>(fOptions.fLineBreak)]);
241 if (idx != std::string_view::npos) {
242 // Line break found, return the string and skip the linebreak itself
243 line.append(buffer, idx);
244 fFilePos -= nbytes - idx;
245 fFilePos += kLineBreakTokenSizes[static_cast<int>(fOptions.fLineBreak)];
246 return true;
247 }
248 line.append(buffer, nbytes);
249 } while (nbytes > 0);
250
251 return !line.empty();
252}
253
255{
256 fFilePos = offset;
257}
258
259void ROOT::Internal::RRawFile::Unmap(void *region, size_t nbytes)
260{
261 if (!fIsOpen)
262 throw std::runtime_error("Cannot unmap, file not open");
263 UnmapImpl(region, nbytes);
264}
size_t fSize
std::string fBuffer
#define h(i)
Definition RSha256.hxx:106
#define R__ASSERT(e)
Definition TError.h:118
Option_t Option_t TPoint TPoint const char GetTextMagnitude GetFillStyle GetLineColor GetLineWidth GetMarkerStyle GetTextAlign GetTextColor GetTextSize void char Point_t Rectangle_t WindowAttributes_t Float_t Float_t Float_t Int_t Int_t UInt_t UInt_t Rectangle_t Int_t Int_t Window_t TString Int_t GCValues_t GetPrimarySelectionOwner GetDisplay GetScreen GetColormap GetNativeEvent const char const char dpyName wid window const char font_name cursor keysym reg const char only_if_exist regb h Point_t winding char text const char depth char const char Int_t count const char ColorStruct_t color const char Pixmap_t Pixmap_t PictureAttributes_t attr const char char ret_data h unsigned char height h offset
#define gROOT
Definition TROOT.h:406
The RRawFileUnix class uses POSIX calls to read from a mounted file system.
The RRawFileWin class uses portable C I/O calls to read from a drive.
The RRawFile provides read-only access to local and remote files.
Definition RRawFile.hxx:43
unsigned char * fBufferSpace
Memory block containing the block buffers consecutively.
Definition RRawFile.hxx:124
static std::string GetLocation(std::string_view url)
Returns only the file location, e.g. "server/file" for http://server/file.
Definition RRawFile.cxx:126
unsigned int fBlockBufferIdx
To be used modulo kNumBlockBuffers, points to the last used block buffer in fBlockBuffers.
Definition RRawFile.hxx:120
RRawFile(std::string_view url, ROptions options)
Definition RRawFile.cxx:61
std::uint64_t fFilePos
The current position in the file, which can be changed by Seek, Read, and Readln.
Definition RRawFile.hxx:134
virtual void * MapImpl(size_t nbytes, std::uint64_t offset, std::uint64_t &mapdOffset)
If a derived class supports mmap, the MapImpl and UnmapImpl calls are supposed to be implemented,...
Definition RRawFile.cxx:108
void Unmap(void *region, size_t nbytes)
Receives a pointer returned by Map() and should have nbytes set to the full length of the mapping.
Definition RRawFile.cxx:259
virtual void ReadVImpl(RIOVec *ioVec, unsigned int nReq)
By default implemented as a loop of ReadAt calls but can be overwritten, e.g. XRootD or DAVIX impleme...
Definition RRawFile.cxx:114
static std::string GetTransport(std::string_view url)
Returns only the transport protocol in lower case, e.g. "http" for HTTP://server/file.
Definition RRawFile.cxx:148
std::uint64_t GetSize()
Returns the size of the file.
Definition RRawFile.cxx:134
static std::unique_ptr< RRawFile > Create(std::string_view url, ROptions options=ROptions())
Factory method that returns a suitable concrete implementation according to the transport in the url.
Definition RRawFile.cxx:73
void Seek(std::uint64_t offset)
Change the cursor fFilePos.
Definition RRawFile.cxx:254
static constexpr std::uint64_t kUnknownFileSize
Derived classes do not necessarily need to provide file size information but they can return "not kno...
Definition RRawFile.hxx:46
size_t ReadAt(void *buffer, size_t nbytes, std::uint64_t offset)
Buffered read from a random position.
Definition RRawFile.cxx:171
bool fIsOpen
Files are opened lazily and only when required; the open state is kept by this flag.
Definition RRawFile.hxx:128
bool Readln(std::string &line)
Read the next line starting from the current value of fFilePos. Returns false if the end of the file ...
Definition RRawFile.cxx:221
void EnsureOpen()
Open the file if not already open. Otherwise noop.
Definition RRawFile.cxx:99
void ReadV(RIOVec *ioVec, unsigned int nReq)
Opens the file if necessary and calls ReadVImpl.
Definition RRawFile.cxx:215
size_t Read(void *buffer, size_t nbytes)
Read from fFilePos offset. Returns the actual number of bytes read.
Definition RRawFile.cxx:164
std::string GetUrl() const
Returns the url of the file.
Definition RRawFile.cxx:144
std::uint64_t fFileSize
The cached file size.
Definition RRawFile.hxx:126
virtual void UnmapImpl(void *region, size_t nbytes)
Derived classes with mmap support must be able to unmap the memory area handed out by Map()
Definition RRawFile.cxx:121
void * Map(size_t nbytes, std::uint64_t offset, std::uint64_t &mapdOffset)
Memory mapping according to POSIX standard; in particular, new mappings of the same range replace old...
Definition RRawFile.cxx:158
TLine * line
std::uint64_t fBufferOffset
Where in the open file does fBuffer start.
Definition RRawFile.hxx:105
unsigned char * fBuffer
Points into the I/O buffer with data from the file, not owned.
Definition RRawFile.hxx:109
size_t CopyTo(void *buffer, size_t nbytes, std::uint64_t offset)
Tries to copy up to nbytes starting at offset from fBuffer into buffer. Returns number of bytes copie...
Definition RRawFile.cxx:46
size_t fBufferSize
The number of currently buffered bytes in fBuffer.
Definition RRawFile.hxx:107
Used for vector reads from multiple offsets into multiple buffers.
Definition RRawFile.hxx:71
std::size_t fOutBytes
The number of actually read bytes, set by ReadV()
Definition RRawFile.hxx:79
On construction, an ROptions parameter can customize the RRawFile behavior.
Definition RRawFile.hxx:59