Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -1,20 +1,24 @@
|
|
| 1 |
"""
|
| 2 |
-
Streamlit
|
| 3 |
-
--------------------------------------
|
| 4 |
-
A Streamlit app
|
| 5 |
-
contents
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
*
|
| 9 |
-
|
| 10 |
-
*
|
| 11 |
-
*
|
| 12 |
-
|
| 13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
"""
|
| 15 |
from __future__ import annotations
|
| 16 |
|
| 17 |
-
#
|
| 18 |
import os, pathlib
|
| 19 |
os.environ.setdefault("STREAMLIT_HOME", "/tmp/.streamlit")
|
| 20 |
os.environ.setdefault("HOME", "/tmp")
|
|
@@ -22,147 +26,152 @@ pathlib.Path(os.environ["STREAMLIT_HOME"]).mkdir(parents=True, exist_ok=True)
|
|
| 22 |
|
| 23 |
import io
|
| 24 |
import zipfile
|
|
|
|
|
|
|
| 25 |
from datetime import datetime
|
| 26 |
from pathlib import Path
|
| 27 |
|
| 28 |
import streamlit as st
|
| 29 |
-
from PIL import Image
|
|
|
|
|
|
|
|
|
|
| 30 |
|
| 31 |
# -----------------------------------------------------------------------------
|
| 32 |
-
# Supported extensions
|
| 33 |
# -----------------------------------------------------------------------------
|
| 34 |
-
IMAGE_EXTS = {
|
| 35 |
-
|
| 36 |
-
}
|
| 37 |
-
TEXT_EXTS = {
|
| 38 |
-
".txt", ".md", ".csv", ".json", ".xml", ".html", ".css", ".js",
|
| 39 |
-
}
|
| 40 |
-
ARCHIVE_EXTS = {".zip", ".tar", ".gz", ".7z"}
|
| 41 |
MEDIA_EXTS = {".mp3", ".wav", ".mp4", ".avi", ".mkv", ".mov"}
|
| 42 |
-
DOC_EXTS = {".pdf", ".doc", ".docx"}
|
| 43 |
-
|
| 44 |
-
ALLOWED_TARGET_EXTS: list[str] = sorted(
|
| 45 |
-
IMAGE_EXTS | TEXT_EXTS | ARCHIVE_EXTS | MEDIA_EXTS | DOC_EXTS
|
| 46 |
-
)
|
| 47 |
|
|
|
|
| 48 |
DISALLOWED_SOURCE_EXTS = {".exe", ".bin"}
|
| 49 |
|
| 50 |
# -----------------------------------------------------------------------------
|
| 51 |
-
#
|
| 52 |
# -----------------------------------------------------------------------------
|
| 53 |
-
|
| 54 |
def sidebar_target_extension() -> str:
|
| 55 |
st.sidebar.header("Settings")
|
| 56 |
query = st.sidebar.text_input("Filter extensions… (optional)")
|
| 57 |
-
|
| 58 |
-
if not
|
| 59 |
st.sidebar.error("No extension matches that filter.")
|
| 60 |
-
|
| 61 |
-
|
| 62 |
-
|
| 63 |
-
index=(filtered or ALLOWED_TARGET_EXTS).index(".png")
|
| 64 |
-
if ".png" in (filtered or ALLOWED_TARGET_EXTS) else 0,
|
| 65 |
-
)
|
| 66 |
-
st.sidebar.markdown(
|
| 67 |
-
"*Images are truly converted. Text files are re‑saved as UTF‑8. "
|
| 68 |
-
"Other combinations fall back to a safe rename.*"
|
| 69 |
)
|
| 70 |
-
return target_ext
|
| 71 |
-
|
| 72 |
|
| 73 |
def uploader():
|
| 74 |
return st.file_uploader(
|
| 75 |
-
"Upload
|
| 76 |
-
accept_multiple_files=True,
|
| 77 |
-
type=None, # accept *all* extensions
|
| 78 |
-
help="Drag‑and‑drop or click to browse.",
|
| 79 |
)
|
| 80 |
|
| 81 |
-
|
| 82 |
# -----------------------------------------------------------------------------
|
| 83 |
-
# Conversion
|
| 84 |
# -----------------------------------------------------------------------------
|
| 85 |
-
|
| 86 |
def convert_image(data: bytes, target_ext: str) -> bytes:
|
| 87 |
-
"""Return `bytes` of the image converted to `target_ext`. Raises if Pillow
|
| 88 |
-
cannot save in that format."""
