Messing around with Github CI and PDF processing.
This program will take two PDF files and match all paragraphs (which SHOULD be numbered) and store the matched pairs in a .csv. file.
Example usage
python interleave.py file1.pdf file2.pdf output.csv
Example output
Document1,Document2
1. First Entry.,1. First Entry
2. Second Entry.,2. Second Entry
3. Third Entry,3. Third Entry
Current Status
102/1088 paragraphs in test data have an anomaly.
A complete list of observed errors is in Errors.csv
Error types:
- Double-Number Parse
- Excessive Heading
- Grouped Response
- Heading Parse
- Missing Character(s)
- Missing Text
- Pagebreak Parse
- Parse Error
- Parsed Count
- Preceding Data
- WTF
Here's a tabular representation of the anomalies.
| 1/2 | 3/4* | 5/6 | 7/8 | Total | |
|---|---|---|---|---|---|
| EJ | 17 | 6 | 8 | 17 | 48 |
| EPA | 15 | 3 | 21 | 15 | 54 |
| Total | 32 | 9 | 29 | 32 | 102 |