diff options
| author | Oliver Sander <oliver.sander@tu-dresden.de> | 2019-10-14 20:08:33 +0200 |
|---|---|---|
| committer | Oliver Sander <oliver.sander@tu-dresden.de> | 2019-10-23 08:51:46 +0000 |
| commit | e076a478e4d2f7d6445215814c86e6ac4654f575 (patch) | |
| tree | ea6c1545e0c8366877ef0f6bd7a924056e629c99 /poppler | |
| parent | 8911527dc38908ee56626fc55d09041475cc5145 (diff) | |
Make FontInfo::scan return a std::vector object
... rather than a pointer to a std::vector. Given that a
std::vector is little more than a pointer and some size
information, there is no need to create std::vector objects
on the heap. Returning them by value is just as fast
(the vector content is not copied), and makes the code
more readable, too.
Diffstat (limited to 'poppler')
| -rw-r--r-- | poppler/FontInfo.cc | 8 | ||||
| -rw-r--r-- | poppler/FontInfo.h | 2 |
2 files changed, 5 insertions, 5 deletions
diff --git a/poppler/FontInfo.cc b/poppler/FontInfo.cc index 7dc62769..87c212ee 100644 --- a/poppler/FontInfo.cc +++ b/poppler/FontInfo.cc @@ -52,17 +52,17 @@ FontInfoScanner::FontInfoScanner(PDFDoc *docA, int firstPage) { FontInfoScanner::~FontInfoScanner() { } -std::vector<FontInfo*> *FontInfoScanner::scan(int nPages) { +std::vector<FontInfo*> FontInfoScanner::scan(int nPages) { Page *page; Dict *resDict; Annots *annots; int lastPage; + std::vector<FontInfo*> result; + if (currentPage > doc->getNumPages()) { - return nullptr; + return result; } - - auto result = new std::vector<FontInfo*>(); lastPage = currentPage + nPages; if (lastPage > doc->getNumPages() + 1) { diff --git a/poppler/FontInfo.h b/poppler/FontInfo.h index 6ed6fb2f..0ecef820 100644 --- a/poppler/FontInfo.h +++ b/poppler/FontInfo.h @@ -92,7 +92,7 @@ public: // Destructor. ~FontInfoScanner(); - std::vector<FontInfo*> *scan(int nPages); + std::vector<FontInfo*> scan(int nPages); private: |