HowdenParser 3.1.0__tar.gz → 3.1.1__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {howdenparser-3.1.0 → howdenparser-3.1.1}/HowdenParser/parser.py +11 -3
- {howdenparser-3.1.0 → howdenparser-3.1.1}/PKG-INFO +1 -1
- {howdenparser-3.1.0 → howdenparser-3.1.1}/pyproject.toml +1 -1
- {howdenparser-3.1.0 → howdenparser-3.1.1}/HowdenParser/__init__.py +0 -0
- {howdenparser-3.1.0 → howdenparser-3.1.1}/HowdenParser/parameter/__init__.py +0 -0
- {howdenparser-3.1.0 → howdenparser-3.1.1}/HowdenParser/parameter/huggingface.py +0 -0
- {howdenparser-3.1.0 → howdenparser-3.1.1}/HowdenParser/parameter/llamaparser.py +0 -0
- {howdenparser-3.1.0 → howdenparser-3.1.1}/HowdenParser/parameter/mistralocr.py +0 -0
- {howdenparser-3.1.0 → howdenparser-3.1.1}/README.md +0 -0
|
@@ -62,7 +62,9 @@ class Parser(BaseParser):
|
|
|
62
62
|
result_type: str,
|
|
63
63
|
mode: bool,
|
|
64
64
|
preserve_layout_alignment_across_pages: bool,
|
|
65
|
-
merge_tables_across_pages_in_markdown: bool
|
|
65
|
+
merge_tables_across_pages_in_markdown: bool,
|
|
66
|
+
hide_footers: bool,
|
|
67
|
+
hide_headers: bool,
|
|
66
68
|
) -> "LlamaParser": ...
|
|
67
69
|
|
|
68
70
|
@overload
|
|
@@ -237,13 +239,17 @@ class LlamaParser(BaseParser, name="llamaparser"):
|
|
|
237
239
|
mode: bool,
|
|
238
240
|
provider_and_model: str,
|
|
239
241
|
merge_tables_across_pages_in_markdown: bool,
|
|
240
|
-
preserve_layout_alignment_across_pages: bool
|
|
242
|
+
preserve_layout_alignment_across_pages: bool,
|
|
243
|
+
hide_footers: bool,
|
|
244
|
+
hide_headers: bool) -> None:
|
|
241
245
|
logging.info("Initializing LlamaParser...")
|
|
242
246
|
self.result_type = result_type
|
|
243
247
|
self.mode = mode
|
|
244
248
|
self.provider_and_model = provider_and_model
|
|
245
249
|
self.merge_tables_across_pages_in_markdown = merge_tables_across_pages_in_markdown
|
|
246
250
|
self.preserve_layout_alignment_across_pages = preserve_layout_alignment_across_pages
|
|
251
|
+
self.hide_footers=hide_footers
|
|
252
|
+
self.hide_headers=hide_headers
|
|
247
253
|
|
|
248
254
|
from llama_parse import LlamaParse, ResultType
|
|
249
255
|
if result_type.lower() in ("md", "markdown"):
|
|
@@ -255,7 +261,9 @@ class LlamaParser(BaseParser, name="llamaparser"):
|
|
|
255
261
|
result_type=result_type,
|
|
256
262
|
premium_mode=self.mode,
|
|
257
263
|
merge_tables_across_pages_in_markdown=self.merge_tables_across_pages_in_markdown,
|
|
258
|
-
preserve_layout_alignment_across_pages=self.preserve_layout_alignment_across_pages
|
|
264
|
+
preserve_layout_alignment_across_pages=self.preserve_layout_alignment_across_pages,
|
|
265
|
+
hide_footers=self.hide_footers,
|
|
266
|
+
hide_headers=self.hide_headers)
|
|
259
267
|
|
|
260
268
|
def parse(self, file_path: Path) -> str:
|
|
261
269
|
documents = self._parser.load_data(str(file_path))
|
|
@@ -14,7 +14,7 @@ build-backend = "poetry.core.masonry.api"
|
|
|
14
14
|
|
|
15
15
|
[tool.poetry]
|
|
16
16
|
name = "HowdenParser"
|
|
17
|
-
version = "3.1.
|
|
17
|
+
version = "3.1.1"
|
|
18
18
|
description = "A simple configuration manager with Pydantic and JSON export."
|
|
19
19
|
authors = [ "JesperThoftIllemannJ <jesper.jaeger@howdendanmark.dk>",]
|
|
20
20
|
readme = "README.md"
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|