|
| 89 |
img = Image.open(io.BytesIO(data))
|
| 90 |
buf = io.BytesIO()
|
| 91 |
-
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
|
| 95 |
-
|
| 96 |
-
|
| 97 |
-
|
| 98 |
-
|
| 99 |
-
|
| 100 |
-
|
| 101 |
-
|
| 102 |
-
|
| 103 |
-
|
| 104 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 105 |
|
| 106 |
|
| 107 |
def convert_file(file: st.runtime.uploaded_file_manager.UploadedFile, target_ext: str) -> tuple[bytes, str]:
|
| 108 |
-
|
| 109 |
-
|
| 110 |
-
orig_ext = Path(file.name).suffix.lower()
|
| 111 |
raw = file.read()
|
| 112 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 113 |
try:
|
| 114 |
if orig_ext in IMAGE_EXTS and target_ext in IMAGE_EXTS:
|
| 115 |
return convert_image(raw, target_ext), "image converted"
|
| 116 |
-
if
|
| 117 |
-
|
| 118 |
-
|
| 119 |
-
|
| 120 |
-
|
| 121 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 122 |
return raw, "renamed only"
|
| 123 |
|
| 124 |
-
|
| 125 |
# -----------------------------------------------------------------------------
|
| 126 |
-
#
|
| 127 |
# -----------------------------------------------------------------------------
|
| 128 |
-
|
| 129 |
def package_zip(files: list[st.runtime.uploaded_file_manager.UploadedFile], target_ext: str) -> io.BytesIO:
|
| 130 |
buf = io.BytesIO()
|
| 131 |
-
with zipfile.ZipFile(buf,
|
| 132 |
for file in files:
|
| 133 |
-
|
| 134 |
-
if
|
| 135 |
-
st.warning(f"
|
| 136 |
continue
|
| 137 |
data, note = convert_file(file, target_ext)
|
| 138 |
-
|
| 139 |
-
zf.writestr(
|
| 140 |
-
st.success(f"
|
| 141 |
buf.seek(0)
|
| 142 |
return buf
|
| 143 |
|
| 144 |
-
|
| 145 |
# -----------------------------------------------------------------------------
|
| 146 |
-
# Main
|
| 147 |
# -----------------------------------------------------------------------------
|
| 148 |
|
| 149 |
def main():
|
| 150 |
-
st.set_page_config("Universal
|
| 151 |
-
st.title("🔄 Universal File
|
| 152 |
-
st.write("Upload files
|
| 153 |
|
| 154 |
target_ext = sidebar_target_extension()
|
| 155 |
files = uploader()
|
| 156 |
|
| 157 |
-
if files and st.button("
|
| 158 |
zip_buf = package_zip(files, target_ext)
|
| 159 |
-
ts = datetime.utcnow().strftime(
|
| 160 |
-
st.download_button(
|
| 161 |
-
|
| 162 |
-
|
| 163 |
-
|
| 164 |
-
st.caption("© 2025 Universal Changer • Streamlit • Hugging Face Spaces")
|
| 165 |
|
|
|
|
| 166 |
|
| 167 |
-
if __name__ ==
|
| 168 |
main()
|
|
|
|
| 1 |
"""
|
| 2 |
+
Streamlit Universal File-Format Converter
|
| 3 |
+
----------------------------------------
|
| 4 |
+
A Streamlit app for Hugging Face Spaces that **actually converts** file
|
| 5 |
+
contents across a wide array of formats, leveraging local libraries
|
| 6 |
+
(no API keys needed):
|
| 7 |
+
|
| 8 |
+
• **Images** via Pillow (JPEG, PNG, GIF, BMP, TIFF, ICO, WebP)
|
| 9 |
+
• **Text & markup** via pypandoc (MD, HTML, LaTeX, DOCX, PDF, etc.)
