spectre_pdf 1.0.0

Native Rust PDF extraction engine: text, markdown for RAG, AcroForm widgets, image decoding, and encrypted PDFs. Lazy parser, persistent Document handle, no C dependencies.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

==============================================================================================================
PDF                               imgs  extracted   pmu ms   spec ms   speedup   pixel match
==============================================================================================================
eu-001.pdf                           1          1     0.08      0.00    23.98x 1/          1
eu-002.pdf                           2          2     3.63      2.15     1.69x 2/          2
eu-003.pdf                           1          1     0.06      0.00    24.14x 1/          1
eu-009a.pdf                          1          1     1.95      1.43     1.37x 1/          1
eu-011.pdf                          32         32    62.57     43.64     1.43x 8/          8
eu-012.pdf                           1          1     0.12      0.01     8.33x 0/          0
eu-015.pdf                           5          5     0.31      0.16     1.93x 3/          3
eu-020.pdf                           4          4    13.38     11.76     1.14x 4/          4
i1040gi.pdf                          2          2    74.23      0.20   375.31x 1/          1
f1099int.pdf                         1          1     0.07      0.01     5.36x 1/          1
==============================================================================================================
TOTAL  speedup 2.63x  (156.4ms / 59.4ms)
  extracted (viewable container): 50/50 = 1.000
  pixel-exact equality vs pymupdf: 22/22 = 1.000
  CMYK preserved (pymupdf converted to RGB, not counted as miss): 1 images