lm-deluge 0.0.6__tar.gz → 0.0.7__tar.gz

{lm_deluge-0.0.6 → lm_deluge-0.0.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: lm_deluge
-Version: 0.0.6
+Version: 0.0.7
 Summary: Python utility for using LLM API models.
 Author-email: Benjamin Anderson <ben@trytaylor.ai>
 Requires-Python: >=3.10
@@ -20,10 +20,8 @@ Requires-Dist: bs4
 Requires-Dist: lxml
 Requires-Dist: pdf2image
 Requires-Dist: pillow
-Provides-Extra: full
-Requires-Dist: pymupdf; extra == "full"
-Requires-Dist: fasttext-wheel; extra == "full"
-Requires-Dist: fasttext-langdetect; extra == "full"
+Requires-Dist: fasttext-wheel
+Requires-Dist: fasttext-langdetect
 
 # lm_deluge
 
@@ -44,11 +42,6 @@ Requires-Dist: fasttext-langdetect; extra == "full"
 pip install lm-deluge
 ```
 
-There are optional goodies. If you want support for PDFs and language-detection via FastText:
-```bash
-pip install "lm-deluge[full]"
-```
-
 The package relies on environment variables for API keys. Typical variables include `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, `COHERE_API_KEY`, `META_API_KEY`, and `GOOGLE_API_KEY`. `LLMClient` will automatically load the `.env` file when imported; we recommend using that to set the environment variables.
 
 ## Quickstart

{lm_deluge-0.0.6 → lm_deluge-0.0.7}/README.md

@@ -17,11 +17,6 @@
 pip install lm-deluge
 ```
 
-There are optional goodies. If you want support for PDFs and language-detection via FastText:
-```bash
-pip install "lm-deluge[full]"
-```
-
 The package relies on environment variables for API keys. Typical variables include `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, `COHERE_API_KEY`, `META_API_KEY`, and `GOOGLE_API_KEY`. `LLMClient` will automatically load the `.env` file when imported; we recommend using that to set the environment variables.
 
 ## Quickstart

{lm_deluge-0.0.6 → lm_deluge-0.0.7}/pyproject.toml

@@ -3,7 +3,7 @@ requires = ["setuptools", "wheel"]
 
 [project]
 name = "lm_deluge"
-version = "0.0.6"
+version = "0.0.7"
 authors = [{ name = "Benjamin Anderson", email = "ben@trytaylor.ai" }]
 description = "Python utility for using LLM API models."
 readme = "README.md"
@@ -26,12 +26,7 @@ dependencies = [
     "bs4",
     "lxml",
     "pdf2image",
-    "pillow"
-]
-
-[project.optional-dependencies]
-full = [
-    "pymupdf",
+    "pillow",
     "fasttext-wheel",
     "fasttext-langdetect",
 ]

{lm_deluge-0.0.6 → lm_deluge-0.0.7}/src/lm_deluge.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: lm_deluge
-Version: 0.0.6
+Version: 0.0.7
 Summary: Python utility for using LLM API models.
 Author-email: Benjamin Anderson <ben@trytaylor.ai>
 Requires-Python: >=3.10
@@ -20,10 +20,8 @@ Requires-Dist: bs4
 Requires-Dist: lxml
 Requires-Dist: pdf2image
 Requires-Dist: pillow
-Provides-Extra: full
-Requires-Dist: pymupdf; extra == "full"
-Requires-Dist: fasttext-wheel; extra == "full"
-Requires-Dist: fasttext-langdetect; extra == "full"
+Requires-Dist: fasttext-wheel
+Requires-Dist: fasttext-langdetect
 
 # lm_deluge
 
@@ -44,11 +42,6 @@ Requires-Dist: fasttext-langdetect; extra == "full"
 pip install lm-deluge
 ```
 
-There are optional goodies. If you want support for PDFs and language-detection via FastText:
-```bash
-pip install "lm-deluge[full]"
-```
-
 The package relies on environment variables for API keys. Typical variables include `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, `COHERE_API_KEY`, `META_API_KEY`, and `GOOGLE_API_KEY`. `LLMClient` will automatically load the `.env` file when imported; we recommend using that to set the environment variables.
 
 ## Quickstart

{lm_deluge-0.0.6 → lm_deluge-0.0.7}/src/lm_deluge.egg-info/SOURCES.txt

@@ -35,7 +35,6 @@ src/lm_deluge/llm_tools/score.py
 src/lm_deluge/llm_tools/translate.py
 src/lm_deluge/util/json.py
 src/lm_deluge/util/logprobs.py
-src/lm_deluge/util/pdf.py
 src/lm_deluge/util/validation.py
 src/lm_deluge/util/xml.py
 tests/test_all_models.py

{lm_deluge-0.0.6 → lm_deluge-0.0.7}/src/lm_deluge.egg-info/requires.txt

@@ -13,8 +13,5 @@ bs4
 lxml
 pdf2image
 pillow
-
-[full]
-pymupdf
 fasttext-wheel
 fasttext-langdetect

lm_deluge-0.0.6/src/lm_deluge/util/pdf.py

@@ -1,45 +0,0 @@
-import io
-
-
-def text_from_pdf(pdf: str | bytes | io.BytesIO):
-    """
-    Extract text from a PDF. Does NOT use OCR, extracts the literal text.
-    The source can be:
-    - A file path (str)
-    - Bytes of a PDF file
-    - A BytesIO object containing a PDF file
-    """
-    try:
-        import pymupdf  # pyright: ignore
-    except ImportError:
-        raise ImportError(
-            "pymupdf is required to extract text from PDFs. Install lm_deluge[pdf] or lm_deluge[full]."
-        )
-    if isinstance(pdf, str):
-        # It's a file path
-        doc = pymupdf.open(pdf)
-    elif isinstance(pdf, (bytes, io.BytesIO)):
-        # It's bytes or a BytesIO object
-        if isinstance(pdf, bytes):
-            pdf = io.BytesIO(pdf)
-        doc = pymupdf.open(stream=pdf, filetype="pdf")
-    else:
-        raise ValueError("Unsupported pdf_source type. Must be str, bytes, or BytesIO.")
-
-    text_content = []
-    for page in doc:
-        blocks = page.get_text("blocks", sort=True)
-        for block in blocks:
-            # block[4] contains the text content
-            text_content.append(block[4].strip())
-            text_content.append("\n")  # Add extra newlines between blocks
-
-    # Join all text content with newlines
-    full_text = "\n".join(text_content).strip()
-    # Replace multiple consecutive spaces with a single space
-    full_text = " ".join(full_text.split())
-    # Clean up any resulting double spaces or newlines
-    full_text = " ".join([x for x in full_text.split(" ") if x])
-    full_text = "\n".join([x for x in full_text.split("\n") if x])
-
-    return full_text