|
| 10 |
+
• **Office docs** via unoconv + LibreOffice headless (PDF, DOCX, PPTX, XLSX)
|
| 11 |
+
• **Audio/video** via ffmpeg-python (MP3, WAV, MP4, AVI, MKV, MOV, etc.)
|
| 12 |
+
• **MIME detection** via python-magic
|
| 13 |
+
|
| 14 |
+
Disallowed uploads: `.exe`, `.bin`
|
| 15 |
+
All outputs are streamed into a ZIP for download.
|
| 16 |
+
|
| 17 |
+
Created 2025-05-22 • v3
|
| 18 |
"""
|
| 19 |
from __future__ import annotations
|
| 20 |
|
| 21 |
+
# Set up a writable Streamlit home BEFORE importing streamlit
|
| 22 |
import os, pathlib
|
| 23 |
os.environ.setdefault("STREAMLIT_HOME", "/tmp/.streamlit")
|
| 24 |
os.environ.setdefault("HOME", "/tmp")
|
|
|
|
| 26 |
|
| 27 |
import io
|
| 28 |
import zipfile
|
| 29 |
+
import tempfile
|
| 30 |
+
import subprocess
|
| 31 |
from datetime import datetime
|
| 32 |
from pathlib import Path
|
| 33 |
|
| 34 |
import streamlit as st
|
| 35 |
+
from PIL import Image
|
| 36 |
+
import pypandoc
|
| 37 |
+
import ffmpeg
|
| 38 |
+
import magic # python-magic for mime detection
|
| 39 |
|
| 40 |
# -----------------------------------------------------------------------------
|
| 41 |
+
# Supported extensions
|
| 42 |
# -----------------------------------------------------------------------------
|
| 43 |
+
IMAGE_EXTS = {".jpg", ".jpeg", ".png", ".gif", ".bmp", ".tiff", ".ico", ".webp"}
|
| 44 |
+
TEXT_EXTS = {".txt", ".md", ".csv", ".json", ".xml", ".html", ".css", ".js"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
MEDIA_EXTS = {".mp3", ".wav", ".mp4", ".avi", ".mkv", ".mov"}
|
| 46 |
+
DOC_EXTS = {".pdf", ".doc", ".docx", ".ppt", ".pptx", ".xls", ".xlsx", ".odt", ".ods"}
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
|
| 48 |
+
ALLOWED_TARGET_EXTS = sorted(IMAGE_EXTS | TEXT_EXTS | MEDIA_EXTS | DOC_EXTS)
|
| 49 |
DISALLOWED_SOURCE_EXTS = {".exe", ".bin"}
|
| 50 |
|
| 51 |
# -----------------------------------------------------------------------------
|
| 52 |
+
# UI elements
|
| 53 |
# -----------------------------------------------------------------------------
|
|
|
|
| 54 |
def sidebar_target_extension() -> str:
|
| 55 |
st.sidebar.header("Settings")
|
| 56 |
query = st.sidebar.text_input("Filter extensions… (optional)")
|
| 57 |
+
choices = [e for e in ALLOWED_TARGET_EXTS if query.lower() in e]
|
| 58 |
+
if not choices:
|
| 59 |
st.sidebar.error("No extension matches that filter.")
