fix: pdf2text imports

dvdblk · Oct 26, 2023 · 90ff310 · 90ff310
1 parent 2492a62
commit 90ff310
Show file tree

Hide file tree

Showing 4 changed files with 9 additions and 7 deletions.
diff --git a/experiments/alex/pdf2text/convert.py b/experiments/alex/pdf2text/convert.py
@@ -11,9 +11,12 @@ class PDFConverter:
 
     Note:
         The full conversion pipeline is as follows:
-        1. `extract` text and formats from PDF (including tables and images)
-            a. `preprocess` the text (remove linebreaks, page numbers, headers, footers, etc.)
-            b. `postprocess` the text (pdf specific postprocessing e.g. remove first few pages of intro, blank pages, etc.)
+        1. `extract` text and formats from PDF (including text from tables and images)
+            a. `preprocess` the text = works on individual text paragraphs, tables, images
+                                        (remove linebreaks, join words, remove whitespace etc.)
+            b. `postprocess` the text = works on full pages of preprocessed text
+                                        (e.g. remove first few pages of intro, blank pages,
+                                        page numbers, etc.)
         2. (optional) `translate`
         3. `format` the text to the desired output
     """

diff --git a/experiments/alex/pdf2text/extract.py b/experiments/alex/pdf2text/extract.py
@@ -1,12 +1,11 @@
 from typing import Any, Generator, List, Tuple
 
 import pdfplumber
+from pdf2text.models import ExtractedPage
 from pdf2text.preprocessing.manager import PreprocessorManager
 from pdfminer.high_level import extract_pages, extract_text
 from pdfminer.layout import LTChar, LTFigure, LTItem, LTRect, LTTextContainer
 
-from .model import ExtractedPage
-
 
 # TODO: Move this to formatter.py
 def table_converter(table):

diff --git a/experiments/alex/pdf2text/preprocessing/decorator.py b/experiments/alex/pdf2text/preprocessing/decorator.py
@@ -1,4 +1,4 @@
-from pdf2text.model import PDFElementType
+from pdf2text.models import PDFElementType
 
 _preprocessors = {
     PDFElementType.TABLE: [],

diff --git a/experiments/alex/pdf2text/preprocessing/manager.py b/experiments/alex/pdf2text/preprocessing/manager.py
@@ -1,6 +1,6 @@
 from typing import Optional
 
-from pdf2text.model import PDFElementType
+from pdf2text.models import PDFElementType
 from pdf2text.preprocessing.decorator import _preprocessors, _register_preprocessor
 from pdf2text.preprocessing.result import (
     ModifiedPreprocessResult,