Zpdf: PDF text extraction in Zig

Post by forgotpwd16 » Tue, Dec 30, 2025, 11:05 PM UTC

dmytrish wrote: post #28

littlestymaar wrote: post #16

- First commit 3hours ago. - commit message: LLM-generated. - README: LLM-generated. I'm not convinced that projects vibe coded over the evening deserve the HN front page… Edit: and of course the author's blog is also full of AI slop… 2026 hasn't even started I already hate it.

...and it does not work. I tried it on ~10 random pdfs, including very simple ones (e.g. a hello world from typst), it segfaults on every single one.

Post by TZubiri » Tue, Dec 30, 2025, 11:08 PM UTC

lulzx wrote: post #30

forgotpwd16 wrote: post #26

74910,74912c187768,187779 > myconv; > [Example 1: If you want to use the code conversion facet codecvt_utf8 to output to cout a UTF-8 multibyte sequence > corresponding to a wide string, but you don’t want to alter the locale for cout, you can write something like: > > § D.27.2 > 1954 > > © ISO/IEC > N4950 > > wstring_convert > myconv; > std::string mbstring = myconv.to_bytes(L"Hello\n"); Is indeed faster but output…

fixed.

Post by forgotpwd16 » Tue, Dec 30, 2025, 11:20 PM UTC

lulzx wrote: post #30

forgotpwd16 wrote: post #26

74910,74912c187768,187779 > myconv; > [Example 1: If you want to use the code conversion facet codecvt_utf8 to output to cout a UTF-8 multibyte sequence > corresponding to a wide string, but you don’t want to alter the locale for cout, you can write something like: > > § D.27.2 > 1954 > > © ISO/IEC > N4950 > > wstring_convert > myconv; > std::string mbstring = myconv.to_bytes(L"Hello\n"); Is indeed faster but output…

fixed.

  01F9020101FC020401F9020301FB02070205020800030209020701FF01F90203020901F9012D020A0201020101FF01FB01FE0208 
  0200012E0219021802160218013202120222 0209021D0212021D012E013202200222000301FA021A0220021C022002160213012E0222000F000301F90206012C

  020301FF02000205020101FC020901F90003020001F9020701F9020E020802000205020A 
  01FC028C0213021B022002230221021800030200012E021902180216021201320221021A012E00030209021D0212021D012E013202200222000301FA021A0220021C022002160213012E0222000F000301F90206012C 
 
  0200020D02030208020901F90203020901FF0203020502080003012B020001F9012B020001F901FA0205020A01FD01FE0208 
  020201300132012E012F021A012F0210021B013202200221012E0222 0209021D0212021D012E013202200222000301FA021A0220021C022002160213012E0222000F000301F90206012C 

Post by DetroitThrow » Tue, Dec 30, 2025, 11:39 PM UTC

TZubiri wrote: post #32

forgotpwd16 wrote: post #26

74910,74912c187768,187779 > myconv; > [Example 1: If you want to use the code conversion facet codecvt_utf8 to output to cout a UTF-8 multibyte sequence > corresponding to a wide string, but you don’t want to alter the locale for cout, you can write something like: > > § D.27.2 > 1954 > > © ISO/IEC > N4950 > > wstring_convert > myconv; > std::string mbstring = myconv.to_bytes(L"Hello\n"); Is indeed faster but output…

In my experience with parsing PDFs, speed has never been an issue, it has always been a matter of quality.

Post by rishabhaiover » Tue, Dec 30, 2025, 11:43 PM UTC

jeffbee wrote: post #12

lulzx wrote: post #2

I built a PDF text extraction library in Zig that's significantly faster than MuPDF for text extraction workloads. ~41K pages/sec peak throughput. Key choices: memory-mapped I/O, SIMD string search, parallel page extraction, streaming output. Handles CID fonts, incremental updates, all common compression filters. ~5,000 lines, no dependencies, compiles in Why it's fast: - Memory-mapped file I/O (no read syscalls) - Z…

What's fast about mmap?