|
| 60 |
+
choices = ALLOWED_TARGET_EXTS
|
| 61 |
+
return st.sidebar.selectbox(
|
| 62 |
+
"Target extension for **all** files", choices, index=choices.index(".pdf") if ".pdf" in choices else 0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 63 |
)
|
|
|
|
|
|
|
| 64 |
|
| 65 |
def uploader():
|
| 66 |
return st.file_uploader(
|
| 67 |
+
"Upload files to convert", type=None, accept_multiple_files=True
|
|
|
|
|
|
|
|
|
|
| 68 |
)
|
| 69 |
|
|
|
|
| 70 |
# -----------------------------------------------------------------------------
|
| 71 |
+
# Conversion functions
|
| 72 |
# -----------------------------------------------------------------------------
|
|
|
|
| 73 |
def convert_image(data: bytes, target_ext: str) -> bytes:
|
|
|
|
|
|
|
| 74 |
img = Image.open(io.BytesIO(data))
|
| 75 |
buf = io.BytesIO()
|
| 76 |
+
fmt = {".jpg":"JPEG", ".jpeg":"JPEG", ".png":"PNG", ".gif":"GIF",
|
| 77 |
+
".bmp":"BMP", ".tiff":"TIFF", ".ico":"ICO", ".webp":"WEBP"}[target_ext]
|
| 78 |
+
img.save(buf, format=fmt)
|
| 79 |
+
return buf.getvalue()
|
| 80 |
+
|
| 81 |
+
|
| 82 |
+
def convert_text_markup(data: bytes, orig_ext: str, target_ext: str) -> bytes:
|
| 83 |
+
text = data.decode("utf-8", errors="ignore")
|
| 84 |
+
return pypandoc.convert_text(text, to=target_ext.lstrip('.'), format=orig_ext.lstrip('.')).encode('utf-8')
|
| 85 |
+
|
| 86 |
+
|
| 87 |
+
def convert_office(temp_dir: str, data: bytes, orig_ext: str, target_ext: str) -> bytes:
|
| 88 |
+
# Use unoconv to convert office files
|
| 89 |
+
suffix_in = orig_ext
|
| 90 |
+
suffix_out = target_ext
|
| 91 |
+
in_path = Path(temp_dir) / f"input{suffix_in}"
|
| 92 |
+
out_path = Path(temp_dir) / f"output{suffix_out}"
|
| 93 |
+
in_path.write_bytes(data)
|
| 94 |
+
subprocess.run(["unoconv", "-f", suffix_out.lstrip('.'), "-o", str(out_path), str(in_path)], check=True)
|
| 95 |
+
return out_path.read_bytes()
|
| 96 |
+
|
| 97 |
+
|
| 98 |
+
def convert_media(data: bytes, target_ext: str) -> bytes:
|
| 99 |
+
# ffmpeg-python streaming
|
| 100 |
+
process = (
|
| 101 |
+
ffmpeg.input('pipe:0')
|
| 102 |
+
.output('pipe:1', format=target_ext.lstrip('.'))
