Module: Rpdfium::Util::TextExtraction

Defined in:
lib/rpdfium/util/text_extraction.rb

Overview

“Linear” text extraction from a collection of chars, layout=False. Equivalent of pdfplumber.utils.text.chars_to_textmap in the variant without preservation of the graphic layout.

Algorithm:

1. Extract words with WordExtractor (same tolerances).
2. Cluster words by `top` with y_tolerance → logical lines.
3. For each line, sort by x0 and join with a single space.
4. Join the lines with "\n".

NOTE on a subtlety: pdfplumber allows using an x_tolerance different from y_tolerance both for word-extraction and for line-clustering. We replicate this flexibility.

Constant Summary collapse

DEFAULT_X_TOLERANCE = 
WordExtractor::DEFAULT_X_TOLERANCE
DEFAULT_Y_TOLERANCE = 
WordExtractor::DEFAULT_Y_TOLERANCE

Class Method Summary collapse

Class Method Details

.extract_text(chars, x_tolerance: DEFAULT_X_TOLERANCE, y_tolerance: DEFAULT_Y_TOLERANCE, keep_blank_chars: false) ⇒ Object 



24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
 
# File 'lib/rpdfium/util/text_extraction.rb', line 24

def extract_text(chars,
                 x_tolerance: DEFAULT_X_TOLERANCE,
                 y_tolerance: DEFAULT_Y_TOLERANCE,
                 keep_blank_chars: false)
  return "" if chars.empty?

  words = WordExtractor.new(
    x_tolerance: x_tolerance,
    y_tolerance: y_tolerance,
    keep_blank_chars: keep_blank_chars
  ).extract_words(chars)
  return "" if words.empty?

  # Cluster the WORDS by top: final output lines.
  # Uses the "line" y_tolerance — pdfplumber here uses the same
  # y_tolerance passed in, consistent with how extract_text behaves.
  line_clusters = Cluster.cluster_objects(words, :top, tolerance: y_tolerance)

  # For each output line, join with a single space.
  line_clusters.map do |line_words|
    line_words.sort_by { |w| w[:x0] }.map { |w| w[:text] }.join(" ")
  end.join("\n")
end