readpdf-cli 0.1.0__py3-none-any.whl

readpdf.py

@@ -0,0 +1,60 @@
+#!/usr/bin/env python3
+"""readpdf - Convert PDF to text for token-efficient AI reading"""
+
+import argparse
+import subprocess
+import sys
+from pathlib import Path
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="Convert PDF to text for token-efficient AI agent reading"
+    )
+    parser.add_argument("pdf", help="PDF file path")
+    parser.add_argument("-o", "--output", help="Output text file (default: stdout)")
+    parser.add_argument("-p", "--page", type=int, help="Extract specific page only")
+    parser.add_argument("--pages", help="Page range, e.g. 3-7", metavar="START-END")
+    args = parser.parse_args()
+
+    pdf_path = Path(args.pdf)
+    if not pdf_path.exists():
+        print(f"Error: {pdf_path} not found", file=sys.stderr)
+        sys.exit(1)
+
+    cmd = ["pdftotext", "-layout"]
+
+    if args.page:
+        cmd += ["-f", str(args.page), "-l", str(args.page)]
+    elif args.pages:
+        try:
+            start, end = args.pages.split("-")
+            cmd += ["-f", start.strip(), "-l", end.strip()]
+        except ValueError:
+            print("Error: --pages format must be START-END, e.g. 3-7", file=sys.stderr)
+            sys.exit(1)
+
+    if args.output:
+        cmd += [str(pdf_path), args.output]
+    else:
+        cmd += [str(pdf_path), "-"]
+
+    try:
+        result = subprocess.run(cmd, capture_output=True, text=True)
+        if result.returncode != 0:
+            print(f"Error: {result.stderr}", file=sys.stderr)
+            sys.exit(1)
+        if not args.output:
+            print(result.stdout, end="")
+    except FileNotFoundError:
+        print(
+            "Error: pdftotext not found.\n"
+            "  macOS:  brew install poppler\n"
+            "  Linux:  apt install poppler-utils",
+            file=sys.stderr,
+        )
+        sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()

readpdf_cli-0.1.0.dist-info/METADATA

@@ -0,0 +1,71 @@
+Metadata-Version: 2.4
+Name: readpdf-cli
+Version: 0.1.0
+Summary: Convert PDF to text for token-efficient AI agent reading
+Project-URL: Homepage, https://github.com/xodn348/readpdf
+License: MIT
+License-File: LICENSE
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+
+# readpdf
+
+Convert PDFs to text for token-efficient AI agent reading.
+
+## Why
+
+Reading PDFs with vision costs 100–500 tokens per page. Converting to text first lets any AI agent read only the sections it needs — saving **up to 90% of tokens**.
+
+## Install
+
+```bash
+pip install readpdf
+```
+
+Requires `pdftotext`:
+```bash
+# macOS
+brew install poppler
+
+# Linux
+apt install poppler-utils
+```
+
+## Usage
+
+```bash
+# Convert to file, then read selectively
+readpdf paper.pdf -o paper.txt
+
+# Extract a single page
+readpdf paper.pdf -o paper.txt -p 3
+
+# Extract a page range
+readpdf paper.pdf -o paper.txt --pages 3-7
+
+# Print to stdout
+readpdf paper.pdf
+```
+
+Works with any AI agent that has shell access: Claude, GPT-4, Gemini, Cursor, etc.
+
+## How it works
+
+```
+Without readpdf:
+  PDF → AI vision → ~200 tokens/page × 30 pages = 6,000 tokens (full doc, every time)
+
+With readpdf:
+  PDF → pdftotext → paper.txt (on disk, 0 tokens)
+  AI reads only offset/limit chunks it needs → ~300 tokens total
+```
+
+**Step 1.** `readpdf paper.pdf -o paper.txt` runs `pdftotext` locally — no AI tokens consumed.
+
+**Step 2.** The AI uses a file-reading tool (e.g. Read with `offset`/`limit`) to load only the relevant lines. Because the text file already exists on disk, the AI never pays the cost of processing the entire PDF.
+
+**Why not MCP?** MCP tool results return the full content back into the AI's context window — same cost as reading directly. A disk file lets the AI pull exactly the slice it needs, nothing more.
+
+## License
+
+MIT

readpdf_cli-0.1.0.dist-info/RECORD

@@ -0,0 +1,6 @@
+readpdf.py,sha256=vkz8m4Gu2VdGrkc4O2bCp0Qt3u2QEh2DgWSyQ8OSO10,1842
+readpdf_cli-0.1.0.dist-info/METADATA,sha256=j_To4Y1I38Pm8NqJXvGzpcD9v7zfXror96YB1gu5jdE,1814
+readpdf_cli-0.1.0.dist-info/WHEEL,sha256=QccIxa26bgl1E6uMy58deGWi-0aeIkkangHcxk2kWfw,87
+readpdf_cli-0.1.0.dist-info/entry_points.txt,sha256=FRsrAhal_wa5bud0rSofbYsEm-4UWcn6liUzZod4jkU,41
+readpdf_cli-0.1.0.dist-info/licenses/LICENSE,sha256=2pRddN1trfdXxktpmiIzZqgb-PwbONYWD0A-LV1coKY,1068
+readpdf_cli-0.1.0.dist-info/RECORD,,

readpdf_cli-0.1.0.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: hatchling 1.29.0
+Root-Is-Purelib: true
+Tag: py3-none-any

readpdf_cli-0.1.0.dist-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+readpdf = readpdf:main

readpdf_cli-0.1.0.dist-info/licenses/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Junhyuk Lee
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.