mirror/tesseract
1
0
Fork 0
mirror of https://github.com/tesseract-ocr/tesseract.git synced 2024-11-24 11:09:06 +08:00
Code Issues Actions 10 Packages Projects Releases Wiki Activity

Replace PointerVector by std::vector in class DocumentCache

Browse Source 
Signed-off-by: Stefan Weil <sw@weilnetz.de>
This commit is contained in:
Stefan Weil 2021-02-26 17:10:49 +01:00
parent 387acd9881
commit 5cac52173e
2 changed files with 20 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
src/ccstruct/imagedata.cpp
View File

@ -582,8 +582,11 @@ bool DocumentData::ReCachePages() {
// A collection of DocumentData that knows roughly how much memory it is using.
DocumentCache::DocumentCache(int64_t max_memory)
: num_pages_per_doc_(0), max_memory_(max_memory) {}
DocumentCache::~DocumentCache() {}
: max_memory_(max_memory) {
}
DocumentCache::~DocumentCache() {
}
// Adds all the documents in the list of filenames, counting memory.
// The reader is used to read the files.
@ -619,9 +622,10 @@ bool DocumentCache::AddToCache(DocumentData* data) {
// Finds and returns a document by name.
DocumentData* DocumentCache::FindDocument(const STRING& document_name) const {
for (int i = 0; i < documents_.size(); ++i) {
if (documents_[i]->document_name() == document_name)
return documents_[i];
for (auto* document : documents_) {
if (document->document_name() == document_name) {
return document;
}
}
return nullptr;
}
@ -636,11 +640,10 @@ int DocumentCache::TotalPages() {
return num_pages_per_doc_ * documents_.size();
}
int total_pages = 0;
int num_docs = documents_.size();
for (int d = 0; d < num_docs; ++d) {
for (auto* document : documents_) {
// We have to load a page to make NumPages() valid.
documents_[d]->GetPage(0);
total_pages += documents_[d]->NumPages();
document->GetPage(0);
total_pages += document->NumPages();
}
return total_pages;
}
@ -683,8 +686,8 @@ const ImageData* DocumentCache::GetPageSequential(int serial) {
// Count up total memory. Background loading makes it more complicated to
// keep a running count.
int64_t total_memory = 0;
for (int d = 0; d < num_docs; ++d) {
total_memory += documents_[d]->memory_used();
for (auto* document : documents_) {
total_memory += document->memory_used();
}
if (total_memory >= max_memory_) {
// Find something to un-cache.

12
src/ccstruct/imagedata.h
View File

@ -22,7 +22,7 @@
#include "points.h" // for FCOORD
#include "genericvector.h" // for GenericVector, PointerVector, FileReader
#include "strngs.h" // for STRING
#include "strngs.h" // for STRING
#include <mutex> // for std::mutex
#include <thread> // for std::thread
@ -362,7 +362,7 @@ class DocumentCache {
return GetPageRoundRobin(serial);
}
const PointerVector<DocumentData>& documents() const {
const std::vector<DocumentData*>& documents() const {
return documents_;
}
// Returns the total number of pages in an epoch. For CS_ROUND_ROBIN cache
@ -387,14 +387,14 @@ class DocumentCache {
int CountNeighbourDocs(int index, int dir);
// A group of pages that corresponds in some loose way to a document.
PointerVector<DocumentData> documents_;
std::vector<DocumentData*> documents_;
// Strategy to use for caching and serializing data samples.
CachingStrategy cache_strategy_;
CachingStrategy cache_strategy_ = CS_SEQUENTIAL;
// Number of pages in the first document, used as a divisor in
// GetPageSequential to determine the document index.
int num_pages_per_doc_;
int num_pages_per_doc_ = 0;
// Max memory allowed in this cache.
int64_t max_memory_;
int64_t max_memory_ = 0;
};
} // namespace tesseract