reformatted the text

Signed-off-by: Peter Staar <taa@zurich.ibm.com>
This commit is contained in:
Peter Staar 2024-10-29 06:02:49 +01:00
parent 7cb7da7ce9
commit 6163409305
3 changed files with 67 additions and 54 deletions

View File

@ -1,16 +1,19 @@
import os
import json
import os
from pathlib import Path
from docling.backend.html_backend import HTMLDocumentBackend
from docling.datamodel.base_models import InputFormat
from docling.datamodel.document import InputDocument, SectionHeaderItem
from docling.datamodel.document import ConversionResult
from docling.datamodel.document import (
ConversionResult,
InputDocument,
SectionHeaderItem,
)
from docling.document_converter import DocumentConverter
GENERATE = False
def test_heading_levels():
in_path = Path("tests/data/html/wiki_duck.html")
in_doc = InputDocument(
@ -35,6 +38,7 @@ def test_heading_levels():
assert item.level == 3
assert found_lvl_2 and found_lvl_3
def get_html_paths():
# Define the directory you want to search
@ -44,14 +48,14 @@ def get_html_paths():
html_files = sorted(directory.rglob("*.html"))
return html_files
def get_converter():
converter = DocumentConverter(
allowed_formats = [InputFormat.HTML]
)
converter = DocumentConverter(allowed_formats=[InputFormat.HTML])
return converter
def verify_export(pred_text: str, gtfile: str):
if not os.path.exists(gtfile) or GENERATE:
@ -67,6 +71,7 @@ def verify_export(pred_text:str, gtfile:str):
assert pred_text == true_text, "pred_itxt==true_itxt"
return pred_text == true_text
def test_e2e_html_conversions():
html_paths = get_html_paths()
@ -82,10 +87,12 @@ def test_e2e_html_conversions():
pred_md: str = doc.export_to_markdown()
assert verify_export(pred_md, str(html_path) + ".md"), "export to md"
pred_itxt:str = doc._export_to_indented_text(max_text_len=70, explicit_tables=False)
assert verify_export(pred_itxt, str(html_path)+".itxt"), "export to indented-text"
pred_itxt: str = doc._export_to_indented_text(
max_text_len=70, explicit_tables=False
)
assert verify_export(
pred_itxt, str(html_path) + ".itxt"
), "export to indented-text"
pred_json: str = json.dumps(doc.export_to_dict(), indent=2)
assert verify_export(pred_json, str(html_path) + ".json"), "export to json"

View File

@ -1,16 +1,19 @@
import os
import json
import os
from pathlib import Path
from docling.backend.msword_backend import MsWordDocumentBackend
from docling.datamodel.base_models import InputFormat
from docling.datamodel.document import InputDocument, SectionHeaderItem
from docling.datamodel.document import ConversionResult
from docling.datamodel.document import (
ConversionResult,
InputDocument,
SectionHeaderItem,
)
from docling.document_converter import DocumentConverter
GENERATE = False
def test_heading_levels():
in_path = Path("tests/data/docx/word_sample.docx")
in_doc = InputDocument(
@ -45,14 +48,14 @@ def get_docx_paths():
pdf_files = sorted(directory.rglob("*.docx"))
return pdf_files
def get_converter():
converter = DocumentConverter(
allowed_formats = [InputFormat.DOCX]
)
converter = DocumentConverter(allowed_formats=[InputFormat.DOCX])
return converter
def verify_export(pred_text: str, gtfile: str):
if not os.path.exists(gtfile) or GENERATE:
@ -84,9 +87,12 @@ def test_e2e_docx_conversions():
pred_md: str = doc.export_to_markdown()
assert verify_export(pred_md, str(docx_path) + ".md"), "export to md"
pred_itxt:str = doc._export_to_indented_text(max_text_len=70, explicit_tables=False)
assert verify_export(pred_itxt, str(docx_path)+".itxt"), "export to indented-text"
pred_itxt: str = doc._export_to_indented_text(
max_text_len=70, explicit_tables=False
)
assert verify_export(
pred_itxt, str(docx_path) + ".itxt"
), "export to indented-text"
pred_json: str = json.dumps(doc.export_to_dict(), indent=2)
assert verify_export(pred_json, str(docx_path) + ".json"), "export to json"