|
| 103 |
+
.run_async(pipe_stdin=True, pipe_stdout=True, pipe_stderr=True)
|
| 104 |
+
)
|
| 105 |
+
out, err = process.communicate(data)
|
| 106 |
+
return out
|
| 107 |
|
| 108 |
|
| 109 |
def convert_file(file: st.runtime.uploaded_file_manager.UploadedFile, target_ext: str) -> tuple[bytes, str]:
|
| 110 |
+
name = Path(file.name)
|
| 111 |
+
orig_ext = name.suffix.lower()
|
|
|
|
| 112 |
raw = file.read()
|
| 113 |
|
| 114 |
+
if orig_ext in DISALLOWED_SOURCE_EXTS:
|
| 115 |
+
raise ValueError(f"Disallowed: {orig_ext}")
|
| 116 |
+
|
| 117 |
+
mime = magic.from_buffer(raw, mime=True) or ''
|
| 118 |
+
|
| 119 |
try:
|
| 120 |
if orig_ext in IMAGE_EXTS and target_ext in IMAGE_EXTS:
|
| 121 |
return convert_image(raw, target_ext), "image converted"
|
| 122 |
+
if mime.startswith('text/') or orig_ext in TEXT_EXTS:
|
| 123 |
+
if orig_ext != target_ext:
|
| 124 |
+
return convert_text_markup(raw, orig_ext, target_ext), "text/markup converted"
|
| 125 |
+
if orig_ext in DOC_EXTS or target_ext in DOC_EXTS:
|
| 126 |
+
with tempfile.TemporaryDirectory() as tmp:
|
| 127 |
+
return convert_office(tmp, raw, orig_ext, target_ext), "office/doc converted"
|
| 128 |
+
if mime.startswith(('audio/','video/')) or orig_ext in MEDIA_EXTS:
|
| 129 |
+
if orig_ext != target_ext:
|
| 130 |
+
return convert_media(raw, target_ext), "media converted"
|
| 131 |
+
except Exception as e:
|
| 132 |
+
st.warning(f"⚠️ Conversion failed for {file.name}: {e}. Falling back to rename.")
|
| 133 |
+
|
| 134 |
+
# Fallback: no conversion, just rename
|
| 135 |
return raw, "renamed only"
|
| 136 |
|
|
|
|
| 137 |
# -----------------------------------------------------------------------------
|
| 138 |
+
# ZIP packaging
|
| 139 |
# -----------------------------------------------------------------------------
|
|
|
|
| 140 |
def package_zip(files: list[st.runtime.uploaded_file_manager.UploadedFile], target_ext: str) -> io.BytesIO:
|
| 141 |
buf = io.BytesIO()
|
| 142 |
+
with zipfile.ZipFile(buf, 'w', zipfile.ZIP_DEFLATED) as zf:
|
| 143 |
for file in files:
|
| 144 |
+
name = Path(file.name)
|
| 145 |
+
if name.suffix.lower() in DISALLOWED_SOURCE_EXTS:
|
| 146 |
+
st.warning(f"Skipping disallowed file: {name.name}")
|
| 147 |
continue
|
| 148 |
data, note = convert_file(file, target_ext)
|
| 149 |
+
out_name = name.with_suffix(target_ext).name
|
| 150 |
+
zf.writestr(out_name, data)
|
| 151 |
+
st.success(f"{note}: {name.name} → {out_name}")
|
| 152 |
buf.seek(0)
|
| 153 |
return buf
|
| 154 |
|
|
|
|
| 155 |
# -----------------------------------------------------------------------------
|
| 156 |
+
# Main
|
| 157 |
# -----------------------------------------------------------------------------
|
| 158 |
|
| 159 |
def main():
|
| 160 |
+
st.set_page_config("Universal Converter", page_icon="🔄", layout="centered")
|
| 161 |
+
st.title("🔄 Universal File-Format Converter")
|
| 162 |
+
st.write("Upload files of any format; choose a new extension; download a ZIP of converted files.")
|
| 163 |
|
| 164 |
target_ext = sidebar_target_extension()
|
| 165 |
files = uploader()
|
| 166 |
|
| 167 |
+
if files and st.button("Convert & Download 🚀"):
|
| 168 |
zip_buf = package_zip(files, target_ext)
|
| 169 |
+
ts = datetime.utcnow().strftime('%Y%m%dT%H%M%SZ')
|
| 170 |
+
st.download_button("⬇️ Download ZIP", zip_buf,
|
| 171 |
+
file_name=f"converted_{ts}.zip",
|
| 172 |
+
mime='application/zip')
|
|
|
|
|
|
|
| 173 |
|
| 174 |
+
st.caption("© 2025 Universal Converter • Streamlit • Hugging Face Spaces")
|
| 175 |
|
| 176 |
+
if __name__ == '__main__':
|
| 177 |
main()
|